Skills and Expertise
Research Items (287)
In Reply We deliberately designed the D-PRESCRIBE study as a pragmatic clinical trial to mirror real-life conditions and test the effectiveness, rather than the efficacy, of the pharmacist-led educational intervention for patients and primary care clinicians.¹ The more similar the participants and conditions in a trial are to real life, the more reproducible the findings will be in practice. For this reason, the trial was rolled out in the community, outside of the academic setting, without financial incentives offered to pharmacists or patients. We recruited pharmacies through partnership with 3 large for-profit pharmacy chains serving rural and urban areas in Quebec. Although only half of the eligible pharmacies agreed to participate, and we could not record the profiles of pharmacies that declined, the characteristics of the 69 pharmacies that enrolled in the trial were similar to the characteristics of a population-based sample of 1742 pharmacies in the province of Quebec.² Pharmacies in the intervention group filled a mean number of 411 prescriptions per day (range, 140-1000), with more than 50% of their clientele composed of adults aged 65 years and older. This large-chain, high-volume prescribing profile of consenting pharmacies compares favorably—and even exceeds—the mean prescription volume reported in 2017 by a random sample of 292 nationally representative community pharmacies in the United States.³ The latter sample of US pharmacies reported a mean prescription volume of 228 prescriptions per day (range, 0-900). Furthermore, the following characteristics of participants in the trial reflected the type of patient that physicians and pharmacists are likely to see in practice: mean age of 75 years (range, 66-96 years) with significant polypharmacy (a mean number of 9 different medications per day [range, 1-28]), frailty rates of 27%, and almost 40% of patients taking more than 10 medications per day. The 2014 EMPOWER trial recruited a comparable sample of pharmacists and patients to an educational deprescribing intervention and achieved discontinuation rates of 27% for chronic benzodiazepine users.⁴ Taken together, the EMPOWER and D-PRESCRIBE trials provide strong evidence that the deprescribing intervention will work effectively if applied by high-volume, for-profit pharmacy chains in the United States.
Background: Super learning is an ensemble machine learning approach used increasingly as an alternative to classical prediction techniques. When implementing super learning, however, not tuning the hyperparameters of the algorithms in it may adversely affect the performance of the super learner. Methods: In this case study, we used data from a Canadian electronic prescribing system to predict when primary care physicians prescribed antidepressants for indications other than depression. The analysis included 73,576 antidepressant prescriptions and 373 candidate predictors. We derived two super learners: one using tuned hyperparameter values for each machine learning algorithm identified through an iterative grid search procedure and the other using the default values. We compared the performance of the tuned super learner to that of the super learner using default values ("untuned") and a carefully constructed logistic regression model from a previous analysis. Results: The tuned super learner had a scaled Brier score (R) of 0.322 (95% [confidence interval] CI = 0.267, 0.362). In comparison, the untuned super learner had a scaled Brier score of 0.309 (95% CI = 0.256, 0.353), corresponding to an efficiency loss of 4% (relative efficiency 0.96; 95% CI = 0.93, 0.99). The previously-derived logistic regression model had a scaled Brier score of 0.307 (95% CI = 0.245, 0.360), corresponding to an efficiency loss of 5% relative to the tuned super learner (relative efficiency 0.95; 95% CI = 0.88, 1.01). Conclusions: In this case study, hyperparameter tuning produced a super learner that performed slightly better than an untuned super learner. Tuning the hyperparameters of individual algorithms in a super learner may help optimize performance.
- Mar 2019
Background: Prescription opioid use and overdose has steadily increased over the past years, resulting in a dramatic increase in opioid-related emergency department (ED) visits and hospitalizations. Methods: This study used a prospective cohort of cancer patients having undergone surgery in Montreal (Quebec) to describe their post-discharge opioid use and identify potential patterns of unplanned health service use (ED visits, hospitalizations). Provincial health administrative claims were used to measure opioid dispensation as well as hospital re-admissions and ED visits. The hospital warehouse, patient chart and patient interview will be used to further describe patient’s medical profile. Marginal structural models will be used to model the association between use of opioids and risk of ED visits and hospitalizations. Inverse probability of treatment and censoring weights will be constructed to properly adjust for confounders that may be unbalanced between the opioid and non–opioid users as well as to account for competing risk due to mortality. Reasons for the re-admissions will also be presented as part of the analyses. Covariates will include patient comorbidities, medication history, and healthcare system characteristics such as nurse-to-patient and attending physician-to-patient ratios. Results (interim): A total of 821 were included in the study; of these, 73% (n=597) were admitted for a cancer procedure. At postoperative discharge, 605 (74%) of patients had at least one opioid dispensation, of which the majority (67%) were oxycodone with hydromorphone being the second most prescribed (28%). Among those who filled a prescription, mean age was 66 (13.4), 68% had no previous history of opioid use, and 10% have had 3 or more dispensing pharmacies in the year prior to admission, compared to less than 1% for the non–opioid users. Overall, 343 people refilled their opioid prescription at least once and 128 at least twice during the 1-year postoperative period. Among cancer patients who were opioid users, 214 ED visits occurred in the 1 year after surgery compared to only 40 for the non-cancer opioid users. Conclusion: This study will help to identify the risk profile of cancer patients who are most likely to continue using opioids for prolonged periods following surgical procedures as well as quantify the impact of opioid use and its associated burden on the healthcare system in order to identify areas for possible interventions.
Background: Drug expenditures are responsible for an increasing proportion of health costs, accounting for $1.1 trillion in annual expenditure worldwide. As hundreds of billions of dollars are being spent each year on overtreatment with prescribed medications that are either unnecessary or are in excess of lowest cost-effective therapy, programs are needed that optimize fiscally appropriate use. We evaluated whether providing physicians with information on the patient out-of-pocket payment consequences of prescribing decisions that were in excess of lowest cost-effective therapy would alter prescribing decisions using the treatment of uncomplicated hypertension as an exemplar. Methods: A single-blind cluster randomized trial was conducted over a 60-month follow-up period in 76 primary care physicians in Quebec, Canada, and their patients with uncomplicated hypertension who were using the MOXXI integrated electronic health record for drug and health problem management. Physicians were randomized to an out-of-pocket expenditure module that provided alerts for comparative out-of-payment costs, thiazide diuretics as recommended first-line therapy, and tools to monitor blood pressure targets and medication compliance, or alternatively the basic MOXXI system. System software and prescription claims were used to analyze the impact of the intervention on treatment choice, adherence, and overall and out-of-pocket payment costs using generalized estimating equations. Results: Three thousand five-hundred ninety-two eligible patients with uncomplicated hypertension were enrolled, of whom 1261 (35.1%) were newly started (incident patient) on treatment during follow-up. There was a statistically significant increase in the prescription of diuretics in the newly treated intervention (26.6%) compared to control patients (19.8%) (RR 1.65, 95% CI 1.17 to 2.33). For patients already treated (prevalent patient), there was a statistically significant interaction between the intervention and patient age, with older patients being less likely to be switched to a diuretic. Among the incident patients, physicians with less than 15 years of experience were much more likely to prescribe a diuretic (OR 10.69; 95% CI 1.49 to 76.64) than physicians with 15 to 25 years (OR 0.67; 95%CI 0.25 to 1.78), or more than 25 years of experience (OR 1.80; 95% CI 1.23 to 2.65). There was no statistically significant effect of the intervention on adherence or out-of-pocket payment cost. Conclusions: The provision of comparative information on patient out-of-pocket payments for treatment of uncomplicated hypertension had a statistically significant impact on increasing the initiation of diuretics in incident patients and switching to diuretics in younger prevalent patients. The impact of interventions to improve the cost-effectiveness of prescribing may be enhanced by also targeting patients with tools to participate in treatment decision-making and by providing physicians with comparative out-of-pocket information on all evidence-based alternatives that would enhance clinical decision-making. Trial registration: ISRCTN96253624.
- Nov 2018
Importance High rates of inappropriate prescribing persist among older adults in many outpatient settings, increasing the risk of adverse drug events and drug-related hospitalizations. Objective To compare the effectiveness of a consumer-targeted, pharmacist-led educational intervention vs usual care on discontinuation of inappropriate medication among community-dwelling older adults. Design, Setting, and Participants A cluster randomized trial (D-PRESCRIBE [Developing Pharmacist-Led Research to Educate and Sensitize Community Residents to the Inappropriate Prescriptions Burden in the Elderly]) that recruited community pharmacies in Quebec, Canada, from February 2014 to September 2017, with follow-up until February 2018, and randomly allocated them to intervention or control groups. Patients included were adults aged 65 years and older who were prescribed 1 of 4 Beers Criteria medications (sedative-hypnotics, first-generation antihistamines, glyburide, or nonsteroidal anti-inflammatory drugs), recruited from 69 community pharmacies. Patients were screened and enrolled before randomization. Interventions Pharmacists in the intervention group were encouraged to send patients an educational deprescribing brochure in parallel to sending their physicians an evidence-based pharmaceutical opinion to recommend deprescribing. The pharmacists in the control group provided usual care. Randomization occurred at the pharmacy level, with 34 pharmacies randomized to the intervention group (248 patients) and 35 to the control group (241 patients). Patients, physicians, pharmacists, and evaluators were blinded to outcome assessment. Main Outcomes and Measures Discontinuation of prescriptions for inappropriate medication at 6 months, ascertained by pharmacy medication renewal profiles. Results Among 489 patients (mean age, 75 years; 66% women), 437 (89%) completed the trial (219 [88%] in the intervention group vs 218 [91%] in the control group). At 6 months, 106 of 248 patients (43%) in the intervention group no longer filled prescriptions for inappropriate medication compared with 29 of 241 (12%) in the control group (risk difference, 31% [95% CI, 23% to 38%]). In the intervention vs control group, discontinuation of inappropriate medication occurred among 63 of 146 sedative-hypnotic drug users (43.2%) vs 14 of 155 (9.0%), respectively (risk difference, 34% [95% CI, 25% to 43%]); 19 of 62 glyburide users (30.6%) vs 8 of 58 (13.8%), respectively (risk difference, 17% [95% CI, 2% to 31%]); and 19 of 33 nonsteroidal anti-inflammatory drug users (57.6%) vs 5 of 23 (21.7%), respectively (risk difference, 35% [95% CI, 10% to 55%]) (P for interaction = .09). Analysis of the antihistamine drug class was not possible because of the small sample size (n = 12). No adverse events requiring hospitalization were reported, although 29 of 77 patients (38%) who attempted to taper sedative-hypnotics reported withdrawal symptoms. Conclusions and Relevance Among older adults in Quebec, a pharmacist-led educational intervention compared with usual care resulted in greater discontinuation of prescriptions for inappropriate medication after 6 months. The generalizability of these findings to other settings requires further research. Trial Registration ClinicalTrials.gov Identifier: NCT02053194
Appendix A. Names and ATC codes of drugs included in the analysis APPENDIX B. ICD–9 codes for antidepressant treatment indication APPENDIX C. Positive likelihood ratio (LR+) and negative likelihood ratio (LR−) of administrative diagnostic codes for the seven most common treatment indications, by antidepressant class
IntroductionData from electronic medical records is now readily available and records information needed in pharmacoepidemiological studies not usually found in administrative data such as risk factors and biometrics. Yet, EMR data leads to measurement error due to primary non-adherence. Bayesian bias correction could provide corrected estimates from administrative data. Objectives and ApproachWe present a method for correcting risk estimates from EMR data using linked data. In our example, we estimate the risk of cardiovascular events from oral-hypoglycemics in patients with type-2 diabetes in Boston, Quebec, and the UK between 2009 and 2012. Using linked EMR and administrative data in Quebec, we compute a positive and negative predicting value of prescription on dispensation for each class of oral-hypoglycemics. The cardiovascular risk is then analysed using a bayesian Weibull survival model adjusted for potential confounders. A similar model is then computed that accounts for exposure measurement error using the PPV and NPV. ResultsThe Quebec and Boston cohorts have similar sizes with 1197 and 2346 patients, but the UK was bigger at 41370 patients. In Quebec's data, there were important differences in PPV and NPV by class of oral-hypoglycemics with PPVs for Biguanides at 0.81, Sulphonylureas at 0.65, and others at 0.50. The pattern for NPV differed with the same classes having respectively values of 0.56, 0.97, and 0.99. Estimates from the naïve model are typical of similar analysis but compared to their correction, they were generally overprecise and biased towards the null. The adjusted estimated were adequately representing the increased uncertainty with hazard ratios for Sulphonylureas going from 1.72 (1.22, 2.41) to 3.19 (1.36, 5.93), and from 1.09 (0.86, 1.39) to 1.05 (0.45, 2.16) for no drugs Conclusion/ImplicationsBayesian adjustment for measurement error allowed us to use linked data to regenerate uncertainty and to correct the bias in our risk estimates. Our approach was impacted by the observed low predictive value of prescribing, by reduced transportability of our PPV and NPV estimates, and other sources of bias.
Introduction Identifying strategies to prevent hospital readmissions remains elusive since the reasons for returning to hospital can include a number of interlinked patient, health provider and system level factors. The impact of patient medications are of significant interest since a large proportion of re-admissions are related to adverse drug events. Objectives and Approach The objective was to determine which factors are associated with non-adherence to in-hospital medications and the impact of non-adherence on re-hospitalization, emergency department visits and death in the 30-days post discharge for patients admitted at two tertiary care academic hospitals in Montreal, Quebec between October 2014 and May 2016. Non-adherence to in-hospital changes was measured by comparing patient discharge prescriptions (patient chart) to medications filled in community 30-days post-discharge (dispensing data) and included i) community medications stopped in-hospital and filled post-discharge, ii) community medications modified in-hospital but not filled at the modified daily-dose, and iii) new medications not filled post-discharge. Results Among 2,895 included patients, mean age was 70 (SD 15) and 58% were males. A median of 4 in-hospital medication changes were made (IQR:3-6) and 54% of patients were non-adherent to at least one change. Multivariable Poisson models suggested that the most important factor associated with the number of new medications not filled post discharge was out of pocket cost; for each additional $10 increase in costs there was a 20% increase in the number of new medications not filled. Multivariable time-varying Cox models suggested that in patients who filled medications post-discharge, selective non-adherence to new and discontinued medications reduced the risk adverse health outcomes in 30-days, while not filling any medications post discharge more than doubled the risk of an adverse event in 30-days. Conclusion/Implications Not only did the majority of patients not follow all medication changes that were made during hospitalization, the extent to which this occurred significantly impacted the risk of hospital re-admissions and ED visits. Policy and patient level interventions should be developed specifically targeting barriers for adherence to medication changes.
IntroductionOver the past 20 years, increases in opioid prescribing rates and average prescription volumes have been documented in both the United States and Canada. As prescription opioid use and overdose has steadily increased in North America, a dramatic rise in hospitalizations resulting from opioid poisonings has also been witnessed. Objectives and ApproachTo describe opioid utilization patterns for patients admitted to a tertiary care hospital in Montreal between 2014 and 2016, and to estimate the association between opioid use and risk of adverse health outcomes in the 90-days post discharge. Opioid utilization was measured using medication dispensing data from the provincial healthcare databases (RAMQ) while hospital/ED visits were obtained from RAMQ medical services. Patient characteristics and discharge prescriptions were obtained from the hospital chart. Time-varying utilization of opioids was modeled as: 1) current use, 2) cumulative duration of past use, and 3) duration of use of past 10 days, using Cox models. ResultsOf the 3,308 included patients mean age was 70 (SD 14), 57% were male and 47% were discharged from surgical units. 856 (26%) patients had a history of opioid use in the 1-year prior to admission, 1528 (46%) were prescribed an opioid at discharge and 1481 (45%) filled an opioid in the 90-days post discharge. Among patients prescribed an opioid at discharge, 79\% filled their prescription post discharge, where opioid naïve patients were less likely to fill their prescriptions compared to those with a history of opioid use (40% vs 81%). Our multivariable Cox models suggested that cumulative duration of opioid exposure in the past 10 days was associated with a 10% increased risk of ED visits and hospitalizations. Conclusion/ImplicationsPatients with a history of opioid use were more likely to both receive an opioid prescription at hospital discharge and fill their prescription post-discharge. Our findings suggest that longer-term utilization patterns of these medications after hospitalization may increase risk of re-admissions and ED visits.
