Stuart R Lipsitz

Massachusetts General Hospital, Boston, Massachusetts, United States

Are you Stuart R Lipsitz?

Claim your profile

Publications (499)2663.58 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: For complex surveys with a binary outcome, logistic regression is widely used to model the outcome as a function of covariates. Complex survey sampling designs are typically stratified cluster samples, but consistent and asymptotically unbiased estimates of the logistic regression parameters can be obtained using weighted estimating equations (WEEs) under the naive assumption that subjects within a cluster are independent. Despite the relatively large samples typical of many complex surveys, with rare outcomes, many interaction terms, or analysis of subgroups, the logistic regression parameters estimates from WEE can be markedly biased, just as with independent samples. In this paper, we propose bias-corrected WEEs for complex survey data. The proposed method is motivated by a study of postoperative complications in laparoscopic cystectomy, using data from the 2009 United States' Nationwide Inpatient Sample complex survey of hospitals. © The Author(s) 2015.
    Statistical Methods in Medical Research 08/2015; DOI:10.1177/0962280215596550 · 2.96 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Medical organizations have increased interest in identifying and improving behaviors that threaten team performance and patient safety. Three hundred and sixty degree evaluations of surgeons were performed at 8 academically affiliated hospitals with a common Code of Excellence. We evaluate participant perceptions and make recommendations for future use. Three hundred and eighty-five surgeons in a variety of specialties underwent 360-degree evaluations, with a median of 29 reviewers each (interquartile range 23 to 36). Beginning 6 months after evaluation, surgeons, department heads, and reviewers completed follow-up surveys evaluating accuracy of feedback, willingness to participate in repeat evaluations, and behavior change. Survey response rate was 31% for surgeons (118 of 385), 59% for department heads (10 of 17), and 36% for reviewers (1,042 of 2,928). Eighty-seven percent of surgeons (95% CI, 75%-94%) agreed that reviewers provided accurate feedback. Similarly, 80% of department heads believed the feedback accurately reflected performance of surgeons within their department. Sixty percent of surgeon respondents (95% CI, 49%-75%) reported making changes to their practice based on feedback received. Seventy percent of reviewers (95% CI, 69%-74%) believed the evaluation process was valuable, with 82% (95% CI, 79%-84%) willing to participate in future 360-degree reviews. Thirty-two percent of reviewers (95% CI, 29%-35%) reported perceiving behavior change in surgeons. Three hundred and sixty degree evaluations can provide a practical, systematic, and subjectively accurate assessment of surgeon performance without undue reviewer burden. The process was found to result in beneficial behavior change, according to surgeons and their coworkers. Copyright © 2015 American College of Surgeons. Published by Elsevier Inc. All rights reserved.
    Journal of the American College of Surgeons 07/2015; DOI:10.1016/j.jamcollsurg.2015.06.017 · 4.45 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Accountable care organizations are designed to improve value by decreasing costs and maintaining quality. Strategies to maximize value are needed for high-risk surgery. We wanted to understand whether certain patient groups were differentially associated with better outcomes at high-volume hospitals in terms of quality and cost. In all, 37,746 patients underwent elective major lung resection in 1,273 hospitals in the Nationwide Inpatient Sample from 2007 to 2011. Patients were stratified by hospital volume quartile and substratified by preoperative mortality risk, age, and chronic obstructive pulmonary disease status. Mortality was evaluated using clustered multivariable hierarchical logistic regression controlling for patient comorbidity, demographics, and procedure. Adjusted cost was evaluated using generalized linear models fit to a gamma distribution. Patients were grouped into volume quartiles based on cases per year (less than 21, 21 to 40, 40 to 78, and more than 78). Patient characteristics and procedure mix differed across quartiles. Overall, mortality decreased across volume quartiles (lowest 1.9% versus highest 1.1%, p < 0.0001). Patients aged more than 80 years were associated with greater absolute and relative mortality rates than patients less than 60 years old in highest volume versus lowest volume hospitals (age more than 80 years, 4.2% versus 1.3%, p < 0.0001, odds ratio 3.31, 95% confidence interval: 1.89 to 5.80; age less than 60 years, 1.0% versus 0.8%, p = 0.19, odds ratio 1.38, 95% confidence interval: 0.74 to 2.56). Patients with high preoperative risk (more than 75th percentile) were also associated with lower absolute mortality in high-volume hospitals. Adjusted costs were not significantly different across quartiles or patient strata. Older patients show a significantly stronger volume-outcome relationship than patients less than 60 years of age. Costs were equivalent across volume quartile and patient strata. Selective patient referral may be a strategy to improve outcomes for elderly patients undergoing lung resection. Copyright © 2015 The Society of Thoracic Surgeons. Published by Elsevier Inc. All rights reserved.
