Impact of a physician-oriented intervention on follow-up in colorectal cancer screening
ABSTRACT Complete diagnostic evaluation or CDE (i.e., colonoscopy or combined flexible sigmoidoscopy plus barium enema X-ray) is often not performed for persons with an abnormal screening fecal occult blood test (FOBT+) result.
This study evaluated the impact of a reminder-feedback and educational outreach intervention on primary care practice CDE recommendation and performance rates. Four hundred seventy primary care physicians (PCPs) in 318 practices participated in the study. Patients were mailed an FOBT kit annually as part of a screening program. Practices were randomly assigned to a Control Group (N = 198) or an Intervention Group (N = 120). During an 18-month pre-randomization period and a 9-month post-randomization period, 2992 screening FOBT+ patients were identified. Intervention practices received the screening program and the intervention. Control practices received only the screening program. Study outcomes were baseline-adjusted CDE recommendation and performance rates.
At baseline, about two-thirds of FOBT+ patients received a CDE recommendation, and about half had a CDE performed. At endpoint, CDE recommendation and performance rates were both significantly higher for the Intervention as compared to the Control practices (OR = 2.28; 95% CI: 1.37, 3.78, and OR = 1.63; 95% CI: 1.06, 2.50, respectively).
The reminder-feedback plus educational outreach intervention significantly increased CDE recommendation and performance.
[Show abstract] [Hide abstract]
ABSTRACT: Background. This study assessed the contribution of organizational structures and processes identified from facility surveys to follow-up for positive Fecal Occult Blood Tests (FOBT+). Methods. We identified 74,104 patients with FOBT+ results from 98 Veterans Health Administration (VHA) facilities between 8/16/09-3/20/11 and followed them until 9/30/11 for completion of colonoscopy. We identified patient characteristics from VHA administrative records, and organizational factors from facility surveys completed by Primary Care and Gastroenterology Chiefs. We estimated predictors of colonoscopy completion within 60 days and 6 months using hierarchical logistic regression models. Results. 30% of patients with FOBT+ results received colonoscopy within 60 days and 49% within 6 months. Having Gastroenterology or Laboratory staff notify Gastroenterology providers directly about FOBT+ cases was a significant predictor of 60-day (odds ratio (OR)=1.85, p=0.01) and 6-month follow-up (OR 1.25, p=0.008). Additional predictors of 60-day follow-up included adequacy of colonoscopy appointment availability (OR 1.43, p=0.01) and frequent individual feedback to Primary Care providers about FOBT+ referral timeliness (OR 1.79, p=0.04). Additional predictors of 6-month follow-up included using guideline-concordant surveillance intervals for low-risk adenomas (OR 1.57, p=0.01) and using group appointments and combined verbal-written methods for colonoscopy preparation instruction (OR 1.48, p=0.0001). Conclusion. Directly notifying Gastroenterology providers about FOBT+ results, employing guideline-concordant adenoma surveillance intervals, and using colonoscopy preparations instruction methods that provide both verbal and written information may increase overall follow-up rates. Enhancing follow-up within 60-days may require increased colonoscopy capacity and feedback to Primary Care providers. Impact. These findings may inform organizational-level interventions to improve FOBT+ follow-up. Copyright © 2014, American Association for Cancer Research.Cancer Epidemiology Biomarkers & Prevention 12/2014; 24(2). DOI:10.1158/1055-9965.EPI-14-1170 · 4.32 Impact Factor
[Show abstract] [Hide abstract]
ABSTRACT: Timely follow-up of fecal occult blood screening with colonoscopy is essential for achieving colorectal cancer mortality reduction. This study evaluates the effectiveness of two ongoing interventions designed to improve colonoscopy uptake after a positive fecal occult blood test (FOBT) result within Ontario's population-wide ColonCancerCheck program. The first was a revision of mailed FOBT lab results to physicians to explicitly define a positive FOBT and to recommend colonoscopy. The second was a letter to participants informing them of the positive FOBT and urging them to seek appropriate follow-up. Prospective cohort study using Ontario's ColonCancerCheck program data sets (2008-2011), linked to provincial administrative health databases. Crude rate ratios were calculated to assess determinants of colonoscopy uptake among an Ontario-wide FOBT-positive cohort with rolling enrolment, followed from October 2008 through February 2011. Segmented time-series regression was used to assess the average additional change in colonoscopy uptake after FOBT-positive status following the introduction of two ongoing interventions among the same cohort. A notification mailed directly to FOBT-positive screening participants was observed to increase colonoscopy uptake, beyond the modest average underlying increase throughout the study period, by an average of 3% per month (multivariable-adjusted RR: 1.03, 95% CI: 1.00-1.06). However, revision of the existing FOBT result notification to physicians was observed to have no effect. Direct participant notification of a positive FOBT result improved adherence with follow-up colonoscopy in Ontario's population-wide ColonCancerCheck program. Further participant-directed interventions may be an effective means of maximizing adherence in population-wide screening.Implementation Science 12/2015; 10(1):226. DOI:10.1186/s13012-015-0226-0 · 3.47 Impact Factor
[Show abstract] [Hide abstract]
ABSTRACT: There is currently no reliable way to choose strategies that are appropriate for implementing guidelines facing different barriers. This study examined trends in guideline implementation by topic over a 10-year period to explore whether and how strategies may be suitable for addressing differing barriers. A scoping systematic review was performed. MEDLINE and EMBASE were searched from 2004 to 2013 for studies that evaluated the implementation of guidelines on arthritis, diabetes, colorectal cancer and heart failure. Data on study characteristics, reason for implementation (new guideline or quality improvement), implementation strategy used, rationale for selecting that strategy and reported impact were extracted and summarized. Interventions were mapped against a published taxonomy of guideline implementation strategies. The search resulted in 1,709 articles; 156 were retrieved and 127 were excluded largely because they did not evaluate guideline implementation, leaving 32 eligible for review (4 arthritis, 3 colorectal cancer, 21 diabetes, 4 heart failure). Six of 7 randomized trials and 8 of 25 observational studies had a low risk of bias. Most studies promoted guideline use for quality improvement (78.0%). Few studies rationalized strategy choice (18.8%). Most employed multiple approaches and strategies, most often educational meetings and print material for professionals or patients. Few studies employed organizational, financial or regulatory approaches. Strategies employed that were unique to the published taxonomy included professional (print material, tailoring guidelines, self-audit training or material) and patient strategies (education, counselling, group interaction, print material, reminders). Most studies achieved positive impact (87.5%). This did not appear to be associated with guideline topic, use of theory or barrier assessment, or number or type of implementation approaches and strategies. While few studies were eligible, limiting insight on how to choose implementation strategies that address guideline-specific barriers, this review identified other important findings. Education for professionals or patients and print material were the most commonly employed strategies for translating guidelines to practice. Mapping of strategies onto the published taxonomy identified gaps in guideline implementation that represent opportunities for future research and expanded the taxonomy.Implementation Science 04/2015; 10(1):54. DOI:10.1186/s13012-015-0247-8 · 3.47 Impact Factor