Objective: To determine the proportion of family medicine patients unwilling to allow their eHealth data to be used for research purposes, and evaluate how patient characteristics and the relevance of research impact that decision. Design: Cross-sectional questionnaire. Setting: Acute care respiratory clinic or an outpatient family medicine clinic in Montreal, Quebec. Participants: Four hundred seventy-four waiting room patients recruited via convenience sampling. Main Outcome Measures: A self-administered questionnaire collected data on age, gender, employment status, education, mother tongue and perceived health status. The main outcome of was self-reported relevance of three research scenarios and willingness or refusal to share their anonymized data. Responses were compared for family practice vs. specialty care patients. Results: The questionnaire was completed by 229 family medicine respondents and 245 outpatient respondents. Almost a quarter of all respondents felt the research was not relevant. Family medicine patients (15.7%) were unwilling to allow their data to be used for at least one scenario vs. 9.4% in the outpatient clinic. Lack of relevance (OR 11.55; 95% CI 5.12–26.09) and being in family practice (OR 2.13; 95% CI 1.06–4.27) increased the likelihood of refusal to share data for research. Conclusion: Family medicine patients were somewhat less willing to share eHealth data, but the overall refusal rate indicates a need to better engage patients in understanding the significance of full access to eHealth data for the purposes of research. Personal relevance of the research had a strong impact on the responses arguing for better efforts to make research more pertinent to patients.
Objectives: The primary objective of this study was to identify the predictors of new-onset psychological distress available in routinely collected administrative health databases for women diagnosed with breast cancer. The secondary objective was to explore whether the predictors vary based on the period of cancer care. Methods: A population-based cohort study followed 16,495 female patients with newly diagnosed breast cancer who did not experience psychological distress during the 14 months before breast cancer surgery. The incidence of psychological distress was reported overall and by type of mental health problem. Time-varying Cox proportional hazards models were developed to identify predictors of new-onset psychological distress during 2 key periods of cancer care: (1) hospital-based treatment during which women undergo treatment with breast surgery, chemotherapy, and/or radiation, and (2) 1-year transitional survivorship when women begin follow-up care. Results: The incidence of psychological distress was 16% within each period. Anxiety was present in 85.1% and 65.5% of new cases during hospital-based treatment and transitional survivorship, respectively. Predictors during both periods were younger age, receipt of axillary lymph node dissection, rheumatologic disease, and baseline menopausal symptoms, as well as new opioid dispensations, emergency department visits, and hospital contacts that occurred during follow-up. Other predictors varied based on the period of cancer care. More advanced breast cancer and type of treatment were associated with onset of psychological distress during hospital-based treatment. Psychological distress during transitional survivorship was predicted by diagnosis of localized breast disease, shorter duration of hospital-based treatment, receipt of additional hospital-based treatment in survivorship, and newly diagnosed comorbidities or symptoms. Conclusions: This study identified the predictors of new-onset psychological distress available in routinely collected administrative health databases, and showed how predictors change between hospital-based treatment and transitional survivorship periods. The results highlight the importance of developing predictive models tailored to the period of cancer care.
IntroductionThe opioid epidemic in North America has, in part, been attributed to an increase in opiate use for non-cancer pain and the prescription of more potent molecules. In contrast, the United Kingdom appears unaffected by this crisis, possibly because of differences in primary care prescribing, or health system policies. Objective To determine if there are differences in opiate prescribing for new users in primary care in the United Kingdom, United States, and Canada. ApproachElectronic health record data from Quebec, Canada (MOXXI), the United States (Partners Health Care, Boston MA), and the United Kingdom (CPRD random sample of 600,000) were used to identify new users of opiates (no prior prescription in 2 years), at least 18 years old between 2006-2016. Cancer patients were excluded after harmonizing equivalent READ and ICD9/10 codes. Generic drug names in each jurisdiction were mapped to the WHO ATC classification, and characterized using morphine milligram equivalents (MME). ResultsOverall 655,877 new users were identified, of whom 78% of 58,286 (U.S.), 88% of 6,251 (Canada), and 96% of 600,000 (UK) were non-cancer patients. Mean age of new users was 49 (SD 16) in the US, 57 (SD 16) in Canada, and 52 (SD 19) in the UK. 57.6% (UK) to 67.3% (US) of new users were women. In the UK, 86.5% of patients were started on codeine (MME:0.15), compared to 43.9% in Canada and 8.5% in the U.S. In the U.S 65.0\% were started on oxycodone (MME:1.5), and 10.9% on hydrocodone (MME:1). In Canada, tramadol (18.2%; MME: 0.1) followed by oxycodone (13.2%) were the next most commonly prescribed drugs. Conclusion/ImplicationsSubstantial differences in opioid prescribing practices for non-cancer pain were observed between the UK and Canadian and United States sites. The predilection to start patients on more potent opiates in North America may be a contributing cause to the opiate epidemic.
- Jul 2018
Purpose: Older Breast Cancer (BC) survivors are an increased risk of osteoporosis due to natural aging and long-term cancer treatment-related toxicity. It is well known that anti-estrogen therapy (AET), especially aromatase inhibitors (AI), is associated with rapid bone loss and thus increases the risk of osteoporosis. This study characterizes patterns and predictors of receiving guideline-recommended bone densitometry (BD) screening at AET initiation. Methods: A retrospective cohort study (1998-2012) of all women ≥65 years of age initiating AET was designed using claims data from Quebec's universal health care. Associations with BD screening were estimated using a generalized estimating equations regression model, adjusting for clustering of patients within physicians. Results: Among 16,480 women initiating AET, 36.1% received a baseline BD. Among AI users, the rate was 58.4%. In the multivariate analysis, age, lower socioeconomic status, tamoxifen use, lack of periodic health exam and having a general practitioner as the AET prescriber were associated with lower odds of BD screening. In terms of quality of care-related variables, lack of guideline-appropriate radiotherapy (OR: 0.69 (95% CI, 0.57-0.83), or chemotherapy consideration (0.82 (95% CI, 0.71-0.94)) and non-adherence to AET (0.76 (95% CI, 0.68-0.84)) were associated with lower odds of receiving BD screening. Women diagnosed with BC after 2003 had significantly better odds of being screened. Conclusion: Despite an increase in rates since 2003, BD screening remains suboptimal, especially for women at higher risk of osteoporosis. Coordination of health care and service-delivery monitoring can potentially optimize long-term management of treatment-related toxicity in older BC survivors.
Background: Peer review is used to determine what research is funded and published, yet little is known about its effectiveness, and it is suspected that there may be biases. We investigated the variability of peer review and factors influencing ratings of grant applications. Methods: We evaluated all grant applications submitted to the Canadian Institutes of Health Research between 2012 and 2014. The contribution of application, principal applicant and reviewer characteristics to overall application score was assessed after adjusting for the applicant's scientific productivity. Results: Among 11 624 applications, 66.2% of principal applicants were male and 64.1% were in a basic science domain. We found a significant nonlinear association between scientific productivity and final application score that differed by applicant gender and scientific domain, with higher scores associated with past funding success and h-index and lower scores associated with female applicants and those in the applied sciences. Significantly lower application scores were also associated with applicants who were older, evaluated by female reviewers only (v. male reviewers only, -0.05 points, 95% confidence interval [CI] -0.08 to -0.02) or reviewers in scientific domains different from the applicant's (-0.07 points, 95% CI -0.11 to -0.03). Significantly higher application scores were also associated with reviewer agreement in application score (0.23 points, 95% CI 0.20 to 0.26), the existence of reviewer conflicts (0.09 points, 95% CI 0.07 to 0.11), larger budget requests (0.01 points per $100 000, 95% CI 0.007 to 0.02), and resubmissions (0.15 points, 95% CI 0.14 to 0.17). In addition, reviewers with high expertise were more likely than those with less expertise to provide higher scores to applicants with higher past success rates (0.18 points, 95% CI 0.08 to 0.28). Interpretation: There is evidence of bias in peer review of operating grants that is of sufficient magnitude to change application scores from fundable to nonfundable. This should be addressed by training and policy changes in research funding.
Purpose Real‐world data for observational research commonly require formatting and cleaning prior to analysis. Data preparation steps are rarely reported adequately and are likely to vary between research groups. Variation in methodology could potentially affect study outcomes. This study aimed to develop a framework to define and document drug data preparation and to examine the impact of different assumptions on results. Methods An algorithm for processing prescription data was developed and tested using data from the Clinical Practice Research Datalink (CPRD). The impact of varying assumptions was examined by estimating the association between 2 exemplar medications (oral hypoglycaemic drugs and glucocorticoids) and cardiovascular events after preparing multiple datasets derived from the same source prescription data. Each dataset was analysed using Cox proportional hazards modelling. Results The algorithm included 10 decision nodes and 54 possible unique assumptions. Over 11 000 possible pathways through the algorithm were identified. In both exemplar studies, similar hazard ratios and standard errors were found for the majority of pathways; however, certain assumptions had a greater influence on results. For example, in the hypoglycaemic analysis, choosing a different variable to define prescription end date altered the hazard ratios (95% confidence intervals) from 1.77 (1.56‐2.00) to 2.83 (1.59‐5.04). Conclusions The framework offers a transparent and efficient way to perform and report drug data preparation steps. Assumptions made during data preparation can impact the results of analyses. Improving transparency regarding drug data preparation would increase the repeatability, reproducibility, and comparability of published results.
Objective Physicians commonly prescribe antidepressants for indications other than depression that are not evidence-based and need further evaluation. However, lack of routinely documented treatment indications for medications in administrative and medical databases creates a major barrier to evaluating antidepressant use for indications besides depression. Thus, the aim of this study was to derive a model to predict when primary care physicians prescribe antidepressants for indications other than depression and to identify important determinants of this prescribing practice. Methods Prediction study using antidepressant prescriptions from January 2003–December 2012 in an indication-based electronic prescribing system in Quebec, Canada. Patients were linked to demographic files, medical billings data, and hospital discharge summary data to create over 370 candidate predictors. The final prediction model was derived on a random 75% sample of the data using 3-fold cross-validation integrated within a score-based forward stepwise selection procedure. The performance of the final model was assessed in the remaining 25% of the data. Results Among 73,576 antidepressant prescriptions, 32,405 (44.0%) were written for indications other than depression. Among 40 predictors in the final model, the most important covariates included the molecule name, the patient’s education level, the physician’s workload, the prescribed dose, and diagnostic codes for plausible indications recorded in the past year. The final model had good discrimination (concordance (c) statistic 0.815; 95% CI, 0.787–0.847) and good calibration (ratio of observed to expected events 0.986; 95% CI, 0.842–1.136). Conclusion In the absence of documented treatment indications, researchers may be able to use health services data to accurately predict when primary care physicians prescribe antidepressants for indications other than depression. Our prediction model represents a valuable tool for enabling researchers to differentiate between antidepressant use for depression versus other indications, thus addressing a major barrier to performing pharmacovigilance research on antidepressants.
Objective: (1) To describe the usage of medication data from the Health Information Exchange (HIE) at the health care system level in the province of Quebec; (2) To assess the accuracy of the medication list obtained from the HIE. Methods: A descriptive study was conducted utilizing usage data obtained from the Ministry of Health at the individual provider level from January 1 to December 31, 2015. Usage patterns by role, type of site, and tool used to access the HIE were investigated. The list of medications of 111 high risk patients arriving at the emergency department of an academic healthcare center was obtained from the HIE and compared with the list obtained through the medication reconciliation process. Results: There were 31 022 distinct users accessing the HIE 11 085 653 times in 2015. The vast majority of pharmacists and general practitioners accessed it, compared to a minority of specialists and nurses. The top 1% of users was responsible of 19% of access. Also, 63% of the access was made using the Viewer application, while using a certified electronic medical record application seemed to facilitate usage. Among 111 patients, 71 (64%) had at least one discrepancy between the medication list obtained from the HIE and the reference list. Conclusions: Early adopters were mostly in primary care settings, and were accessing it more frequently when using a certified electronic medical record. Further work is needed to investigate how to resolve accuracy issues with the medication list and how certain tools provide different features.
Objective: To develop an enriched set of core competencies for health services and policy research (HSPR) doctoral training that will help graduates maximize their impact across a range of academic and nonacademic work environments and roles. Data sources/study setting: Data were obtained from multiple sources, including literature reviews, key informant interviews, stakeholder consultations, and Expert Working Group (EWG) meetings between January 2015 and March 2016. The study setting is Canada. Study design: The study used qualitative methods and an iterative development process with significant stakeholder engagement throughout. Data collection/extraction methods: The literature reviews, key informant interviews, existing data on graduate career trajectories, and EWG deliberations informed the identification of career profiles for HSPR graduates and the competencies required to succeed in these roles. Stakeholder consultations were held to vet, refine, and validate the competencies. Principal findings: The EWG reached consensus on six sectors and eight primary roles in which HSPR doctoral graduates can bring value to employers and the health system. Additionally, 10 core competencies were identified that should be included or further emphasized in the training of HSPR doctoral students to increase their preparedness and potential for impact in a variety of roles within and outside of traditional academic workplaces. Conclusion: The results offer an expanded view of potential career paths for HSPR doctoral graduates and provide recommendations for an expanded set of core competencies that will better equip graduates to maximize their impact on the health system.