    The Annals of thoracic surgery 06/2015; DOI:10.1016/j.athoracsur.2015.03.076 · 3.65 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We previously identified a range of 4344-5028 annual operations per 100,000 people to be related to desirable health outcomes. From this and other evidence, the Lancet Commission on Global Surgery recommends a minimum rate of 5000 operations per 100,000 people. We evaluate rates of growth and estimate the time it will take to reach this minimum surgical rate threshold. We aggregated country-level surgical rate estimates from 2004 to 2012 into the twenty-one Global Burden of Disease (GBD) regions. We calculated mean rates of surgery proportional to population size for each year and assessed the rate of growth over time. We then extrapolated the time it will take each region to reach a surgical rate of 5000 operations per 100,000 population based on linear rates of change. All but two regions experienced growth in their surgical rates during the past 8 years. Fourteen regions did not meet the recommended threshold in 2012. If surgical capacity continues to grow at current rates, seven regions will not meet the threshold by 2035. Eastern Sub-Saharan Africa will not reach the recommended threshold until 2124. The rates of growth in surgical service delivery are exceedingly variable. At current rates of surgical and population growth, 6.2 billion people (73 % of the world's population) will be living in countries below the minimum recommended rate of surgical care in 2035. A strategy for strengthening surgical capacity is essential if these targets are to be met in a timely fashion as part of the integrated health system development.
    World Journal of Surgery 06/2015; DOI:10.1007/s00268-015-3113-6 · 2.35 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Postpartum hemorrhage is a leading cause of maternal death worldwide. Rapid provision of uterotonics after childbirth is recommended to reduce the incidence and severity of postpartum hemorrhage. Data obtained through direct observation of childbirth practices, collected in a study of the World Health Organization's Safe Childbirth Checklist in Karnataka, India, were used to measure if oxytocin prepared for administration and available at the bedside before birth was associated with decreased time to administration after birth. This was an observational study of provider behavior: data were obtained during a baseline assessment of health worker practices prior to introduction of the Safe Childbirth Checklist, representing behavior in the absence of any intervention. Analysis was based on 330 vaginal deliveries receiving oxytocin at any point postpartum. Oxytocin was prepared and available at bedside for approximately 39% of deliveries. We found that advance preparation and bedside availability of oxytocin was associated with increased likelihood of oxytocin administration within 1 minute after delivery (adjusted risk ratio = 4.89, 95% CI = 2.61, 9.16), as well as with decreased overall time to oxytocin administration after delivery (2.9 minutes sooner in adjusted models, 95% CI = -5.0, -0.9). Efforts to reduce postpartum hemorrhage should include recommendations and interventions to ensure advance preparation and bedside availability of oxytocin to facilitate prompt administration of the medicine after birth. © Moucheraud et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly cited. To view a copy of the license, visit http://creativecommons.org/licenses/by/3.0/
    Global Health: Science and Practice 06/2015; 3(2):300-4. DOI:10.9745/GHSP-D-14-00239
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Face transplantation is an increasingly feasible option for patients with severe disfigurement. Donors and recipients are currently matched based on immune compatibility, skin characteristics, age, and gender. Aesthetic outcomes of the match are not always optimal and not possible to study in actual cases due to ethical and logistical challenges. We have used a reproducible and inexpensive three-dimensional virtual face transplantation (VFT) model to study this issue. Sixty-one VFTs were performed using reconstructed high-resolution computed tomography angiographs of male and female subjects aged 20-69 years. Twenty independent reviewers evaluated the level of disfigurement of the posttransplant models. Absolute differences in 9 soft-tissue measurements and 16 bony cephalometric measurements from each of the VFT donor and recipient pretransplant model pairs were correlated to the reviewers' evaluation of disfigurement after VFT through a multivariate logistic regression model. Five soft-tissue measurements and 3 bony measurements were predictive of the rating of disfigurement after VFT (odds ratio; 95% confidence interval): trichion-to-nasion facial height (1.106; 1.066-1.148), endocanthal width (1.096; 1.051-1.142), exocanthal width (1.067; 1.036-1.099), mouth/chelion width (1.064; 1.019-1.110), subnasale-to-menton facial height (1.029; 1.003-1.056), inner orbit width (1.039; 1.009-1.069), palatal plane/occlusal plane angle (1.148; 1.047-1.258), and sella-nasion/mandibular plane angle (1.079; 1.013-1.150). This study provides early evidence for the importance of soft-tissue and bony measurements in planning of facial transplantation. With future improvements to immunosuppressive regimens and increased donor availability, these measurements may be used as an additional criterion to optimize posttransplant outcomes.