Table S1. Mean absolute prediction errors (MAPE), proportions of subjects with absolute prediction errors (APE) within clinical acceptability and measurement error, and mean squared absolute prediction errors (MSAPE), for each of eight strata. Reporting on subjects whose last HbA1c value excluded from analysis. Figure S1a. Mean absolute prediction errors for strata (A) new users on metformin only medication, (B) female new users, (C) female metformin‐only new users, (D) female metformin‐only new users aged a) ≤ 58, b) 59–70, c) ≥ 71 years, (E) new users with a) single medication change and b) two or more changes in medication, (F) new users whose distance in time between missing HbA1c measure and its closest neighbour was a) > 50th percentile b) > 75th percentile and c) > 90th percentile, where prediction error is absolute difference between predicted and actual values of last HbA1c observation and n is number of subjects being reported on, whose last value was excluded. Figure S1b. Proportion (%) with absolute prediction errors < 0.5 (clinical important difference) for strata (A) new users on metformin only medication, (B) female new users, (C) female metformin‐only new users, (D) female metformin‐only new users aged a) ≤ 58, b) 59–70, c) ≥ 71 years, (E) new users with a) single medication change and b) two or more changes in medication, (F) new users whose distance in time between missing HbA1c measure and its closest neighbour was a) > 50th percentile b) > 75th percentile and c) > 90th percentile, where prediction error is absolute difference between predicted and actual values of last HbA1c observation and n is number of subjects being reported on, whose last value was excluded. Figure S1c. Proportion (%) with absolute prediction errors < 0.4 (measurement error) for strata (A) new users on metformin only medication, (B) female new users, (C) female metformin‐only new users, (D) female metformin‐only new users aged a) ≤ 58, b) 59–70, c) ≥ 71 years, (E) new users with a) single medication change and b) two or more changes in medication, (F) new users whose distance in time between missing HbA1c measure and its closest neighbour was a) > 50th percentile b) > 75th percentile and c) > 90th percentile, where prediction error is absolute difference between predicted and actual values of last HbA1c observation and n is number of subjects being reported on, whose last value was excluded. Figure S1d. Mean squared absolute prediction errors for strata (A) new users on metformin only medication, (B) female new users, (C) female metformin‐only new users, (D) female metformin‐only new users aged a) ≤ 58, b) 59–70, c) ≥ 71 years, (E) new users with a) single medication change and b) two or more changes in medication, (F) new users whose distance in time between missing HbA1c measure and its closest neighbour was a) > 50th percentile b) > 75th percentile and c) > 90th percentile, where prediction error is absolute difference between predicted and actual values of last HbA1c observation and n is number of subjects being reported on, whose last value was excluded. Figure S2a. Mean absolute prediction errors for subgroups (from stratum G) of new users with 2–3, 4–5, 6–8 or >8 HbA1c measures for both stable (< 2 HbA1c units) and unstable disease (≥ 2 HbA1c units), where prediction error is absolute difference between predicted and actual values of last HbA1c observation and n is number of subjects being reported on, whose last value was excluded. Figure S2b. Proportion (%) with absolute prediction errors < 0.5 (clinical important difference) for subgroups (from stratum G) of new users with 2–3, 4–5, 6–8 or >8 HbA1c measures for both stable (< 2 HbA1c units) and unstable disease (≥ 2 HbA1c units), where prediction error is absolute difference between predicted and actual values of last HbA1c observation and n is number of subjects being reported on, whose last value was excluded. Figure S2c. Proportion (%) with absolute prediction errors < 0.4 (measurement error) for subgroups (from stratum G) of new users with 2–3, 4–5, 6–8 or >8 HbA1c measures for both stable (< 2 HbA1c units) and unstable disease (≥ 2 HbA1c units), where prediction error is absolute difference between predicted and actual values of last HbA1c observation and n is number of subjects being reported on, whose last value was excluded. Figure S2d. Mean squared absolute prediction errors for subgroups (from stratum G) of new users with 2–3, 4–5, 6–8 or >8 HbA1c measures for both stable (< 2 HbA1c units) and unstable disease (≥ 2 HbA1c units), where prediction error is difference between absolute predicted and actual values of last HbA1c observation and n is number of subjects being reported on, whose last value was excluded. Table S2. Mean absolute prediction errors (MAPE), proportions of subjects with absolute prediction errors (APE) within clinical acceptability and measurement error, and mean squared absolute prediction errors (MSAPE), for each of eight strata. Reporting on subjects whose middle HbA1c value excluded from analysis. Figure S3a. Mean absolute prediction errors for strata (A) new users on metformin only medication, (B) female new users, (C) female metformin‐only new users, (D) female metformin‐only new users aged a) ≤ 58, b) 59–70, c) ≥ 71 years, (E) new users with a) single medication change and b) two or more changes in medication, (F) new users whose distance in time between missing HbA1c measure and its closest neighbour was a) >50th percentile b) >75th percentile and c) >90th percentile, where prediction error is absolute difference between predicted and actual values of middle HbA1c observation and n is number of subjects being reported on, whose middle value was excluded. Figure S3b. Proportion (%) with absolute prediction errors < 0.5 (clinical important difference) for strata (A) new users on metformin only medication, (B) female new users, (C) female metformin‐only new users, (D) female metformin‐only new users aged a) ≤ 58, b) 59–70, c) ≥ 71 years, (E) new users with a) single medication change and b) two or more changes in medication, (F) new users whose distance in time between missing HbA1c measure and its closest neighbour was a) > 50th percentile b) > 75th percentile and c) > 90th percentile, where prediction error is absolute difference between predicted and actual values of middle HbA1c observation and n is number of subjects being reported on, whose middle value was excluded. Figure S3c. Proportion (%) with absolute prediction errors < 0.4 (measurement error) for strata (A) new users on metformin only medication, (B) female new users, (C) female metformin‐only new users, (D) female metformin‐only new users aged a) ≤ 58, b) 59–70, c) ≥ 71 years, (E) new users with a) single medication change and b) two or more changes in medication, (F) new users whose distance in time between missing HbA1c measure and its closest neighbour was a) > 50th percentile b) > 75th percentile and c) > 90th percentile, where prediction error is absolute difference between predicted and actual values of middle HbA1c observation and n is number of subjects being reported on, whose middle value was excluded. Figure S3d. Mean squared absolute prediction errors for strata (A) new users on metformin only medication, (B) female new users, (C) female metformin‐only new users, (D) female metformin‐only new users aged a) ≤ 58, b) 59–70, c) ≥ 71 years, (E) new users with a) single medication change and b) two or more changes in medication, (F) new users whose distance in time between missing HbA1c measure and its closest neighbour was a) > 50th percentile b) > 75th percentile and c) > 90th percentile, where prediction error is absolute difference between predicted and actual values of middle HbA1c observation and n is number of subjects being reported on, with middle value excluded. Figure S4a. Mean absolute prediction errors for subgroups (from stratum G) of new users with 2–3, 4–5, 6–8 or >8 HbA1c measures for both stable (< 2 HbA1c units) and unstable disease (≥ 2 HbA1c units), where prediction error is absolute difference between predicted and actual values of middle HbA1c observation and n is number of subjects being reported on, whose middle value was excluded. Figure S4b. Proportion (%) with absolute prediction errors < 0.5 (clinical important difference) for subgroups (from stratum G) of new users with 2–3, 4–5, 6–8 or >8 HbA1c measures for both stable (< 2 HbA1c units) and unstable disease (≥ 2 HbA1c units), where prediction error is absolute difference between predicted and actual values of middle HbA1c observation and n is number of subjects being reported on, whose middle value was excluded. Figure S4c. Proportion (%) with absolute prediction errors < 0.4 (measurement error) for subgroups (from stratum G) of new users with 2–3, 4–5, 6–8 or >8 HbA1c measures for both stable (< 2 HbA1c units) and unstable disease (≥ 2 HbA1c units), where prediction error is absolute difference between predicted and actual values of middle HbA1c observation and n is number of subjects being reported on, whose middle value was excluded. Figure S4d. Mean squared absolute prediction errors for subgroups (from stratum G) of new users with 2–3, 4–5, 6–8 or >8 HbA1c measures for both stable (< 2 HbA1c units) and unstable disease (≥ 2 HbA1c units), where prediction error is absolute difference between predicted and actual values of middle HbA1c observation and n is number of subjects being reported on, whose middle value was excluded.
- Nov 2017
Objectives: To characterize rates, reasons for, and associated predictors for emergency department (ED) visits after breast cancer (BC) surgery. Methods: All women over 65 years undergoing curative surgery for non-metastatic incident BC (1998-2012) were identified using Quebec's universal healthcare administrative databases. Reasons for ED visits within 45days of operation were reported. Associated factors were estimated using Cox regression. Results: Of 24,463 patients, 12.8% had postoperative ED visits. Most frequent reasons were: superficial infection, noninfectious gastrointestinal, trauma or wound (other than breast), noninfectious respiratory, and breast wound disruption. Significant predictors included localized (aHR, 1.24, CI 1.04-1.49) or regional disease (aHR 1.64, CI 1.41-1.92), mastectomy (aHR 1.22, CI 1.10-1.34), each operation before definitive oncologic control (aHR 1.12, CI 1.03-1.21), lower institutional volume (aHR 1.23, CI 1.09-1.38), having 6-10 prescriptions (aHR 1.23, CI 1.15-1.31) or >10 (aHR 1.53, CI 1.33-1.77), benzodiazepine use (aHR 1.09, CI 1.01-1.18), anticoagulant use (aHR 1.29, CI 1.13-1.46), cardiovascular disease (aHR 1.15, CI 1.05-1.26), diabetes (aHR 1.11, CI 1.00-1.24), past hospitalization (aHR 1.25, CI 1.17-1.34), and lower income (aHR 1.12, CI 1.04-1.20). Conclusion: Identification of risk factors in older patients before BC surgery could help prevent postoperative ED visits.
Background and objective: Many countries require hospitals to implement medication reconciliation for accreditation, but the process is resource-intensive, thus adherence is poor. We report on the impact of prepopulating and aligning community and hospital drug lists with data from population-based and hospital-based drug information systems to reduce workload and enhance adoption and use of an e-medication reconciliation application, RightRx. Methods: The prototype e-medical reconciliation web-based software was developed for a cluster-randomized trial at the McGill University Health Centre. User-centered design and agile development processes were used to develop features intended to enhance adoption, safety, and efficiency. RightRx was implemented in medical and surgical wards, with support and training provided by unit champions and field staff. The time spent per professional using RightRx was measured, as well as the medication reconciliation completion rates in the intervention and control units during the first 20 months of the trial. Results: Users identified required modifications to the application, including the need for dose-based prescribing, the role of the discharge physician in prescribing community-based medication, and access to the rationale for medication decisions made during hospitalization. In the intervention units, both physicians and pharmacists were involved in discharge reconciliation, for 96.1% and 71.9% of patients, respectively. Medication reconciliation was completed for 80.7% (surgery) to 96.0% (medicine) of patients in the intervention units, and 0.7% (surgery) to 82.7% of patients in the control units. The odds of completing medication reconciliation were 9 times greater in the intervention compared to control units (odds ratio: 9.0, 95% confidence interval, 7.4-10.9, P < .0001) after adjusting for differences in patient characteristics. Conclusion: High rates of medication reconciliation completion were achieved with automated prepopulation and alignment of community and hospital medication lists.
Purpose To assess the accuracy of using diagnostic codes from administrative data to infer treatment indications for antidepressants prescribed in primary care. Methods Validation study of administrative diagnostic codes for 13 plausible indications for antidepressants compared with physician‐documented treatment indications from an indication‐based electronic prescribing system in Quebec, Canada. The analysis included all antidepressant prescriptions written by primary care physicians between January 1, 2003 and December 31, 2012 using the electronic prescribing system. Patient prescribed antidepressants were linked to physician claims and hospitalization data to obtain all diagnoses recorded in the past year. Results Diagnostic codes had poor sensitivity for all treatment indications, ranging from a high of only 31.2% (95% CI, 26.8%‐35.9%) for anxiety/stress disorders to as low as 1.3% (95% CI, 0.0%‐5.2%) for sexual dysfunction. Sensitivity was notably worse among older patients and patients with more chronic comorbidities. Physician claims data were a better source of diagnostic codes for antidepressant treatment indications than hospitalization data. Conclusions Administrative diagnostic codes are poor proxies for antidepressant treatment indications. Future work should determine whether the use of other variables in administrative data besides diagnostic codes can improve the ability to predict antidepressant treatment indications.
Abstract Objective To determine if a prototype pharmacists’ services evaluation programme that uses linked community pharmacy claims and health administrative data to measure pharmacists’ performance can be used to identify characteristics of pharmacies providing higher quality of care. Design Population-based cohort study using community pharmacy claims from 1 November 2009 to 30 June 2010. Setting All community pharmacies in Quebec, Canada. Participants 1742 pharmacies dispensing 8 655 348 antihypertensive prescriptions to 760 700 patients. Primary outcome measure Patient adherence to antihypertensive medications. Predictors Pharmacy level: dispensing workload, volume of pharmacist-provided professional services (eg, refusals to dispense, pharmacotherapy recommendations), pharmacy location, banner/chain, pharmacist overlap and within-pharmacy continuity of care. Patient level: sex, age, income, patient prescription cost, new/chronic therapy, single/multiple antihypertensive medications, single/multiple prescribers and single/multiple dispensing pharmacies. Dispensing level: prescription duration, time of day dispensed and antihypertensive class. Multivariate alternating logistic regression estimated predictors of the primary outcome, accounting for patient and pharmacy clustering. Results 9.2% of dispensings of antihypertensive medications were provided to non-adherent patients. Male sex, decreasing age, new treatment, multiple prescribers and multiple dispensing pharmacies were risk factors for increased non-adherence. Pharmacies that provided more professional services were less likely to dispense to non-adherent hypertensive patients (OR: 0.60; 95% CI: 0.57 to 0.62) as were those with better scores on the Within-Pharmacy Continuity of Care Index. Neither increased pharmacists’ services for improving antihypertensive adherence per se nor increased pharmacist overlap impacted the odds of non-adherence. However, pharmacist overlap was strongly correlated with dispensing workload. There was significant unexplained variability among pharmacies belonging to different banners and chains. Conclusions Pharmacy administrative claims data can be used to calculate pharmacy-level characteristics associated with improved quality of care. This study supports the importance of pharmacist’s professional services and continuity of pharmacist’s care.
Purpose: To compare the more complex technique, functional principal component analysis (FPCA), to simpler methods of estimating values of sparse and irregularly spaced continuous variables at given time points in longitudinal data using a diabetic patient cohort from UK primary care. Methods: The setting for this study is the Clinical Practice Research Datalink (CPRD), a UK general practice research database. For 16,034 diabetic patients identified in CPRD, with at least 2 measures in a 30-month period, HbA1c was estimated after temporarily omitting (i) the final and (ii) middle known values using linear interpolation, simple linear regression, arithmetic mean, random effects, and FPCA. Performance of each method was assessed using mean prediction error. The influence on predictive accuracy of (1) more homogeneous populations and (2) number and range of known HbA1c values was explored. Results: When estimating the last observation, the predictive accuracy of FPCA was highest with over half of predicted values within 0.4 units, equivalent to laboratory measurement error. Predictive accuracy improved when estimating the middle observation with almost 60% predicted values within 0.4 units for FPCA. These results were marginally better than that achieved by simpler approaches, such as last-occurrence-carried-forward linear interpolation. This pattern persisted with more homogeneous populations as well as when variability in HbA1c measures coupled with frequency of data points were considered. Conclusions: When estimating change from baseline to prespecified time points in electronic medical records data, a marginal benefit to using the more complex modelling approach of FPCA exists over more traditional methods.
Purpose: Unmanaged distress has been shown to adversely affect survival and quality of life in breast cancer survivors. Fortunately, distress can be managed and even prevented with appropriate evidence-based interventions. Therefore, the objective of this systematic review was to synthesize the published literature around predictors of distress in female breast cancer survivors to help guide targeted intervention to prevent distress. Methods: Relevant studies were located by searching MEDLINE, Embase, PsycINFO, and CINAHL databases. Significance and directionality of associations for commonly assessed candidate predictors (n ≥ 5) and predictors shown to be significant (p ≤ 0.05) by at least two studies were summarized descriptively. Predictors were evaluated based on the proportion of studies that showed a significant and positive association with the presence of distress. Results: Forty-two studies met the target criteria and were included in the review. Breast cancer and treatment-related predictors were more advanced cancer at diagnosis, treatment with chemotherapy, longer primary treatment duration, more recent transition into survivorship, and breast cancer recurrence. Manageable treatment-related symptoms associated with distress included menopausal/vasomotor symptoms, pain, fatigue, and sleep disturbance. Sociodemographic characteristics that increased the risk of distress were younger age, non-Caucasian ethnicity, being unmarried, and lower socioeconomic status. Comorbidities, history of mental health problems, and perceived functioning limitations were also associated. Modifiable predictors of distress were lower physical activity, lower social support, and cigarette smoking. Conclusions: This review established a set of evidence-based predictors that can be used to help identify women at higher risk of experiencing distress following completion of primary breast cancer treatment.
Importance: In 2016 recommendations for safer prescribing practices were circulated to all doctors in one of Canada's largest provinces, by the college of physicians, following a coroner's inquest into a vehicular death related to Z-drug use. We sought to determine how frequently Z-drug prescriptions in our institution were not adhering to these recommendations. Design: Retrospective cohort study. Setting: McGill University Health Centre, an 832-bed tertiary care institution in Montréal, Canada. Participants: All adult non-obstetrical patients admitted between April 1, 2015 and March 31, 2016. Exposure: The receipt of at least one dose of Z-drug as determined by pharmacy records. Main outcomes and measures: Adherence to four recommendations related to starting dose, maximal dose, concomitant drug administration, and duration of use were evaluated. Results: 1,409 unique patients received a Z-drug during 1,783 admissions representing use in 9.3% of non-obstetrical patients. Standing orders were seen in 42% (745/1783) of admissions. Non-conformity with the coroner's recommendations was common. Overall, 672/1783 (38%) admissions involved a patient receiving more than the recommended daily maximum dose (643/999 older patients, 64%). Of 607 admissions which were longer than 10 days, 257 (39%) involved a prescription which exceeded 10 days. Conclusions and relevance: A coroner's recommendation that doctors receive instructions about safe Z-drug prescribing is unprecedented, and was likely required given that use of Z-drugs occurs at doses and durations that often exceed best practice recommendations. Similar interventions may be required in other jurisdictions.
- Apr 2017
Fragmentation of care delivery remains one of the main challenges of healthcare systems. In this study, we examine health information technology (HIT) that is electronically integrated across care physicians, specialists, hospitals, and pharmacists and its impact on quality of care. Drawing on the structure–process–outcome framework, we hypothesize that integrated HIT will have both a direct and an indirect effect on quality of care. By providing more complete information to physicians and allowing them to detect and manage health problems early in the care cycle, integrated HIT is expected to directly improve quality of care. Integrated HIT is also hypothesized to improve quality of care indirectly by increasing continuity of care and reducing the odds of therapeutic errors. We test these hypotheses through a contemporaneous natural experiment of a matched set of 15,626 outpatients receiving ambulatory care through an integrated HIT system and 15,626 outpatients in a control group. The results show that the patients treated with the integrated system had better quality of care in the follow-up period than patients of the control group. Integrated HIT also had an indirect effect on quality of care through an increase in continuity of care. Our study contributes to research by examining how an integrated HIT system—rather than specific HIT components—influences patient-related outcomes, and how this effect is mediated by two key processes: continuity of care and therapeutic duplication errors. Our work also helps decision makers to better understand the impact of HIT in a complex and highly decentralized and fragmented care delivery system.
The use of simulation-based training has established itself in healthcare but its implementation has been varied and mostly limited to technical and non-technical skills training. This article discusses the possibilities of the use of simulation as part of an overarching approach to improving patient safety, and represents the views of the Simnovate Patient Safety Domain Group, an international multidisciplinary expert group dedicated to the improvement of patient safety. The application and integration of simulation into the various facets of a learning healthcare system is discussed, with reference to relevant literature and the different modalities of simulation which may be employed. The selection and standardisation of outcomes is highlighted as a key goal if the evidence base for simulation-based patient safety interventions is to be strengthened. This may be achieved through the establishment of standardised reporting criteria. If such safety interventions can be proven to be effective, financial incentives are likely to be necessary to promote their uptake, with the intention that up-front cost to payers or insurers be recouped in the longer term but reductions in complications and lengths of stay.
The use of simulation-based training has established itself in healthcare but its implementation has been varied and mostly limited to technical and non-technical skills training. This article discusses the possibilities of the use of simulation as part of an overarching approach to improving patient safety, and represents the views of the Simnovate Patient Safety Domain Group, an international multidisciplinary expert group dedicated to the improvement of patient safety. The application and integration of simulation into the various facets of a learning healthcare system is discussed, with reference to relevant literature and the different modalities of simulation which may be employed. The selection and standardisation of outcomes is highlighted as a key goal if the evidence base for simulation-based patient safety interventions is to be strengthened. This may be achieved through the establishment of standardised reporting criteria. If such safety interventions can be proven to be effective, financial incentives are likely to be necessary to promote their uptake, with the intention that up-front cost to payers or insurers be recouped in the longer term but reductions in complications and lengths of stay.
Objective To examine off-label indications for antidepressants in primary care and determine the level of scientific support for off-label prescribing. Design Descriptive study of antidepressant prescriptions written by primary care physicians using an indication based electronic prescribing system. Setting Primary care practices in and around two major urban centres in Quebec, Canada. Participants Patients aged 18 years or older who visited a study physician between 1 January 2003 and 30 September 2015 and were prescribed an antidepressant through the electronic prescribing system. Main outcome measures Prevalence of off-label indications for antidepressant prescriptions by class and by individual drug. Among off-label antidepressant prescriptions, the proportion of prescriptions in each of the following categories was measured: strong evidence supporting use of the prescribed drug for the respective indication; no strong evidence for the prescribed drug but strong evidence supporting use of another drug in the same class for the indication; or no strong evidence supporting use of the prescribed drug and all other drugs in the same class for the indication. Results 106 850 antidepressant prescriptions were written by 174 physicians for 20 920 adults. By class, tricyclic antidepressants had the highest prevalence of off-label indications (81.4%, 95% confidence interval, 77.3% to 85.5%), largely due to a high off-label prescribing rate for amitriptyline (93%, 89.6% to 95.7%). Trazodone use for insomnia was the most common off-label use for antidepressants, accounting for 26.2% (21.9% to 30.4%) of all off-label prescriptions. For only 15.9% (13.0% to 19.3%) of all off-label prescriptions, the prescribed drug had strong scientific evidence for the respective indication. For 39.6% (35.7% to 43.2%) of off-label prescriptions, the prescribed drug did not have strong evidence but another antidepressant in the same class had strong evidence for the respective indication. For the remaining 44.6% (40.2% to 49.0%) of off-label prescriptions, neither the prescribed drug nor any other drugs in the class had strong evidence for the indication. Conclusions When primary care physicians prescribed antidepressants for off-label indications, these indications were usually not supported by strong scientific evidence, yet often another antidepressant in the same class existed that had strong evidence for the respective indication. There is an important need to generate and provide physicians with evidence on off-label antidepressant use to optimise prescribing decisions.
- Jan 2017
- Hawaii International Conference on System Sciences
Background Health care is provided under the conditions in which people live and under the rules and regulations of a prevailing health system. As a consequence, ‘local’ circumstances are an important determinant of the actual care that can be provided and its effects on the health of individuals and populations. This plays in particular, but not exclusively, a role in community-based primary health care. Although this is generally accepted, there is little insight in the impact of the setting and context in which health care is provided on the outcome of care. Aim This paper argues the case to use this natural variation within and between countries as an opportunity to be used as a form of natural experiment in health research. Arguments We argue that analysing and comparing outcomes across settings, that is comparative outcomes of interventions that have been performed under different health care conditions will improve the understanding of how the real-life setting in which health care is provided – including the health system, the socio-economic circumstances and prevailing cultural values – do determine outcome of care. Recommendations To facilitate comparison of research findings across health systems and different socio-economic and cultural contexts, we recommend a more detailed reporting of the conditions and circumstances under which health research has been performed. A set of core variables is proposed for studies in primary health care.
Background Whether Web-based technologies can improve disease self-management is uncertain. My Asthma Portal (MAP) is a Web-based self-management support system that couples evidence-based behavioral change components (self-monitoring of symptoms, physical activity, and medication adherence) with real-time monitoring, feedback, and support from a nurse case manager. Objective The aim of this study was to compare the impact of access to a Web-based asthma self-management patient portal linked to a case-management system (MAP) over 6 months compared with usual care on asthma control and quality of life. Methods A multicenter, parallel, 2-arm, pilot, randomized controlled trial was conducted with 100 adults with confirmed diagnosis of asthma from 2 specialty clinics. Asthma control was measured using an algorithm based on overuse of fast-acting bronchodilators and emergency department visits, and asthma-related quality of life was assessed using the Mini-Asthma Quality of Life Questionnaire (MAQLQ). Secondary mediating outcomes included asthma symptoms, depressive symptoms, self-efficacy, and beliefs about medication. Process evaluations were also included. Results A total of 49 individuals were randomized to MAP and 51 to usual care. Compared with usual care, participants in the intervention group reported significantly higher asthma quality of life (mean change 0.61, 95% CI 0.03 to 1.19), and the change in asthma quality of life for the intervention group between baseline and 3 months (mean change 0.66, 95% CI 0.35 to 0.98) was not seen in the control group. No significant differences in asthma quality of life were found between the intervention and control groups at 6 (mean change 0.46, 95% CI –0.12 to 1.05) and 9 months (mean change 0.39, 95% CI –0.2 to 0.98). For poor control status, there was no significant effect of group, time, or group by time. For all self-reported measures, the intervention group had a significantly higher proportion of individuals, demonstrating a minimal clinically meaningful improvement compared with the usual care group. Conclusions This study supported the use of MAP to enhance asthma quality of life but not asthma control as measured by an administrative database. Implementation of MAP beyond 6 months with tailored protocols for monitoring symptoms and health behaviors as individuals’ knowledge and self-management skills improve may result in long-term gains in asthma control. ClinicalTrial International Standard Randomized Controlled Trial Number (ISRCTN): 34326236; http://www.isrctn.com/ISRCTN34326236 (Archived by Webcite at http://www.webcitation.org/6mGxoI1R7).
Background In women with hormone receptor positive breast cancer, adjuvant endocrine therapy (AET) is associated with a significant survival advantage. Nonadherence is a particular challenge in older women, even though they stand to benefit the most from AET. Therefore, a novel eHealth tool (OPTIMUM) that integrates real-time analysis of health administrative claims data was developed to provide point-of-care decision support for clinicians. Objectives The objectives of the study are to determine the effectiveness of a patient-specific, real-time eHealth alert delivered at point-of-care in reducing rates of AET discontinuation and to understand patient-level factors related to AET discontinuation as well as to assess integration of eHealth alerts regarding deviations from best practices in administration of AET by cancer care teams. Methods A prospective, 2-group controlled comparison pilot study will be conducted at 2 urban, McGill University–affiliated hospitals, the Royal Victoria Hospital and St. Mary’s Hospital. A minimum of 43 patients per study arm will be enrolled through site-level allocation. Follow-up is 1.5 years. Health care professionals at the intervention site will have access to the eHealth tool, which will report to them in real-time medical events with known associations to AET discontinuation, an AET adherence monitor, and a discontinuation alert. Cox proportional hazard ratios with 95% confidence intervals will estimate risks of AET discontinuation. Tests for significance will be 2-sided with a significance level of P<.05. Results This protocol has been approved and funded by the Canadian Institutes of Health Research. The study will evaluate site-level differences between AET discontinuation and AET adherence and assess care team actions at the intervention site. Participant enrollment into this project is expected to start September 2016 with primary data ready to present by June 2018. Conclusion This study will offer an opportunity to verify the feasibility of integrating an eHealth tool that aims to improve the long-term management of breast cancer in a high-risk population by allowing more timely intervention to prevent or rapidly address AET discontinuation.
Background: Older patients with breast cancer represent a vulnerable population at higher risk of experiencing distress and pain, as well as medication-related adverse events from pharmacological treatment of these symptoms. The purpose of this study is to estimate the prevalence of psychotropic (anxiolytic, antidepressant, and antipsychotic) and opioid medication use by older women diagnosed with breast cancer. Methods: This population-based cohort study followed 19,353 women older than 65 years diagnosed with incident, nonmetastatic breast cancer in Quebec, Canada. Data were obtained from provincial, universal health and drug insurance plans covering all medical and pharmaceutical care. Descriptive statistics were calculated for demographic information, breast cancer characteristics, and treatments. Psychotropic and opioid medication use was assessed across the care trajectory: precancer baseline, active care, and first-year survivorship. Results: There was a marked increase in the prevalence of medication use from precancer baseline to active care, followed by a decrease into first-year survivorship. Anxiolytics were used most often across the care trajectory (36.3%, 50.6%, and 44.4% at baseline, active care, and survivorship, respectively). In contrast, antipsychotic and opioid medications were sought primarily during active care (4.5- and 7-fold increases from baseline, respectively), with opioid use during active care increasing dramatically over the study period (9.0% to 40.9% from 1998 to 2010). Unlike other drugs, antidepressant use peaked in active care but persisted into survivorship (14.7%, 22.4%, and 22.3% at baseline, active care, and survivorship, respectively). Conclusions: A substantial proportion of older patients with breast cancer use psychotropic and opioid medications. The different patterns of medication use represent distress and pain experienced by patients across the care trajectory. Given that medication use in this vulnerable population is associated with an increased risk of adverse events, a multidimensional approach integrating psychological interventions in cancer care may better address psychosocial needs of older patients with breast cancer.
The Canadian Institutes of Health Research's Institute of Health Services and Policy Research (IHSPR) has set out an ambitious direction for the next five years. We aim to build the scientific leadership for learning health systems in Canada, tap into the transformative potential of eHealth for Canadian healthcare, find a better system to support aging in the community, and provide research intelligence on the question of how to finance and fund the health system of the future.
- Aug 2016
Objective: To describe the usage of a novel application (The FLOW) that allows mobile devices to be used for rounding and handoffs. Materials and methods: The FLOW provides a view of patient data and the capacity to enter short notes via personal mobile devices. It was deployed using a "bring-your-own-device" model in 4 pilot units. Social network analysis (SNA) was applied to audit trails in order to visualize usage patterns. A questionnaire was used to describe user experience. Results: Overall, 253 health professionals used The FLOW with their personal mobile devices from October 2013 to March 2015. In pediatric and neonatal intensive care units (ICUs), a median of 26-26.5 notes were entered per user per day. Visual network representation of app entries showed that usage patterns were different between the ICUs. In 127 questionnaires (50%), respondents reported using The FLOW most often to enter notes and for handoffs. The FLOW was perceived as having improved patient care by 57% of respondents, compared to usual care. Most respondents (86%) wished to continue using The FLOW. Discussion: This study shows how a handoff and rounding tool was quickly adopted in pediatric and neonatal ICUs in a hospital setting where patient charts were still paper-based. Originally developed as a tool to support informal documentation using smartphones, it was adapted to local practices and expanded to print sign-out documents and import notes within the medicolegal record with desktop computers. Interestingly, even if not supported by the nursing administrative authorities, the level of use for data entry among nurses and doctors was similar in all units, indicating close collaboration in documentation practices in these ICUs.