    05/2015; 3(5):1. DOI:10.1097/GOX.0000000000000343
  • [Show abstract] [Hide abstract]
    ABSTRACT: To assess whether hospital rates of secondary complications could serve as a performance benchmark and examine associations with mortality. Failure to rescue (death after postoperative complication) is a challenging target for quality improvement. Secondary complications (complications after a first or "index" complication) are intermediate outcomes in the rescue process that may provide specific improvement targets and give us insight into how rescue fails. We used American College of Surgeons' National Surgical Quality Improvement Program data (2008-2012) to define hospital rates of secondary complications after 5 common index complications: pneumonia, surgical site infection (SSI), urinary tract infection, transfusion/bleed events, and acute myocardial infarction (MI). Hospitals were divided into quintiles on the basis of risk- and reliability-adjusted rates of secondary complications, and these rates were compared along with mortality. A total of 524,860 patients were identified undergoing one of the 62 elective, inpatient operations. After index pneumonia, secondary complication rates varied from 57.99% in the highest quintile to 22.93% in the lowest [adjusted odds ratio (OR), 4.64; confidence interval (CI), 3.95-5.45). Wide variation was seen after index SSI (58.98% vs 14.81%; OR, 8.53; CI, 7.41-9.83), urinary tract infection (38.41% vs 8.60%; OR, 7.81; CI, 6.48-9.40), transfusion/bleeding events (27.14% vs 12.88%; OR, 2.54; CI, 2.31-2.81), and acute MI (64.45% vs 23.86%, OR, 6.87; CI, 5.20-9.07). Hospitals in the highest quintile had significantly greater mortality after index pneumonia (10.41% vs 6.20%; OR, 2.17; CI, 1.6-2.94), index MI (18.25% vs 9.65%; OR, 2.67; CI, 1.80-3.94), and index SSI (2.75% vs 0.82%; OR, 3.93; CI, 2.26-6.81). Hospital-level rates of secondary complications (failure to arrest complications) vary widely, are associated with mortality, and may be useful for quality improvement and benchmarking.
    Annals of surgery 04/2015; DOI:10.1097/SLA.0000000000001227 · 7.19 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Reducing maternal and neonatal deaths are important global health priorities. We have previously shown that up to a country-level caesarean delivery rate (CDRs) of roughly 19·0%, cesarean delivery rates and maternal mortality ratio (MMR) and neonatal mortality rate (NMR) were inversely correlated. We investigated the absolute reductions in maternal and neonatal deaths if countries with low CDR increased their rates to a range of greater than 7·2% but less than or equal to 19·1%. We calculated maternal and neonatal deaths in 2013 and 2012, respectively, for countries with CDR 7·2% or less (N=45) with available data from the World Bank Development Indicators. We modelled the expected reduction in deaths in these countries if they had the 25th and 75th MMR and NMR percentiles observed for countries (N=48) with CDRs ranging from greater than 7·2% but less than or equal to 19·1%. This model assumes that if countries with low CDRs increased their rates of caesarean delivery to greater than 7·2% but less than or equal to 19·1%, they would achieve levels of MMR and NMR observed in countries with those CDRs. We estimate 176 078 (95% CI 163 258-188 898) maternal and 1 117 257 (95% CI 1 033 611-1 200 902) neonatal deaths occurred in 45 countries with low CDRs in 2013 and 2012, respectively. If these countries had the 25th and 75th MMR and NMR percentiles (MMR, IQR 36-190; NMR, 9-24) observed in countries (N=48) with a CDR ranging from greater than 7·2% but less than or equal to 19·1%, there would be a potential reduction of 109 762-163 513 and 279 584-803 129 maternal and neonatal deaths, respectively. Increasing caesarean delivery in countries with low CDRs could avert as many as 163 513 maternal deaths and 803 129 neonatal deaths annually. These findings assume that as health systems develop the capacity to deliver surgical care, there is a concurrent improvement in the quality of care and in the ability to rescue women and neonates who would otherwise die. Improving access to safe caesarean delivery should be a central focus in surgical care globally. None. Copyright © 2015 Elsevier Ltd. All rights reserved.
    The Lancet 04/2015; 385:S33. DOI:10.1016/S0140-6736(15)60828-5 · 45.22 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Recent work has indicated an increase in surgical services, especially in resource poor settings. However, the rate of growth is poorly understood and likely insufficient to meet public health needs. We previously identified a range of 4344 to 5028 operations per 100 000 population annually to be related to desirable health outcomes. From this and other evidence, the Lancet Commission on Global Surgery recommends a minimum rate of 5000 operations per 100 000 population. We evaluate rates of growth in surgery and estimate the time it will take to reach this minimum surgical rate threshold.
    The Lancet 04/2015; 385:S14. DOI:10.1016/S0140-6736(15)60809-1 · 45.22 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: It was previously estimated that 234·2 million operations were performed worldwide in 2004. The association between surgical rates and population health outcomes is not clear. We re-estimated global surgical volume to track changes over time and assess rates associated with healthy populations. We gathered demographic, health, and economic data for 194 WHO member states. Surgical volumes were obtained from published studies and other reports from 2005 onwards. We estimated rates of surgery for all countries without available data based on health expenditure in 2012 and assessed the proportion of surgery comprised by caesarean delivery. The rate of surgery was plotted against life expectancy to describe the association between surgical care and this health indicator. We identified 66 countries reporting surgical data between 2005 and 2013. We estimate that 312·9 million operations (95% CI 266·2-359·5) took place in 2012-a 33·6% increase over 8 years; the largest proportional increase occurred in countries spending US$400 or less per capita on health care. Caesarean delivery comprised 29·8% (5·8 million operations) of the total surgical volume in poor health expenditure countries compared with 10·8% (7·8 million operations) in low health expenditure countries and 2·7% (5·1 million operations) in high health expenditure countries. We noted a correlation between increased life expectancy and increased surgical rates up to 1533 operations per 100 000 people, with significant but less dramatic improvement above this rate. Surgical volume is large and continues to grow in all economic environments. A single procedure-caesarean delivery-comprised almost a third of surgical volume in the most resource-limited settings. Surgical care is an essential part of health care and is associated with increased life expectancy, yet many low-income countries fail to achieve basic levels of service. Improvements in capacity and delivery of surgical services must be a major component of health system strengthening. None. Copyright © 2015 Elsevier Ltd. All rights reserved.
    The Lancet 04/2015; 385:S11. DOI:10.1016/S0140-6736(15)60806-6 · 45.22 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The global volume of surgery is estimated at 312.9 million operations annually, but rates of surgery vary dramatically. Identifying surgical rates associated with improved health outcomes would be useful for benchmarking and targeted health system strengthening. We identified rates of surgery associated with a life expectancy (LE) of 74-75 years, a maternal mortality ratio (MMR) of less than or equal to 100 per 100,000 live births, and the estimated need for surgery in the seven global burden of disease (GBD) super-regions based on the prevalence of surgical conditions. We compared our findings to surgical rates from Chile, China, Costa Rica, and Cuba ("4C"), countries with moderate resources but high health outcomes. The median surgical rates associated with LE of 74-75 years (N = 17) and MMR below 100 (N = 109) are 4392 (IQR 2897-4873) and 5028 (IQR 4139-6778) operations per 100,000 people annually, respectively. The mean surgical rate estimated for the seven super-regions was 4723 (95 % CI 3967-5478) operations per 100,000 people annually. The "4C" countries had a mean surgical rate of 4344 (95 % CI 2620-6068) operations per 100,000 people annually. Thirteen of the twenty-one GBD regions, accounting for 78 % of the world's population, do not achieve rates of surgery at the lowest end of this range. We identified a narrow range of surgical rates associated with important health indicators. This target range can be used for benchmarking of surgical services, and as part of a policy aimed at strengthening health care systems and surgical capacity.