- Jul 2016
- Medication-Related Falls in Older People
It is widely accepted that electronic prescribing and integrated drug information systems can reduce avoidable errors in prescribing and dispensing. An optimal system that both enhances the uptake by clinicians and improves the safety of prescribing would have features and functionality that includes the integration of patient demographic information, retrieval of all currently active drugs, automated alerts for relevant prescribing problems, integration of electronic prescriptions and drug discontinuation orders into pharmacy systems, and monitoring of patient adherence. Targeted alerts about the overuse of psychotropic drugs in the elderly would be particularly relevant since this situation represents a potentially preventable cause of injuries. This chapter describes the results of a cluster randomized trial using a cutting-edge custom-designed computer-assisted alert for the detection and management of psychotropic drug use. The results showed that physicians who were exposed to the personalized patient alerts were more likely to reduce drugs that were shown to contribute to fall risk, with greater changes occurring when greater risk was displayed. These positive findings are promising, since prescribers will do the right thing when given the information in an easily understandable format. Computerized decision support is a powerful tool to deliver value-added information to consumers and clinicians at the point of care. The next steps involve establishing the standards for delivering evidence into practice, determining what data is needed for personalized predictions, and determining how to deploy decision-support systems across multiple platforms.
Using data from a single Canadian health center, this study examines the annual costs of prescribing laxatives in medical and surgical units. Constipation is common: present in up to 15% of healthy adults,1 39% of medical inpatients on admission, and develops over the course of hospitalization in 43%2 Given the frequency of bowel symptoms and provider diligence in treating constipation, laxative use in the hospital is common. While relatively inexpensive themselves, the indirect costs of laxatives include: pharmacy inventory management and distribution; nursing administration time; a contribution to polypharmacy; and downstream investigations (eg, Clostridium difficile testing) in the case of laxative-induced diarrhea. Evidence supporting the efficacy of certain laxatives is lacking, particularly docusate sodium/calcium,3 and so we quantified local patterns of laxative use, and estimated some of the associated costs.
- Jun 2016
In Reply Unscientific prescribing constitutes 4 of 5 off-label uses,1- 3 and this unscientific prescribing has resulted in a 54% increased risk of adverse drug events compared with on-label uses.2 To crack down on unscientific prescribing, drug regulatory bodies need to demand strong scientific evidence from pharmaceutical companies to safeguard the public.
- May 2016
This pharmacoepidemiology study uses national Canadian electronic medical record data to describe treatment indications for antidepressants and temporal trends in antidepressant prescribing for depression in Quebec, Canada, from 2006 through 2015. Antidepressant use in the United States has increased over the last 2 decades.¹ A suspected reason for this trend is that primary care physicians are increasingly prescribing antidepressants for nondepressive indications, including unapproved (off-label) indications that have not been evaluated by regulatory agencies.² However, the frequency with which physicians prescribe antidepressants for nondepressive indications is unknown because treatment indications are rarely documented. We analyzed the prevalence of treatment indications for antidepressants and assessed temporal trends in antidepressant prescribing for depression.
Objective To evaluate the impact of comparative performance feedback to community pharmacists on provision of professional services and the quality of patients’ medication use. Design Randomised, controlled, single-blind trial. Setting All 1833 community pharmacies in the Quebec province, Canada. Participants 1814 pharmacies not opting out and with more than 5 dispensings of the target medications during the 6-month baseline were randomised by a 2×2 factorial design to feedback first for hypertension adherence (907 control, 907 intervention) followed by randomisation for asthma adherence (791 control, 807 intervention). 1422 of 1814 pharmacies had complete information available during the follow-up for hypertension intervention (706 intervention, 716 control), and 1301 of 1598 had the follow-up information for asthma (657 intervention, 644 control). Intervention Using provincial billing data to measure performance, mailed comparative feedback reported the pharmacy-level percentage of dispensings to patients non-adherent to antihypertensive medications or overusing asthma rescue inhalers. Primary and secondary outcome measures The number of hypertension/asthma services billed per pharmacy and percentage of dispensings to non-adherent patients over the 12 months post intervention. Results Feedback on the asthma measure led to increased provision of asthma services (control 0.2, intervention 0.4, RR 1.58, 95% CI 1.02 to 2.46). However, this did not translate into reductions in patients’ overuse of rescue inhalers (control 45.5%, intervention 44.6%, RR 0.99, 95% CI 0.98 to 1.01). For non-adherence to antihypertensive medications, feedback resulted in no difference in either provision of hypertension services (control 0.7, intervention 0.8, RR 1.25, 95% CI 0.86 to 1.82) or antihypertensive treatment adherence (control 27.9%, intervention 28.0%, RR 1.0, 95% CI 0.99 to 1.00). Baseline performance did not influence results, and there was no evidence of a cumulative effect with repeated feedback. Conclusions Comparative pharmacy performance feedback increased the provision of asthma pharmacists’ services but did not improve the performance on medication-use measures. Billing data can be used to evaluate the impact of billable services rendered by pharmacists on the quality of patients’ medication use.
Background: Health services and policy research is the innovation engine of a health care system. In 2000, the Canadian Institutes of Health Research (CIHR) was formed to foster the growth of all sciences that could improve health care. We evaluated trends in health services and policy research funding, in addition to determinants of funding success. Methods: All applications submitted to CIHR strategic and open operating grant competitions between 2001 and 2011 were included in our analysis. Age, sex, size of research team, critical mass, season, year and research discipline were retrieved from application information. A cohort of 4725 applicants successfully funded between 2001 and 2005 were followed for 5 years to evaluate predictors of continuous funding. Multivariate generalized estimating equation logistic regression was used to estimate predictors of funding success and sustained funding. Results: Between 2001 and 2011, 80 163 applications were submitted to open and strategic grant competitions. Over time, grant applications increased from 327 to 1137 per year, and annual funding increased from $12.6 to $48.0 million. Grant applications from young male researchers were more likely to be funded than those from female researchers (odds ratio [OR] 1.40, 95% confidence interval [CI] 1.01-1.95), as were applications from larger research teams and institutions with a large critical mass. Only 24.0% of scientists whose first funded grant was in health services and policy research had sustained 5-year funding, compared with 52.8% of biomedical scientists (OR 0.34, 95% CI 0.24-0.49). Interpretation: The CIHR has successfully increased the amount of health services and policy research in Canada. To enhance conditions for success, researchers should be encouraged to work in teams, request longer duration grants, resubmit unsuccessful applications and affiliate themselves with institutions with a greater critical mass.
Objective: Evaluate methodological advantages and limitations of an international pharmacosurveillance system based on electronic health records (EHR). Study design: Type 2 diabetes was used as an exemplar. Cohorts of newly treated diabetics were followed in each country from 2009 to 2012 using local electronic health record systems. Cox proportional hazards models were used to assess the risk of cardiovascular events. Settings: Quebec, CA; Massachusetts, USA; Manchester, UK RESULTS: 44,913 newly treated diabetics were identified. 82.6% (U.S.) to 93.1% (Canada) were started on biguanides. 13% of patients failed to fill initial prescriptions. An increased risk of cardiovascular events with sulfonylureas was observed when dispensing (HR: 2.83) versus EHR prescribing (HR: 2.47) data was used. The addition of clinical data produced a 3- to 10-fold increase in co-morbidity for obesity and renal disease, but had no impact on the risk of different hypoglycemic therapies. The risk of cardiovascular events with sulfonylureas was higher in the US (HR: 3.4; 95% CI: 2.1, 5.5) compared to England (HR: 1.3; 95% CI: 1.1, 1.6). Conclusion: An international surveillance system based on electronic health records may provide more timely information about drug safety, and new opportunities to estimate potential sources of bias and health system effects on drug-related outcomes.
- Mar 2016
Background: Earlier studies on the association between antidepressant use and miscarriage have obtained conflicting results after accounting for the role of depression, and none have taken into account the high risk of induced abortions in women using antidepressants. Methods: We identified 41,964 pregnant women delivering between 1998 and 2002 using Quebec's health administration databases. We compared women prescribed antidepressants in the first trimester and with a recorded diagnosis of depression before pregnancy to: (i) women with neither antidepressant use nor a depression diagnosis before or during pregnancy; (ii) women with a depression diagnosis before pregnancy, but no antidepressants prescribed in the first trimester; and (iii) women prescribed hypothyroid medication in the first trimester, but not antidepressants. We used log binomial regression to assess the adjusted relative risk of miscarriage, corrected for induced abortion risk. Results: The miscarriage risk uncorrected for induced abortions was 16%, 10% and 9% for depressed women exposed to antidepressants; unexposed depressed women; and unexposed, non-depressed women, respectively. These decreased to 11%, 8% and 7% after correction for induced abortions. In multivariable analysis, the corrected risk of miscarriage relative to unexposed, non-depressed women was 1.3 (1.1 to 1.5) for antidepressant-exposed women and 1.1 (1.0 to 1.2) for unexposed depressed women. The miscarriage relative risk for antidepressant users compared to unexposed depressed women was thus 1.2 (1.0 to 1.4). Conclusion: Antidepressant use in the first trimester is associated with an increased risk of miscarriage when compared to either non-depressed or depressed unexposed women, even after accounting for induced abortions.
Background Patient information and education, such as decision aids, are gradually moving toward online, computer-based environments. Considerable research has been conducted to guide content and presentation of decision aids. However, given the relatively new shift to computer-based support, little attention has been given to how multimedia and interactivity can improve upon paper-based decision aids. Objective The first objective of this review was to summarize published literature into a proposed classification of features that have been integrated into computer-based decision aids. Building on this classification, the second objective was to assess whether integration of specific features was associated with higher-quality decision making. Methods Relevant studies were located by searching MEDLINE, Embase, CINAHL, and CENTRAL databases. The review identified studies that evaluated computer-based decision aids for adults faced with preference-sensitive medical decisions and reported quality of decision-making outcomes. A thematic synthesis was conducted to develop the classification of features. Subsequently, meta-analyses were conducted based on standardized mean differences (SMD) from randomized controlled trials (RCTs) that reported knowledge or decisional conflict. Further subgroup analyses compared pooled SMDs for decision aids that incorporated a specific feature to other computer-based decision aids that did not incorporate the feature, to assess whether specific features improved quality of decision making. Results Of 3541 unique publications, 58 studies met the target criteria and were included in the thematic synthesis. The synthesis identified six features: content control, tailoring, patient narratives, explicit values clarification, feedback, and social support. A subset of 26 RCTs from the thematic synthesis was used to conduct the meta-analyses. As expected, computer-based decision aids performed better than usual care or alternative aids; however, some features performed better than others. Integration of content control improved quality of decision making (SMD 0.59 vs 0.23 for knowledge; SMD 0.39 vs 0.29 for decisional conflict). In contrast, tailoring reduced quality of decision making (SMD 0.40 vs 0.71 for knowledge; SMD 0.25 vs 0.52 for decisional conflict). Similarly, patient narratives also reduced quality of decision making (SMD 0.43 vs 0.65 for knowledge; SMD 0.17 vs 0.46 for decisional conflict). Results were varied for different types of explicit values clarification, feedback, and social support. Conclusions Integration of media rich or interactive features into computer-based decision aids can improve quality of preference-sensitive decision making. However, this is an emerging field with limited evidence to guide use. The systematic review and thematic synthesis identified features that have been integrated into available computer-based decision aids, in an effort to facilitate reporting of these features and to promote integration of such features into decision aids. The meta-analyses and associated subgroup analyses provide preliminary evidence to support integration of specific features into future decision aids. Further research can focus on clarifying independent contributions of specific features through experimental designs and refining the designs of features to improve effectiveness.
Effective medication reconciliation is critical in reducing the risk of preventable adverse drug events. Medical trainees are often responsible for medication reconciliation on admission, transfer and discharge of the most vulnerable patients; therefore, it is important that trainees are educated on this aspect of quality care. We conducted a systematic review using MEDLINE and EMBASE databases to identify education initiatives targeted at improving trainee skill and knowledge in carrying out medication reconciliation. Studies published in English or French between July 1980 and July 2013, where the primary focus of the article was the role of medical trainees in conducting medication reconciliation, and where trainee-specific data was reported, were included. Included articles must have reported trainee-specific data. Given the anticipated heterogeneity and array of outcomes, we were unable to employ a specific tool in assessing the risk of bias across studies. Seven studies met pre-specified eligibility criteria, indicating the lack of published education initiatives targeted towards improving trainee knowledge and experience. Four described an education intervention targeted towards students completing internal medicine clerkship, while the remaining 3 were implemented among residents. Although no two interventions were the same, 5 out of 7 included an experiential component. Varying success was achieved with medication reconciliation education interventions. While some noted improved competence and/or confidence amongst trainees, namely undergraduate medical students, others noted little effect resulting from the intervention.
We propose an integrated semantic web framework consisting of formal ontologies, web services, a reasoner and a rule engine that together recommend appropriate level of patient-care based on the defined semantic rules and guidelines. The classification of healthcare-associated infections within the HAIKU (Hospital Acquired Infections – Knowledge in Use) framework enables hospitals to consistently follow the standards along with their routine clinical practice and diagnosis coding to improve quality of care and patient safety. The HAI ontology (HAIO) groups over thousands of codes into a consistent hierarchy of concepts, along with relationships and axioms to capture knowledge on hospital-associated infections and complications with focus on the big four types, surgical site infections (SSIs), catheter-associated urinary tract infection (CAUTI); hospital-acquired pneumonia, and blood stream infection. By employing statistical inferencing in our study we use a set of heuristics to define the rule axioms to improve the SSI case detection. We also demonstrate how the occurrence of an SSI is identified using semantic e-triggers. The e-triggers will be used to improve our risk assessment of post-operative surgical site infections (SSIs) for patients undergoing certain type of surgeries (e.g. coronary artery bypass graft surgery (CABG)).
Importance Off-label use of prescription drugs has been identified as an important contributor to preventable adverse drug events (ADEs) in children. Despite concerns regarding adverse outcomes, to date, no systematic investigation of the effects of off-label drug use in adult populations has been performed.Objective To monitor and evaluate off-label use of prescription drugs and its effect on ADEs in an adult population.Design, Setting, and Participants A cohort of 46 021 patients who received 151 305 incident prescribed drugs was assembled from primary care clinics in Quebec, Canada, using the Medical Office of the XXIst Century electronic health record, which supports documentation of treatment indications and treatment outcomes. Prescriptions dispensed from January 1, 2005, through December 30, 2009, were followed up from the date of the prescription to the date the drug use was discontinued, the end of treatment, or the end of follow-up (December 30, 2010). Data were analyzed from January 5, 2012, to March 15, 2015.Exposures Off-label prescription drug use with and without strong scientific evidence.Main Outcomes and Measures Adverse drug events in off-label use with and without strong scientific evidence. Analysis used multivariate marginal Cox proportional hazards regression for clustered data with the drug as the unit of analysis.Results A total of 3484 ADEs were found in the 46 021 study patients, with an incidence rate of 13.2 per 10 000 person-months. The rate of ADEs for off-label use (19.7 per 10 000 person-months) was higher than that for on-label use (12.5 per 10 000 person-months) (adjusted hazard ratio [AHR], 1.44; 95% CI, 1.30-1.60). Off-label use lacking strong scientific evidence had a higher ADE rate (21.7 per 10 000 person-months) compared with on-label use (AHR, 1.54; 95% CI, 1.37-1.72). However, off-label use with strong scientific evidence had the same risk for ADEs as on-label use (AHR, 1.10; 95% CI, 0.88-1.38). The risks for ADEs were higher for drugs approved from 1981 to 1995 (14.4 per 10 000 person-months; AHR, 1.62; 95% CI, 1.45-1.80) and for those used by women (14.3 per 10 000 person-months; AHR, 1.17; 95% CI, 1.06-1.28), patients receiving 5 to 7 drugs (12.1 per 10 000 person-months; AHR, 3.23; 95% CI, 2.66-3.92), and patients receiving cardiovascular drugs (15.9 per 10 000 person-months; AHR, 3.30; 95% CI, 2.67-4.08) and anti-infectives (66.2 per 10 000 person-months; AHR, 6.33; 95% CI, 4.58-8.76). Patients with a 1-unit increase in the continuity of care index had a 19% increase in ADEs (AHR, 1.19; 95% CI, 1.12-1.26).Conclusions and Relevance Off-label use of prescription drugs is associated with ADEs. Caution should be exercised in prescribing drugs for off-label uses that lack strong scientific evidence. Future electronic health records should be designed to enable postmarket surveillance of treatment indications and treatment outcomes to monitor the safety of on- and off-label uses of drugs.
Background: Recent advances in wheeled mobility technology are multiplying opportunities for community integration and improved quality of life. The mobility needs of older wheelchair users are particularly complex due to a constellation of chronic conditions and comorbidities that may compromise optimal use of the device. The purpose of the Mobility Outcomes via Information Technologies (MOvIT) project is to examine the feasibility of automated calls for the systematic monitoring for adverse outcomes associated with wheelchair use. Methods: A two-phase mixed methods approach was used. Phase I involved user-centered development and face validation of a monitoring questionnaire with end-users (seven wheelchair users and five healthcare providers). Phase II tested the feasibility of monitoring outcomes using automated calls to administer the MOvIT questionnaire 1 and 3 months after wheelchair delivery with a prospective cohort of older adults (50-84 years of age). When problems were identified, the computer monitoring system notified a clinical coordinator who followed up with respondents requiring interventions. Feasibility data were extracted from the web database and from individual interviews covering perceived ease of use, usefulness and intention to use the MOvIT questionnaire in the future. Results: The MOvIT monitoring questionnaire developed in phase I tracks nine potential wheelchair-related adverse outcomes considered important for end-users: 1) non-use of wheelchair, 2) pain, 3) skin condition, 4) positioning, 5) wheelchair incidents, 6) psychosocial issues, 7) restricted wheelchair participation, 8) limited wheelchair skills and knowledge, and 9) technical problems. In phase II, 92 individuals who received a wheelchair were eligible, 71 out of 92 accepted (77 %) and 65 out of 71 (92 %) completed the 3-month follow-up. In the sample of 65 participants, a wheelchair-related adverse outcome was confirmed by a rehabilitation professional for 58.5 %, and at least one recommendation was given to 66.2 % during the 3-month monitoring period. A majority of participants found the intervention useful (82.8 %) and said they intended to use the MOvIT monitoring questionnaire in the future (81.5 %). Participants made suggestions to make the calls more adaptive to various ability profiles. Conclusions: Automated calls tailored for individuals with mobility limitations and associated comorbidities are a promising approach to reach clients who need post-rehabilitation support.
There is consensus that the Canadian health system needs to be reformed and that we are failing to achieve this objective (Romanow 2002; The Standing Senate Committee on Social Affairs Science and Technology 2002; Harvey et al. 2013; Lewis 2015). The authors provide a refreshing perspective on how this could be accomplished: a new model for conditional federal investment that is linked to quality, efficiency and outcomes; incentives for improvement that don't invoke dysfunctional competition; and a national system of outcome measurement. It is a well-reasoned and thoughtful set of solutions that could fit with the complex multi-jurisdictional character of Canadian healthcare. The key will be how to get there.
- Jun 2015
Determining the factors that will predict long-term asthma control is essential for improving health outcomes and decreasing the burden on the health care system. Patient-reported outcomes (PROs) on health behaviors can provide valuable information about future asthma control but have rarely been considered in previous analyses. To develop statistical models for evaluating the predictors of long-term asthma control using PROs such as scores of the Asthma Control Test and the Asthma Self-Efficacy Scale. Of 1,437 individuals contacted, 566 (39%) at baseline and 486 (34%) at follow-up completed the questionnaires, including 4 PROs (Asthma Control Test, Asthma Self-Efficacy Scale, Mini-Asthma Quality of Life Questionnaire, and Beliefs about Medication Questionnaire). Long-term asthma control was evaluated by assessing overuse of rescue medication and emergency department visits. A multivariate logistic generalized estimating equation model was fitted to evaluate the possible effect of the studied factors on asthma control. The complete case generalized estimating equation analysis included 286 participants who had complete PROs at the 2 evaluation times. After adjusting for socioeconomic status and smoking status, the Mini-Asthma Quality of Life Questionnaire was a significant predictor of asthma exacerbation. For each 1-point increase on the Mini-Asthma Quality of Life Questionnaire, there was a 0.25 decrease in the odds of a patient's asthma getting out of control. These findings suggest opportunities to decrease the burden on health care by tailoring interventions that combine PROs with other clinical and sociodemographic variables. Copyright © 2015 American College of Allergy, Asthma & Immunology. Published by Elsevier Inc. All rights reserved.
Medication safety for older persons represents an ongoing challenge. Inappropriate prescriptions - those with a high risk of evidence-based harm - persist in up to 25 % of seniors, and account for a significant proportion of avoidable emergency department visits. This project is the sequel to the EMPOWER study, in which a novel consumer-targeted written knowledge transfer tool aimed at empowering older adults to act as drivers of benzodiazepine de-prescription resulted in a 27 % reduction of inappropriate benzodiazepine use at 6-month follow-up (number needed to treat (NNT) = 4). Failure to discontinue in the EMPOWER study was attributable to re-emerging symptoms among participants, prescribing inertia, and lack of knowledge and skills for substituting alternate therapy among physicians and pharmacists. To maximize de-prescription of inappropriate therapy, educational medication-risk reduction initiatives should be tested that simultaneously include patients, physicians and pharmacists. The objective of this trial is to: 1) test the beneficial effect of a new de-prescribing paradigm enlisting pharmacists to transfer knowledge to both patients and prescribers in a 2-pronged approach to reduce inappropriate prescriptions, compared to usual care and 2) evaluate the transferability of the EMPOWER study concept to other classes of inappropriate prescriptions. We intend to conduct a 3-year pragmatic cluster randomized parallel-group controlled trial to test the effect of the new de-prescribing intervention compared to usual care for reducing 4 classes of inappropriate prescriptions from the 2012 Beers criteria among 450 community-dwelling older adults with polypharmacy. Inappropriate prescriptions will include benzodiazepines, sulfonylurea hypoglycemic agents, first generation antihistamines and non-steroidal anti-inflammatory drugs. The study population is community-dwelling older adults recruited from community pharmacies in Quebec, Canada. The intervention was developed based on a systematic review of the evidence for each medication. Participants in the experimental group will receive the written educational program following randomization and have their pharmacist send their physicians an evidence-based pharmaceutical opinion to recommend de-prescription and be followed for a year. The control group will be wait-listed for 6 months. System change to effectively reduce medication risk among community-dwelling seniors requires a coordinated approach targeting physicians, pharmacists and patients. This trial will test the feasibility and effectiveness of a tripartite approach to de-prescribing. Registered via ClinicalTrials.gov on 31 January 2014, identifier: NCT02053194 .
Background: Despite the potential and interest of using technology for delivering specialized psychiatric services to young adults, surprisingly limited attention has been paid to systematically assess their perspectives in this regard. For example, limited knowledge exists on the extent to which young people receiving specialized services for a first-episode psychosis (FEP) are receptive to using new technologies as part of mental health care, and to which types of technology-enabled mental health interventions they are amenable to. Objective: The purpose of this study is to assess the interest of young adults with FEP in using technology to receive mental health information, services, and supports. Methods: This study uses a cross-sectional, descriptive survey design. A convenience sample of 67 participants between the ages of 18 and 35 were recruited from two specialized early intervention programs for psychosis. Interviewer-administered surveys were conducted between December 2013 and October 2014. Descriptive statistics are reported. Results: Among the 67 respondents who completed the survey, the majority (85%, 57/67) agreed or strongly agreed with YouTube as a platform for mental health-related services and supports. The top five technology-enabled services that participants were amenable to were (1) information on medication (96%, 64/67); (2) information on education, career, and employment (93%, 62/67); (3) decision-making tools pertaining to treatment and recovery (93%, 62/67); (4) reminders for appointments via text messaging (93%, 62/67); and (5) information about mental health, psychosis, and recovery in general (91%, 61/67). The top self-reported barriers to seeking mental health information online were lack of knowledge on how to perform an Internet search (31%, 21/67) and the way information is presented online (27%, 18/67). Two thirds (67%; 45/67) reported being comfortable in online settings, and almost half (48%; 32/67) reported a preference for mixed formats when viewing mental health information online (eg, text, video, visual graphics). Conclusions: Young people diagnosed with FEP express interest in using the Internet, social media, and mobile technologies for receiving mental health-related services. Increasing the awareness of young people in relation to various forms of technology-enabled mental health care warrants further attention. A consideration for future research is to obtain more in-depth knowledge on young people's perspectives, which can help improve the design, development, and implementation of integrated technological health innovations within the delivery of specialized mental health care.
- Apr 2015
The objective of this study was to identify physicians' and pharmacists' perceptions of the challenges and benefits to implementing a nationwide electronic prescribing network linking medical clinics and community pharmacies in Quebec, Canada. Forty-nine people (12 general practitioners, 2 managers, 33 community pharmacists, and 2 pharmacy staff members) from 40 points of care (10 primary care clinics (42% of all the connected sites) and 30 community pharmacies (44%)) were interviewed in 2013. Verbatim transcripts were analyzed using thematic analysis. A low level of network use was observed. Most pharmacists processed e-prescriptions by manual entry instead of importing electronically. They reported concerns about potential errors generated by importing e-prescriptions, mainly due to the instruction field. Paper prescriptions were still perceived as the best means for safe and effective processing of prescriptions in pharmacies. Speed issues when validating e-prescription messages were seen as an irritant by physicians, and resulted in several of them abandoning transmission. Displaying the medications based on the dispensing data was identified as the main obstacle to meaningful use of medication histories. Numerous challenges impeded realization of the benefits of this network. Standards for e-prescription messages, as well as rules for message validation, need to be improved to increase the potential benefits of e-prescriptions. Standard drug terminology including the concept of clinical medication should be developed, and the implementation of rules in local applications to allow for the classification and reconciliation of medication lists from dispensing data should be made a priority. © The Author 2015. Published by Oxford University Press on behalf of the American Medical Informatics Association. All rights reserved. For Permissions, please email: email@example.com.
Valid measurement of outcomes such as disease prevalence using health care utilization data is fundamental to the implementation of a "learning health system." Definitions of such outcomes can be complex, based on multiple diagnostic codes. The literature on validating such data demonstrates a lack of awareness of the need for a stratified sampling design and corresponding statistical methods. We propose a method for validating the measurement of diagnostic groups that have: (1) different prevalences of diagnostic codes within the group, and (2) low prevalence. We describe an estimation method whereby (1) low-prevalence diagnostic codes are oversampled, and the positive predictive value (PPV) of the diagnostic group is estimated as a weighted average of the PPV of each diagnostic code, and (2) claims that fall within a low-prevalence diagnostic group are oversampled relative to claims that are not, and bias-adjusted estimators of sensitivity and specificity are generated. We illustrate our proposed method using an example from population health surveillance in which diagnostic groups are applied to physician claims to identify cases of acute respiratory illness. Failure to account for the prevalence of each diagnostic code within a diagnostic group leads to the underestimation of the PPV, because low-prevalence diagnostic codes are more likely to be false positives. Failure to adjust for oversampling of claims that fall within the low-prevalence diagnostic group relative to those that do not leads to the overestimation of sensitivity and underestimation of specificity.This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License, where it is permissible to download and share the work provided it is properly cited. The work cannot be changed in any way or used commercially. http://creativecommons.org/licenses/by-nc-nd/3.0.
Computer-based decision support has been effective in providing alerts for preventive care. Our objective was to determine whether a personalized asthma management computer-based decision support increases the quality of asthma management and reduces the rate of out-of-control episodes. A cluster-randomized trial was conducted in Quebec, Canada among 81 primary care physicians and 4447 of their asthmatic patients. Patients were followed from the first visit for 3-33 months. The physician control group used the Medical Office of the 21st century (MOXXI) system, an integrated electronic health record. A custom-developed asthma decision support system was integrated within MOXXI and was activated for physicians in the intervention group. At the first visit, 9.8% (intervention) to 12.9% (control) of patients had out-of-control asthma, which was defined as a patient having had an emergency room visit or hospitalization for respiratory-related problems and/or more than 250 doses of fast-acting β-agonist (FABA) dispensed in the past 3 months. By the end of the trial, there was a significant increase in the ratio of doses of inhaled corticosteroid use to fast-acting β-agonist (0.93 vs. 0.69: difference: 0.27; 95% CI: 0.02-0.51; P = 0.03) in the intervention group. The overall out-of-control asthma rate was 54.7 (control) and 46.2 (intervention) per 100 patients per year (100 PY), a non-significant rate difference of -8.7 (95% CI: -24.7, 7.3; P = 0.29). The intervention's effect was greater for patients with out-of-control asthma at the beginning of the study, a group who accounted for 44.7% of the 5597 out-of-control asthma events during follow-up, as there was a reduction in the event rate of -28.4 per 100 PY (95% CI: -55.6, -1.2; P = 0.04) compared to patients with in-control asthma at the beginning of the study (-0.08 [95% CI: -10.3, 8.6; P = 0.86]). This study evaluated the effectiveness of a novel computer-assisted ADS system that facilitates systematic monitoring of asthma control status, follow-up of patients with out of control asthma, and evidence-based, patient-specific treatment recommendations. We found that physicians were more likely to use ADS for out-of-control patients, that in the majority of these patients, they were advised to add an inhaled corticosteroid or a leukotriene inhibitor to the patient s treatment regimen, and the intervention significantly increased the mean ratio of inhaled corticosteroids to FABA during follow-up. It also reduced the rate of out-of-control episodes during follow up among patients whose asthma was out-of-control at the time of study entry. Future research should assess whether coupling patient-specific treatment recommendations, automated follow-up, and home care with comparative feedback on quality and outcomes of care can improve guideline adoption and care outcomes. A primary care-personalized asthma management system reduced the rate of out-of-control asthma episodes among patients whose asthma was poorly controlled at the study's onset. Trial Registration Clinicaltrials.gov Identifier: NCT00170248 http://clinicaltrials.gov/ct2/show/NCT00170248?term=Asthma&spons=McGill+University&state1=NA%3ACA%3AQC&rank=2. © The Author 2015. Published by Oxford University Press on behalf of the American Medical Informatics Association.
Background: Medications are an effective intervention for managing and preventing health problems but their benefit can be undermined by non-adherence or adverse drug events (ADEs). Since these issues may be interconnected, efforts to improve non-adherence should also include reduction of ADEs. We have developed the ISTOP-ADE system (Information Systems-enabled Outreach for Preventing Adverse Drug Events), which enables timely monitoring and managing of ADEs. The objectives of this study are to determine whether the ISTOP-ADE system, compared to routine care, will reduce: a) the probability of discontinuing the use of prognosis-altering medications; b) the probability of a patient experiencing a severe ADE; c) the proportion of patients experiencing ADEs, preventable ADEs and ameliorable ADEs; and d) health services utilization. Methods/design: We will randomly assign 2,200 adult ambulatory patients in the province of Québec who have been prescribed an incident medication for the management or prevention of a chronic health condition, to routine care or the ISTOP-ADE system. The ISTOP-ADE system consists of an interactive voice response system (IVRS) paired with pharmacist support. The IVRS will call patients at 3 and 17 days post-prescription to determine if they are experiencing any problems and connect them with a pharmacist when required or desired by the patient. We will evaluate medication persistence at 180 days and health-care utilization using provincial administrative data. Two blinded physicians will ascertain ADE status through a case review. Discussion: We expect the ISTOP-ADE intervention to be feasible and to improve the quality of patient care through improved medication adherence, reduced ADE duration and reduced number of ADEs resulting in an emergency department or inpatient encounter. This in turn could lower health-care utilization, saving costs and lowering the burden on emergency departments and family practices. The success of ISTOP-ADE would present opportunities to implement this intervention through health systems, health insurance agents and commercial pharmacies. Trial registration: ClinicalTrials.gov Identifier: NCT02059044. Date registered: 10 January 2014.