    The Lancet 04/2015; 385:S12. DOI:10.1016/S0140-6736(15)60807-8 · 45.22 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Trauma team training provides instruction on crisis management through debriefing and discussion of teamwork and leadership skills during simulated trauma scenarios. The effects of team leader's nontechnical skills (NTSs) on technical performance have not been thoroughly studied. We hypothesized that team's and team leader's NTSs correlate with technical performance of clinical tasks. Retrospective cohort study. Brigham and Women's Hospital, STRATUS Center for Surgical Simulation PARTICIPANTS: A total of 20 teams composed of surgical residents, emergency medicine residents, emergency department nurses, and emergency services assistants underwent 2 separate, high-fidelity, simulated trauma scenarios. Each trauma scenario was recorded on video for analysis and divided into 4 consecutive sections. For each section, 2 raters used the Non-Technical Skills for Surgeons framework to assess NTSs of the team. To evaluate the entire team's NTS, 2 additional raters used the Modified Non-Technical Skills Scale for Trauma system. Clinical performance measures including adherence to guidelines and time to perform critical tasks were measured independently. NTSs performance by both teams and team leaders in all NTS categories decreased from the beginning to the end of the scenario (all p < 0.05). There was significant correlation between team's and team leader's cognitive skills and critical task performance, with correlation coefficients between 0.351 and 0.478 (p < 0.05). The NTS performance of the team leader highly correlated with that of the entire team, with correlation coefficients between 0.602 and 0.785 (p < 0.001). The NTSs of trauma teams and team leaders deteriorate as clinical scenarios progress, and the performance of team leaders and teams is highly correlated. Cognitive NTS scores correlate with critical task performance. Increased attention to NTSs during trauma team training may lead to sustained performance throughout trauma scenarios. Decision making and situation awareness skills are critical for both team leaders and teams and should be specifically addressed to improve performance. Copyright © 2015 Association of Program Directors in Surgery. Published by Elsevier Inc. All rights reserved.
    Journal of Surgical Education 03/2015; DOI:10.1016/j.jsurg.2015.01.020 · 1.39 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Little is known about the sustainability and long-term effect of surgical safety checklists when implemented in resource-limited settings. A previous study demonstrated the marked, short-term effect of a structured hospital-wide implementation of a surgical safety checklist in Moldova, a lower-middle-income country, as have studies in other low-resource settings. To assess the long-term reduction in perioperative harm following the introduction of a checklist-based surgical quality improvement program in a resource-limited setting and to understand the long-term effects of such programs. Twenty months after the initial implementation of a surgical safety checklist and the provision of pulse oximetry at a referral hospital in Moldova, a lower-middle-income, resource-limited country in Eastern Europe, we conducted a prospective study of perioperative care and outcomes of 637 consecutive patients undergoing noncardiac surgery (the long-term follow-up group), and we compared the findings with those from 2106 patients who underwent surgery shortly after implementation (the short-term follow-up group). Preintervention data were collected from March to July 2010. Data collection during the short-term follow-up period was performed from October 2010 to January 2011, beginning 1 month after the implementation of the launch period. Data collection during the long-term follow-up period took place from May 25 to July 6, 2012, beginning 20 months after the initial intervention. The primary end points of interest were surgical morbidity (ie, the complication rate), adherence to safety process measures, and frequency of hypoxemia. Between the short- and long-term follow-up groups, the complication rate decreased 30.7% (P = .03). Surgical site infections decreased 40.4% (P = .05). The mean (SD) rate of completion of the checklist items increased from 88% (14%) in the short-term follow-up group to 92% (11%) in the long-term follow-up group (P < .001). The rate of hypoxemic events continued to decrease (from 8.1 events per 100 hours of oximetry for the short-term follow-up group to 6.8 events per 100 hours of oximetry for the long-term follow-up group; P = .10). Sustained use of the checklist was observed with continued improvements in process measures and reductions in 30-day surgical complications almost 2 years after a structured implementation effort that demonstrated marked, short-term reductions in harm. The sustained effect occurred despite the absence of continued oversight by the research team, indicating the important role that local leadership and local champions play in the success of quality improvement initiatives, especially in resource-limited settings.