- Sep 2014
A third of breast cancers (BC) occur in women ≥65 years (seniors). Anti-estrogen therapy (AET) significantly reduces BC recurrence and death. This study characterizes determinants of adherence to AET in seniors with BC. Provincial cancer registry and administrative claims data were accessed for all non-metastatic BC diagnosed in Quebec (1998–2005) to identify seniors treated for 5 years with AET. Multivariate linear regression was used to assess the association with patient, disease, and physician characteristics and the 5-year medication possession ratio (MPR) for each patient. 4,715 women were included (mean age: 72.9). Mean MPR was 83.5%, 79% of patients reached a 5-year MPR of ≥80%, and 34% discontinued AET at some point during treatment. The cumulative probability of discontinuation was 33.8% (mean time to discontinuation 2.3 years). The MPR decreased with increasing age and non-BC related hospitalizations, p < 0.05. Each new medication added during the 5-years decreased the MPR by 0.3% (p < 0.05). Women with in situ disease, on antidepressants at baseline, or treated with Tamoxifen had a lower MPR by 6.5% (p = 0.0002), 4.7% (p = 0.003) and 6% (p = 0.001), respectively. Switching AET type was associated with a lower MPR by 5.3% (p = 0.002) if the switch occurred during the first year. Optimal 5-year adherence to AET in seniors with BC remained a challenge and medication discontinuation rates were high. Advanced age, increasing number of hospitalizations, in situ disease, baseline use of antidepressants, Tamoxifen (versus aromatase inhibitors), early switches of AET type, and newly added medications significantly reduced the MPR.
Objectives: This study had two aims: to measure the prevalence of long-term prescribing of high doses of antipsychotics and antipsychotic polypharmacy in a large Canadian province and to estimate the relative contributions of patient-, physician-, and hospital-level factors. Methods: Government hospital discharge, physician, and pharmaceutical claims data were linked to identify individuals with schizophrenia who in 2004 had antipsychotics available to them for at least 11 months. Individuals on a high dose throughout that period, as well as individuals on multiple concurrent antipsychotics (polypharmacy), were identified. Logistic and generalized linear mixed models using patient-, physician-, and hospital-level predictors were estimated. Results: Among the 12,150 individuals identified, 11.9% were on a high dose and 10.4% on antipsychotic polypharmacy continually, with 3.7% in both groups. After adjustment for potential confounders, analyses showed that systematic propensity for physicians to prescribe high doses accounted for 10.9% of the remaining unexplained variance, and physicians as a group who prescribed high doses across a hospital or psychiatry department accounted for 3.0%. For antipsychotic polypharmacy the corresponding percentages were 9.7% and 6.2%. Even after adjustment, the variation in high-dose prescribing and antipsychotic polypharmacy remained substantial. Conclusions: Long-term high-dose and antipsychotic polypharmacy prescribing appeared partly driven by some physicians' and some hospitals' propensities to prescribe in this way independently of patient characteristics. Given the weight of the evidence against high-dose prescribing and antipsychotic polypharmacy, measures addressed to physicians and hospitals most likely to prescribe high doses, antipsychotic polypharmacy, or both should be considered.
Importance The American Board of Internal Medicine Foundation Choosing Wisely Campaign recommends against the use of benzodiazepine drugs for adults 65 years and older. The effect of direct patient education to catalyze collaborative care for reducing inappropriate prescriptions remains unknown.Objective To compare the effect of a direct-to-consumer educational intervention against usual care on benzodiazepine therapy discontinuation in community-dwelling older adults.Design, Setting, and Participants Cluster randomized trial (EMPOWER [Eliminating Medications Through Patient Ownership of End Results] study [2010-2012, 6-month follow-up]). Community pharmacies were randomly allocated to the intervention or control arm in nonstratified, blocked groups of 4. Participants (303 long-term users of benzodiazepine medication aged 65-95 years, recruited from 30 community pharmacies) were screened and enrolled prior to randomization: 15 pharmacies randomized to the educational intervention included 148 participants and 15 pharmacies randomized to the “wait list” control included 155 participants. Participants, physicians, pharmacists, and evaluators were blinded to outcome assessment.Interventions The active arm received a deprescribing patient empowerment intervention describing the risks of benzodiazepine use and a stepwise tapering protocol. The control arm received usual care.Main Outcomes and Measures Benzodiazepine therapy discontinuation at 6 months after randomization, ascertained by pharmacy medication renewal profiles.Results A total of 261 participants (86%) completed the 6-month follow-up. Of the recipients in the intervention group, 62% initiated conversation about benzodiazepine therapy cessation with a physician and/or pharmacist. At 6 months, 27% of the intervention group had discontinued benzodiazepine use compared with 5% of the control group (risk difference, 23% [95% CI, 14%-32%]; intracluster correlation, 0.008; number needed to treat, 4). Dose reduction occurred in an additional 11% (95% CI, 6%-16%). In multivariate subanalyses, age greater than 80 years, sex, duration of use, indication for use, dose, previous attempt to taper, and concomitant polypharmacy (10 drugs or more per day) did not have a significant interaction effect with benzodiazepine therapy discontinuation.Conclusions and Relevance Direct-to-consumer education effectively elicits shared decision making around the overuse of medications that increase the risk of harm in older adults.Trial Registration clinicaltrials.gov Identifier: NCT01148186
Primary nonadherence is probably an important contributor to suboptimal disease management, but methodological challenges have limited investigation of it. To estimate the incidence of primary nonadherence in primary care and the drug, patient, and physician characteristics that are associated with nonadherence. A prospective cohort of patients and all their incident prescriptions from primary care electronic health records between 2006 and 2009 linked to provincial drug insurer data on all drugs dispensed from community-based pharmacies were assembled. Quebec, Canada. 15 961 patients in a primary care network of 131 physicians. Primary nonadherence was defined as not filling an incident prescription within 9 months. Multivariate alternating logistic regression was used to estimate predictors of nonadherence and account for patient and physician clustering. Overall, 31.3% of the 37 506 incident prescriptions written for the 15 961 patients were not filled. Drugs in the upper quartile of cost were least likely to be filled (odds ratio [OR], 1.11 [95% CI, 1.07 to 1.17]), as were skin agents, gastrointestinal drugs, and autonomic drugs, compared with anti-infectives. Reduced odds of nonadherence were associated with increasing patient age (OR per 10 years, 0.89 [CI, 0.85 to 0.92]), elimination of prescription copayments for low-income groups (OR, 0.37 [CI, 0.32 to 0.41]), and a greater proportion of all physician visits with the prescribing physician (OR per 0.5 increase, 0.77 [CI, 0.70 to 0.85]). Patients' rationale for choosing not to fill their prescriptions could not be measured. Primary nonadherence is common and may be reduced by lower drug costs and copayments, as well as increased follow-up care with prescribing physicians for patients with chronic conditions. Canadian Institutes of Health Research.
Decision support systems linked to administrative databases provide a unique opportunity to monitor adherence to guidelines and target disease management strategies towards patients not receiving guideline-based therapy. The objective of this study was to evaluate the discrepancy between actual asthma treatments prescribed by primary care physicians compared to those recommended by evidence-based guidelines using a decision support tool linked to a provincial health administrative database. The drug and medical services information of individuals with asthma was identified from the provincial health database and was pushed through an asthma decision support system (ADSS). Recommendations aimed at optimising asthma treatment were generated on two index dates, 15 September 2007 (index date 1) and 15 March 2008 (index date 2). Primary care settings in a large Canadian metropolitan area. Individuals with asthma and provincial health insurance primary and secondary outcome measures: well controlled asthma. 16 803 eligible individuals were identified on index date 1, and 18 103 on index date 2. The distribution of recommendation categories was similar on both index dates. 94% were classified as well controlled and 7% as not well controlled. Among well-controlled individuals, the largest proportion was in the maintain treatment category (63.8%), followed by the maintain/decrease treatment category (28.2%) and the decrease treatment category (2.7%). Almost all individuals who were not well controlled had the recommendation to increase treatment (88%) with a small proportion in the refer category (1%). The ADSS was able to identify subgroups of patients from an administrative database that could benefit from a medication review and possible change. Decision support systems linked to an administrative database can be used to identify individuals with uncontrolled asthma or prescriptions that deviate from recommended treatment. When connected to the point of care, this can provide an opportunity for physicians to intervene early.
Breast cancer treatment guidelines state that radiotherapy (rt) can reasonably be omitted in selected women 70 years of age and older if they take adjuvant endocrine therapy (aet) for 5 years. We aimed to assess persistence and adherence to aet in women 70 years of age and older, and to examine differences between rt receivers and non-receivers. Quebec's medical service and pharmacy claims databases were used to identify seniors undergoing breast-conserving surgery (1998-2005) and initiating aet. Cox proportional hazards models were used to identify predictors of aet non-persistence. Of 3180 women who initiated aet (mean age: 77.5 years), 28% did not receive rt. During the subsequent 5 years, 32% of patients who initiated aet did not persist, 2% filled only a single prescription, and 22% switched medications. Compared with rt receivers, non-receivers discontinued more often (35.5% vs. 30.1%) and earlier (1.4 years vs. 1.6 years). They also became nonadherent earlier (medication possession ratio < 80% at year 3 vs. at year 5). Predictors of nonpersistence included rt omission [hazard ratio (hr): 1.26; 95% confidence interval (ci): 1.09 to 1.46]; age (hr per decade increase: 1.15; 95% ci: 1.01 to 1.31); new medications (hr per medication: 1.01; 95% ci: 1.00 to 1.02); and hospitalizations during aet, (hr per hospitalization: 1.08; 95% ci: 1.05 to 1.11). In a subanalysis of rt non-receivers, significant predictors included hospitalizations (hr: 1.07; 95% ci: 1.02 to 1.12) and medications at aet start (hr: 0.94; 95% ci: 0.91 to 0.97). Suboptimal use of aet was observed in at least one third of women. In rt non-receivers, aet use was worse than it was in rt receivers. Initiation of new medications and hospitalizations increased the risk of non-persistence.
Medication reconciliation at admission, transfer and discharge has been designated as a required hospital practice to reduce adverse drug events. However, implementation challenges have resulted in poor hospital adherence. The aim of this study was to assess the processes required to carry out medication reconciliation: the health professionals involved, and the tasks and time devoted to medication reconciliation in general hospital settings. A time-and-motion study design was used. Using a systematic sample of patients admitted and discharged from geriatric, medical and surgical units in two academic centers, health professionals involved in medication reconciliation were observed and timed. Descriptive statistics were used to summarize the number of professionals, tasks performed, and mean time devoted. 1 to 3 professionals from 2 disciplines (medicine and pharmacy) were involved in the medication reconciliation process. Geriatric reconciliations took the most time to complete at admission (mean: 92.2 minutes (SD = 44.3)) and discharge (mean: 29.0 minutes (SD = 23.8)), then internal medicine at admission (mean: 46.2 minutes (SD = 21.1) and 19.4 (SD = 11.7) minutes at discharge) and general surgery minutes at discharge (mean: 9.9 (SD = 18.2)). Considerable differences in order, type and number of tasks performed were noted between and within units. Tasks independent of direct patient interaction took more than twice the time required to complete than tasks requiring patient interaction. Lack of coordination, specialized training and agreement on the roles and responsibilities of professionals are among the most probable reasons for work-flow inefficiencies, possibly variability in quality, and time required for the current medication reconciliation process. A better understanding of the admission process in general surgery is required. Standardization and use of electronic tools could improve efficiency and hospital adherence.
Given the high prevalence of diabetes, guidelines are updated frequently to reflect optimal treatment recommendations. Our study aims to measure the response of primary care physicians to changes in choice of initial therapy for patients with type 2 diabetes in relationship to a change in Canadian Diabetes Association (CDA) Guidelines in 2008. We also assessed patients' and physicians' factors which may affect this change. Historical cohort study of primary care physicians' participating in an electronic medical record research network in Quebec, Canada. 111 primary care physicians and 1279 newly treated patients with diabetes with a prescription of an oral hypoglycemic agent (OHA) between January 20 2003 and December 29 2011 were included. Multivariate GEE logistic regression was used to estimate the impact of guideline change on treatment choice controlling for patients' and physicians' characteristics. After the new CDA guidelines, there was an increase in incident use of metformin from 89.7% to 94.6% (OR 1.86, 95% CI 1.20-2.90) with an accompanying reduction in the use of thiazolidinediones (OR 0.21, 95% CI 0.08-0.55), and reduction in the initiation of sulfonylureas (OR 0.78, 95% CI 0.43-1.09). Physicians' attitudes to evidence-based practice did not significantly modify response to a change in guidelines recommendations. However, older patients and those with renal failure were less likely to receive metformin. Metformin initiation in newly diagnosed diabetes patients has increased post 2008 CDA guidelines. However, due to the nature of the study design, we can not determine whether the observed change in metformin prescribing was causally related to the change in the guideline.
- Aug 2013
To (1) compile an initial list of physician-related practice indicators (PRINDs) that contribute to causing or preventing suboptimal care (SOCR) and adverse events (AEs) and (2) determine the extent to which one national exam assessed these PRINDs. In 2009-2010, the authors searched the literature and surveyed 17 physician experts to compile a list of PRINDs of SOCR and avoidable AEs. They then conducted a content analysis of the 2008 and 2009 Medical Council of Canada (MCC) Qualifying Examinations (QEs). The authors identified 92 unique PRINDs, of which 59 were behaviors or decisions expected of all physicians and suitable for assessment on a general medical examination. Of these, 36 (61%) were tested on the 2008 and 2009 MCC QEs. The mean number of PRINDs tested per exam was highest for Part I Knowledge (32.2), followed by Part I clinical decision making (CDM) (18.4) and Part II clinical performance (objective structured clinical examination [OSCE]) (9.8). The percentage of questions or cases per exam testing a PRIND (e.g., 14/36 [39%] for CDM and 5.26/12 [44%] for OSCE) differed from the percentage of the total test score attributed to PRINDs (e.g., 10.8/36 [30%] for CDM and 68.5/1,522.3 [5%] for OSCE). PRINDs represent candidates' abilities to avoid SOCR and AEs and constitute an important aspect of medical practice to be assessed on licensing or certifying examinations to best protect the public. The different scoring methods used to measure such knowledge and skills warrant further consideration.