    JAMA SURGERY 03/2015; 150(5). DOI:10.1001/jamasurg.2014.3848 · 4.30 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Failure-to-rescue (FTR or death after postoperative complication) is thought to explain surgical mortality excesses across hospitals, and FTR is an emerging performance measure and target for quality improvement. We compared the FTR population to preoperatively identifiable subpopulations for their potential to close the mortality gap between lowest- and highest-mortality hospitals. Patients undergoing small bowel resection, pancreatectomy, colorectal resection, open abdominal aortic aneurysm repair, lower extremity arterial bypass, and nephrectomy were identified in the 2007 to 2011 Nationwide Inpatient Sample. Lowest- and highest-mortality hospitals were defined using risk- and reliability-adjusted mortality quintiles. Five target subpopulations were established a priori: the FTR population, predicted high-mortality risk (predicted highest-risk quintile), emergency surgery, elderly (>75 years old), and diabetic patients. Across the lowest mortality quintile (n = 282 hospitals, 56,893 patients) and highest-mortality quintile (282 hospitals, 45,784 patients), respectively, the size of target subpopulations varied only for the FTR population (20.2% vs 22.4%, p = 0.002) but not for other subpopulations. Variation in mortality rates across lowest- and highest-mortality hospitals was greatest for the high-mortality risk (7.5% vs 20.2%, p < 0.0001) and FTR subpopulations (7.8% vs 18.9%, p < 0.0001). The FTR and high-risk populations had comparable sensitivity (81% and 75%) and positive predictive value (19% and 20%, respectively) for mortality. In Monte Carlo simulations, the mortality gap between the lowest- and highest-mortality hospitals was reduced by nearly 75% when targeting the FTR population or the high-risk population, 78% for the emergency surgery population, but less for elderly (51%) and diabetic (17%) populations. Preoperatively identifiable patients with high estimated mortality risk may be preferable to the FTR population as a target for surgical mortality reduction. Copyright © 2015 American College of Surgeons. Published by Elsevier Inc. All rights reserved.
    Journal of the American College of Surgeons 03/2015; 220(6). DOI:10.1016/j.jamcollsurg.2015.02.036 · 4.45 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Unlike majority of current statistical models and methods focusing on mean response for highly skewed longitudinal data, we present a novel model for such data accommodating a partially linear median regression function, a skewed error distribution and within subject association structures. We provide theoretical justifications for our methods including asymptotic properties of the posterior and associated semiparametric Bayesian estimators. We also provide simulation studies to investigate the finite sample properties of our methods. Several advantages of our method compared with existing methods are demonstrated via analysis of a cardiotoxicity study of children of HIV-infected mothers. © The Author 2015. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
    Biostatistics 03/2015; 16(3). DOI:10.1093/biostatistics/kxv005 · 2.24 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The test of independence of row and column variables in a (J×K) contingency table is a widely used statistical test in many areas of application. For complex survey samples, use of the standard Pearson chi-squared test is inappropriate due to correlation among units within the same cluster. Rao and Scott (1981, Journal of the American Statistical Association 76, 221-230) proposed an approach in which the standard Pearson chi-squared statistic is multiplied by a design effect to adjust for the complex survey design. Unfortunately, this test fails to exist when one of the observed cell counts equals zero. Even with the large samples typical of many complex surveys, zero cell counts can occur for rare events, small domains, or contingency tables with a large number of cells. Here, we propose Wald and score test statistics for independence based on weighted least squares estimating equations. In contrast to the Rao-Scott test statistic, the proposed Wald and score test statistics always exist. In simulations, the score test is found to perform best with respect to type I error. The proposed method is motivated by, and applied to, post surgical complications data from the United States' Nationwide Inpatient Sample (NIS) complex survey of hospitals in 2008. © 2015, The International Biometric Society.