Errors in community medication histories increase the risk of adverse events. The objectives of this study were to estimate the extent to which access to community-based pharmacy records provided more information about prescription drug use than conventional medication histories. A prospective cohort of patients with public drug insurance who visited the emergency departments (ED) in two teaching hospitals in Montreal, Quebec was recruited. Drug lists recorded in the patients' ED charts were compared with pharmacy records of dispensed medications retrieved from the public drug insurer. Patient and drug-related predictors of discrepancies were estimated using general estimating equation multivariate logistic regression. 613 patients participated in the study (mean age 63.1 years, 59.2% women). Pharmacy records identified 41.5% more prescribed medications than were noted in the ED chart. Concordance was highest for anticoagulants, cardiovascular drugs and diuretics. Omissions in the ED chart were more common for drugs that may be taken episodically. Patients with more than 12 medications (OR 2.92, 95% CI 1.71 to 4.97) and more than one pharmacy (OR 3.85, 95% CI 1.80 to 6.59) were more likely to have omissions in the ED chart. The development of health information exchanges could improve the efficiency and accuracy of information about community medication histories if they enable automated access to dispensed medication records from community pharmacies, particularly for the most vulnerable populations with multiple morbidities. Pharmacy records identified a substantial number of medications that were not in the ED chart. There is potential for greater safety and efficiency with automated access to pharmacy records.
Radiotherapy (RT) after breast conserving surgery (BCS) represents the standard for local control of breast cancer (BC). However, variations in practice persist. We aimed to characterize the rate of RT consideration (or referral) after BCS and identify predictors in Quebec, Canada, where universal health insurance is in place. A historical prospective cohort study using the provincial hospital discharge and medical services databases was conducted. All women with incident, non-metastatic BC (stages I-III) undergoing BCS (1998-2005) were identified. Odds ratios (ORs) and 95 % confidence intervals (CIs) for RT consideration were estimated with a generalized estimating equations regression model, adjusting for clustering of patients within physicians. Of the 27,483 women selected, 90 % were considered for RT and 84 % subsequently received it. Relative to women 50-69 years old, younger and older women were less likely to be considered: ORs of 0.82 (95 % CI 0.73-0.93) and 0.10 (0.09-0.12), respectively. Emergency room visits and hospitalizations unrelated to BC were associated with decreased odds of RT consideration: 0.85 (0.76-0.94) and 0.83 (0.71-0.97). Women with regional BC considered for chemotherapy were more likely to be considered for RT: 3.41 (2.83-4.11). RT consideration odds increased by 7 % (OR of 1.07, 95 % CI 1.03-1.10) for every ten additional BCSs performed by the surgeon in the prior year. Social isolation, comorbidities, and greater distance to a referral center lowered the odds. Demographic and clinical patient-related risk factors, health service use, gaps in other aspects of BC management, and surgeon's experience predicted RT consideration.
To identify the factors that can predict physicians' use of electronic prescribing. All primary care physicians who practised in a single geographic region in Quebec were invited to use a free, advanced, research-based electronic prescribing and drug management system. This natural experiment was studied with an expansion of the Technology Acceptance Model (TAM), which was used to explain early adopters' use of this electronic prescribing technology. Quebec city region. A total of 61 primary care physicians who practised in a single geographic region where there was no electronic prescribing. Actual use of electronic prescribing; physicians' perceptions of and intentions to use electronic prescribing; physician and practice characteristics. During the 9-month study period, 61 primary care physicians located in 26 practice sites used electronic prescribing to write 15 160 electronic prescriptions for 18 604 patients. Physician electronic prescribing rates varied considerably, from a low of 0 to a high of 75 per 100 patient visits, with a mean utilization rate of 30 per 100 patient visits. Overall, 34% of the variance in the use of electronic prescribing was explained by the expanded TAM. Computer experience (P = .001), physicians' information-acquisition style (P = .01), and mean medication use in the practice (P = .02) were significant predictors. Other TAM factors that generally predict new technology adoption (eg, intention to use, perceived ease of use, and perceived usefulness) were not predictive in this study. The adoption of electronic prescribing was a highly challenging task, even among early adopters. The insight that this pilot study provides into the determinants of the adoption of electronic prescribing suggests that novel physician-related factors (eg, information-acquisition style) and practice-related variables (eg, prevalence of medication use) influence the adoption of electronic prescribing.
Guidelines recommend radiotherapy (rt) after breast-conserving surgery (bcs) for optimal control of ductal carcinoma in situ (dcis). The aim of the present study was to characterize the rates of rt consideration and administration, and to identify factors influencing those rates in a cohort of women diagnosed between 1998 and 2005 in Quebec. Quebec's medical service claims and discharge abstract database were used. Using consultation for rt as an indicator for rt consideration, odds ratios (ors) and 95% confidence intervals (cis) were estimated using a generalized estimating equations regression model. Of 4139 women analyzed (mean age: 58 years), 3435 (83%) received a consultation for rt, and 3057 of them (89%) proceeded with treatment. The rate of rt consideration increased by 7.1% over the study period, with notable differences in the various age groups. Relative to women 50-69 years of age, the ors for being considered for rt were, respectively, 0.89 (95% ci: 0.71 to 1.12), 0.71 (95% ci: 0.55 to 0.92), and 0.20 (95% ci: 0.14 to 0.31) for women younger than 50, 70-79, and 80 years of age and older. Distance to a designated breast care centre lowered the probability of rt consideration, but the presence of comorbidities did not. A surgeon's volume of bcss increased the probability of being considered for rt by 7% for every 10 such procedures performed (or: 1.07; 95% ci: 1.04 to 1.11). Consideration for rt has increased over time. However, older women (despite being in good health) and those living far from a designated breast care centre or having a low-case-volume surgeon were less likely to be considered for rt.
Background Although the adoption of health information technology (HIT) has advanced in Canada over the past decade, considerable challenges remain in supporting the development, broad adoption, and effective use of HIT in the public health system. Policy makers and practitioners have long recognized that improvements in HIT infrastructure are necessary to support effective and efficient public health practice. The objective of this study was to identify aspects of health information technology (HIT) policy related to public health in Canada that have succeeded, to identify remaining challenges, and to suggest future directions to improve the adoption and use of HIT in the public health system. Methods A qualitative case study was performed with 24 key stakeholders representing national and provincial organizations responsible for establishing policy and strategic direction for health information technology. Results Identified benefits of HIT in public health included improved communication among jurisdictions, increased awareness of the need for interoperable systems, and improvement in data standardization. Identified barriers included a lack of national vision and leadership, insufficient investment, and poor conceptualization of the priority areas for implementing HIT in public health. Conclusions The application of HIT in public health should focus on automating core processes and identifying innovative applications of HIT to advance public health outcomes. The Public Health Agency of Canada should develop the expertise to lead public health HIT policy and should establish a mechanism for coordinating public health stakeholder input on HIT policy.
OBJECTIVE: To develop and test an educational tool for older adults that increases risk perception about benzodiazepines through knowledge acquisition and change in beliefs. METHODS: A written educational tool was mailed to 144 benzodiazepine consumers aged ≥65 years recruited from community pharmacies. Knowledge and beliefs about inappropriate prescriptions were queried prior to and 1-week after the intervention. Primary outcome was a change in risk perception. Explanatory variables were a change in knowledge and beliefs about medications. Self-efficacy for tapering and intent to discuss discontinuation were also measured. RESULTS: Post-intervention, 65 (45.1%) participants perceived increased risk. Increased risk perceptions were explained by better knowledge acquisition (mean change score 0.9, 95% CI (0.5, 1.3)), and a change in beliefs (BMQ differential mean change score -5.03, 95% CI (-6.4, -3.6)), suggesting elicitation of cognitive dissonance. Self-efficacy for tapering, (mean change score 31.2, 95% CI (17.9, 44.6)), and intent to discuss discontinuation of benzodiazepine with a doctor (83.1% vs 44.3%, p<0.001) were higher among participants who perceived increased risk. CONCLUSION: Risk perception surrounding inappropriate prescriptions can be altered through direct delivery of an educational tool to aging consumers. PRACTICE IMPLICATIONS: Patients should be targeted directly with information to catalyze discontinuation of inappropriate prescriptions.
Background: Currently, far too many older adults consume inappropriate prescriptions, which increase the risk of adverse drug reactions and unnecessary hospitalizations. A health education program directly informing patients of prescription risks may promote inappropriate prescription discontinuation in chronic benzodiazepine users. Methods/design: This is a cluster randomized controlled trial using a two-arm parallel-design. A total of 250 older chronic benzodiazepine users recruited from community pharmacies in the greater Montreal area will be studied with informed consent. A participating pharmacy with recruited participants represents a cluster, the unit of randomization. For every four pharmacies recruited, a simple 2:2 randomization is used to allocate clusters into intervention and control arms. Participants will be followed for 1 year. Within the intervention clusters, participants will receive a novel educational intervention detailing risks and safe alternatives to their current potentially inappropriate medication, while the control group will be wait-listed for the intervention for 6 months and receive usual care during that time period. The primary outcome is the rate of change in benzodiazepine use at 6 months. Secondary outcomes are changes in risk perception, self-efficacy for discontinuing benzodiazepines, and activation of patients initiating discussions with their physician or pharmacist about safer prescribing practices. An intention-to-treat analysis will be followed.The rate of change of benzodiazepine use will be compared between intervention and control groups at the individual level at the 6-month follow-up. Risk differences between the control and experimental groups will be calculated, and the robust variance estimator will be used to estimate the associated 95% confidence interval (CI). As a sensitivity analysis (and/or if any confounders are unbalanced between the groups), we will estimate the risk difference for the intervention via a marginal model estimated via generalized estimating equations with an exchangeable correlation structure. Discussion: Targeting consumers directly as catalysts for engaging physicians and pharmacists in collaborative discontinuation of benzodiazepine drugs is a novel approach to reduce inappropriate prescriptions. By directly empowering chronic users with knowledge about risks, we hope to imitate the success of individually targeted anti-smoking campaigns. Trial registration: ClinicalTrials.gov identifier: NCT01148186.
Objective To assess the usability of a computerised drug monitoring programme for ambulatory patients receiving outpatient prescriptions. Materials and methods A prospective cohort of 200 patients received two automated calls after a new drug prescription (day 3 and day 17) to screen for unfilled prescriptions and medication problems. Usability was assessed objectively and subjectively with the coding of technical (eg, voice recognition problems) and respondent burden (eg, failing to follow instructions) problems observed during the calls, and with an interview 21 days after the prescription. Associations between personal factors, usability and call outcome were examined with logistic regression models. Results The automated calls successfully reached 70.0% of enrolled patients. Older age increased the likelihood of experiencing technical (OR 2.18, 95% CI 1.22 to 3.88) and respondent burden problems (OR 3.32, 95% CI 1.88 to 5.87), as well as unsuccessful calls (OR 2.16, 95% CI 1.19 to 3.91). Patients with higher education experienced less respondent burden problems (OR 0.44, 95% CI 0.21 to 0.91), but they were more prone to have unsuccessful calls (OR 2.65, 95% CI 1.07 to 6.56) and less likely to find them useful (OR 0.23 95% CI 0.08 to 0.68). Older adults perceived the calls as easy to use and useful, although they reported lower intention to use the automated calls in the future (OR 0.32, 95% CI 0.15 to 0.70). Discussion As reported in previous studies, we found that older adults tend to have more difficulty when interacting with automated calls. Evidence about the association between education and usability was mixed. Conclusions Our results highlight practical suggestions to improve the feasibility and usability of automated calls in primary care screening programmes.
Background: Clozapine remains the antipsychotic of choice for people who, having met the criteria for a diagnosis of schizophrenia or a related psychotic disorder, do not respond adequately to other antipsychotic medications. Utilization rates appear highly variable across jurisdictions, with an overall tendency toward underuse. This paper describes patterns of clozapine use in the province of Québec, Canada. Methods: Individuals with a diagnosis of schizophrenia were identified using linked government medical claims and hospitalization records for 2003 and 2004. Linked data on their filled prescriptions in 2004 were then used to determine clozapine-use rates at the level of the province, the region, and the hospital at which individuals received most of their services. Individual predictors of clozapine use were identified using logistic regression. Results: Only 6.7% of the 29,155 individuals identified with schizophrenia received clozapine for six months or longer in 2004. Utilization rates ranged from 3.9 to 9.0% among regions with 1,000 or more people with schizophrenia. Over 8% of 61 hospitals did not prescribe clozapine at all. People with schizophrenia taking clozapine experienced 3.4 fewer days of hospitalization per year than those not taking clozapine-representing a cost offset of about $1,800 per year. Medication costs were higher, however, by about $3,000 per year. Conclusions: Given the increasingly clear benefits of clozapine for people who do not respond to other antipsychotics, measures to increase access to clozapine for people who can benefit from it are likely to be cost effective and are urgently needed.
Healthcare-Associated Infections (HAI) impose a substantial health and financial burden. Surveillance for HAI is essential to develop and evaluate prevention and control efforts. The traditional approaches to HAI surveillance are often limited in scope and efficiency by the need to manually obtain and integrate data from disparate paper charts and information systems. The considerable effort required for discovery and integration of relevant data from multiple sources limits the current effectiveness of HAI surveillance. Knowledge-based systems can address this problem of contextualizing data to support integration and reasoning. In order to facilitate knowledge-based decision making in this area, availability of a reference vocabulary is crucial. The existing terminologies in this domain still suffer from inconsistencies and confusion in different medical/clinical practices, and there is a need for their further improvement and clarification. To develop a common understanding of the infection control domain and to achieve data interoperability in the area of hospital-acquired infections, we present the HAI Ontology (HAIO) to improve knowledge processing in pervasive healthcare environments, as part of the HAIKU (Hospital Acquired Infections – Knowledge in Use) system. The HAIKU framework assists physicians and infection control practitioners by providing recommendations regarding case detection, risk stratification and identification of diagnostic factors.
- Sep 2012
43 Background: Postoperative radiotherapy (RT) after breast conserving surgery (BCS) represents the standard of care for local control of breast cancer (BC). Despite wide dissemination of clinical guidelines, variations in practice persist. Our objective was to identify patient, disease, and physician characteristics that predict lack of consideration for RT after BCS. Methods: Cancer registry data and administrative claims for all BCs diagnosed in Quebec from 1998 to 2005 were collected. Receipt of a consultation for RT in women with non-metastatic BC treated with BCS was measured. Multivariate logistic regression was used to assess the association between patient, disease, and physician characteristics and having an RT consult. Results: 27,483 women were included. Mean age was 59 years, 76.5% had no comorbidities, and 27.6% had stage III BC. Overall, 90.1% of women were considered for RT within 1 year of diagnosis. Patients at age extremes were less likely to be considered as compared to women 50-69: th...
Computerised drug alerts are expected to reduce patients' risk of adverse drug events. However, physicians over-ride most drug alerts, because they believe that the benefit exceeds the risk. The purpose of this study was to determine the drug alert, patient and physician characteristics associated with the: (1) occurrence of psychotropic drug alerts for elderly patients and the (2) response to these alerts by their primary care physicians. Primary care, Quebec, Canada. Prospective cohort study. Sixty-one physicians using an electronic prescribing and drug alert decision-support system in their practice, and 3413 elderly patients using psychotropic drugs. PRIMARY AND SECONDARY MEASURES: Psychotropic drug class, alert severity, patient risk for fall injuries and physician experience, practice volume and computer use were evaluated in relationship to the likelihood of having: (1) a psychotropic drug alert, (2) the prescription revised in response to an alert. Cluster-adjusted alternating logistic regression was used to assess multilevel predictors of alert occurrence and response. In total 13 080 psychotropic drug alerts were generated in 8931 visits. Alerts were more likely to be generated for male patients at higher risk of fall-related injury and for physicians who established the highest alert threshold. In 9.9% of alerts seen, the prescription was revised. The highest revision rate was for antipsychotic alerts (22.6%). Physicians were more likely to revise prescriptions for severe alerts (OR 2.03; 95%CI 1.39 to 2.98), if patients had cognitive impairmen