    Biometrics 03/2015; DOI:10.1111/biom.12297 · 1.52 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: To evaluate the learning curves of three high-volume procedures, from distinct surgical specialties. Tertiary care academic hospital. A prospectively collected database comprising all medical records of patients undergoing isolated coronary artery bypass grafting (CABG), total knee replacement (TKR) and bilateral reduction mammoplasty (BRM) at the Brigham and Women's Hospital, USA, 1996-2010. Multivariate generalised estimating equation (GEE) regression models were used to adjust for patient risk and clustering of procedures by surgeon. Operative efficiency. A total of 1052 BRMs, 3254 CABGs and 3325 TKRs performed by 30 surgeons were analysed. Median number of procedures per surgeon was 61 (range 11-502), 290 (52-973) and 99 (10-1871) for BRM, CABG and TKR, respectively. Mean operative times were 134.4 (SD 34.5), 180.9 (62.3) and 101.9 (30.3) minutes, respectively. For each procedure, attending surgeon experience was associated with significant reductions in operative time (p<0.05). After 15 years of experience, BRM operative time decreased by 69.8 min (38.3%), CABG operative time decreased by 17.5 min (7.8%) and TKR operative time decreased by 94.4 min (48.4%). Common trends in surgical learning exist. Dependent on the procedure, experience can serve as a powerful driver of improvement or have clinically insignificant impacts on operative time. Published by the BMJ Publishing Group Limited. For permission to use (where not already granted under a licence) please go to http://group.bmj.com/group/rights-licensing/permissions.
    BMJ Open 03/2015; 5(3):e006679. DOI:10.1136/bmjopen-2014-006679 · 2.06 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The purpose of this study is to assess which personal characteristics and external factors are important contributors to eventual success. The authors distributed a survey to all members of the American Association of Plastic Surgeons and asked responders to rate the importance of 10 preselected qualities in contributing to their personal success. Survey outcomes were analyzed across different demographic groups. Of the 580 American Association of Plastic Surgeons members who were surveyed, 295 returned completed surveys. Overall analysis indicates that hard work, compassion, and manual dexterity are the 3 most important attributes. Many significant differences are observed across demographic groups, indicating potential biases among the survey responders. Notably, we find that male surgeons attribute mentorship to success much more so than female surgeons (Column Trend Exact [CTE], P = 0.021), whereas female surgeons are more likely to attribute their success to hard work (CTE, P = 0.023). Similarly, those who have been program directors credit their success to mentoring more so than nonprogram directors (CTE, P < 0.00001). The authors also found that senior surgeons, as measured by years in practice, place greater emphasis on mentoring and career opportunities than younger surgeons (Mantel-Haenszel Trend, P = 0.003 and 0.0009, respectively). It is also interesting to note that individual talent qualities tend to be favored by more senior surgeons and those with more distant ties to academia. The authors believe that recognizing the relative importance of such factors, and their associated biases, is essential for the process of selecting and developing future successful plastic surgeons.
    03/2015; 3(3):e327. DOI:10.1097/GOX.0000000000000229
  • [Show abstract] [Hide abstract]
    ABSTRACT: Bernoulli (or binomial) regression using a generalized linear model with a log link function, where the exponentiated regression parameters have interpretation as relative risks, is often more appropriate than logistic regression for prospective studies with common outcomes. In particular, many researchers regard relative risks to be more intuitively interpretable than odds ratios. However, for the log link, when the outcome is very prevalent, the likelihood may not have a unique maximum. To circumvent this problem, a 'COPY method' has been proposed, which is equivalent to creating for each subject an additional observation with the same covariates except the response variable has the outcome values interchanged (1's changed to 0's and 0's changed to 1's). The original response is given weight close to 1, while the new observation is given a positive weight close to 0; this approach always leads to convergence of the maximum likelihood algorithm, except for problems with convergence due to multicollinearity among covariates. Even though this method produces a unique maximum, when the outcome is very prevalent, and/or the sample size is relatively small, the COPY method can yield biased estimates. Here, we propose using the jackknife as a bias-reduction approach for the COPY method. The proposed method is motivated by a study of patients undergoing colorectal cancer surgery. Copyright © 2014 John Wiley & Sons, Ltd.
    Statistics in Medicine 02/2015; 34(3). DOI:10.1002/sim.6348 · 2.04 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To identify the timing and relative frequency of common postoperative complications in a contemporary, diverse surgical population and develop a mnemonic for teaching and clinical decision support. We enrolled a cohort of general and vascular surgical patients undergoing elective, inpatient surgery in the American College of Surgeons National Surgical Quality Improvement Program database between 2005 and 2011. Index complications were noted by postoperative day (POD). Timing and incidence were compared within each day. Among 614,525 patients, 51,173 (9.88%) experienced the following index complications over 30 days: pneumonia (n = 5947), urinary tract infection (n = 9459), superficial surgical site infection (sSSI) (n = 20,460), deep/organ space surgical site infection (dSSI) infection (n = 11,847), venous thromboembolism (n = 4478), kidney injury (n = 2620), and myocardial infarction (n = 1813). Median time to complication differed significantly for index complications (p < 0.0001). On POD 0, the most common complication was myocardial infarction (incidence 4.26/10,000 patient days; 95% CI: 3.75-4.78). On POD 1 and 2, pneumonia was the most common complication, with peak incidence on POD 2 (20.36; 95% CI: 19.22-21.51). On POD 3, pneumonia (16.3; 95% CI: 15.27-17.33) and urinary tract infection (15.5; 95% CI: 14.49-16.51) were significantly more common than other complications. On POD 4, the most common complication was sSSI (16.24; 95% CI: 15.20-17.28). From POD 5 to POD 30, sSSI and dSSI were the 2 most common complications. Risk of venous thromboembolism declined only slightly through POD 30. We propose a mnemonic for postoperative complication timing and frequency, independent of fever, as follows: Waves (myocardial infarction), Wind (pneumonia), Water (urinary tract), Wound (sSSI and dSSI), and Walking (venous thromboembolism) in the order of likelihood. Copyright © 2014 Association of Program Directors in Surgery. Published by Elsevier Inc. All rights reserved.
    Journal of Surgical Education 12/2014; 72(3). DOI:10.1016/j.jsurg.2014.11.004 · 1.39 Impact Factor

Publication Stats

15k Citations
2,663.58 Total Impact Points

Institutions

  • 2015
    • Massachusetts General Hospital
      • Department of Surgery
      Boston, Massachusetts, United States
  • 2005–2015
    • Brigham and Women's Hospital
      • • Department of Medicine
      • • Center for Surgery and Public Health
      Boston, Massachusetts, United States
    • Creighton University
      Omaha, Nebraska, United States
    • University of South Florida St. Petersburg
      St. Petersburg, Florida, United States
    • New York University
      New York City, New York, United States
  • 2014
    • The University of Chicago Medical Center
      • Section of Urology
      Chicago, Illinois, United States
  • 1992–2014
    • Harvard University
      • Department of Health Policy and Management
      Cambridge, Massachusetts, United States
  • 1990–2014
    • Harvard Medical School
      • • Department of Surgery
      • • Department of Medicine
      Boston, Massachusetts, United States
  • 2012
    • St. Joseph Hospital, Orange
      Orange, California, United States
  • 2010–2012
    • Beverly Hospital, Boston MA
      BVY, Massachusetts, United States
    • University of Miami Miller School of Medicine
      • Department of Pediatrics
      Miami, Florida, United States
  • 1990–2012
    • Dana-Farber Cancer Institute
      • • Lank Center for Genitourinary Oncology
      • • Department of Biostatistics and Computational Biology
      Boston, Massachusetts, United States
  • 2011
    • Emory University
      • Goizueta Business School
      Atlanta, GA, United States
    • Georgetown University
      • Department of Urology
      Washington, Washington, D.C., United States
    • Cedars-Sinai Medical Center
      • Cedars Sinai Medical Center
      Los Ángeles, California, United States
  • 1999–2010
    • Medical University of South Carolina
      • • Department of Medicine
      • • Division of Biostatistics and Epidemiology
      Charleston, South Carolina, United States
  • 2009
    • CUNY Graduate Center
      New York, New York, United States
  • 2008–2009
    • Partners HealthCare
      Boston, Massachusetts, United States
  • 2007
    • Texas A&M University
      • Department of Statistics
      College Station, Texas, United States
  • 1999–2005
    • University of Rochester
      • Division of Pediatric Cardiology
      Rochester, New York, United States
  • 2003
    • University Center Rochester
      Рочестер, Minnesota, United States
  • 2002
    • University of North Carolina at Chapel Hill
      • Department of Biostatistics
      North Carolina, United States
  • 1998
    • Fox Chase Cancer Center
      Filadelfia, Pennsylvania, United States
  • 1991–1998
    • Massachusetts Department of Public Health
      Boston, Massachusetts, United States
  • 1997
    • Montefiore Medical Center
      New York City, New York, United States
  • 1995
    • Albert Einstein College of Medicine
      • Department of Medicine
      New York City, NY, United States