Search tips
Search criteria

Results 1-13 (13)

Clipboard (0)

Select a Filter Below

more »
Year of Publication
Document Types
1.  Trends in Diabetes Incidence Among 7 Million Insured Adults, 2006–2011 
American Journal of Epidemiology  2014;181(1):32-39.
An observational cohort analysis was conducted within the Surveillance, Prevention, and Management of Diabetes Mellitus (SUPREME-DM) DataLink, a consortium of 11 integrated health-care delivery systems with electronic health records in 10 US states. Among nearly 7 million adults aged 20 years or older, we estimated annual diabetes incidence per 1,000 persons overall and by age, sex, race/ethnicity, and body mass index. We identified 289,050 incident cases of diabetes. Age- and sex-adjusted population incidence was stable between 2006 and 2010, ranging from 10.3 per 1,000 adults (95% confidence interval (CI): 9.8, 10.7) to 11.3 per 1,000 adults (95% CI: 11.0, 11.7). Adjusted incidence was significantly higher in 2011 (11.5, 95% CI: 10.9, 12.0) than in the 2 years with the lowest incidence. A similar pattern was observed in most prespecified subgroups, but only the differences for persons who were not white were significant. In 2006, 56% of incident cases had a glycated hemoglobin (hemoglobin A1c) test as one of the pair of events identifying diabetes. By 2011, that number was 74%. In conclusion, overall diabetes incidence in this population did not significantly increase between 2006 and 2010, but increases in hemoglobin A1c testing may have contributed to rising diabetes incidence among nonwhites in 2011.
PMCID: PMC4288120  PMID: 25515167
diabetes mellitus; glycated hemoglobin; hemoglobin A1c; incidence; trends
2.  An Interactive Computer Session to Initiate Physical Activity in Sedentary Cardiac Patients: Randomized Controlled Trial 
Physical activity (PA) improves many facets of health. Despite this, the majority of American adults are insufficiently active. Adults who visit a physician complaining of chest pain and related cardiovascular symptoms are often referred for further testing. However, when this testing does not reveal an underlying disease or pathology, patients typically receive no additional standard care services. A PA intervention delivered within the clinic setting may be an effective strategy for improving the health of this population at a time when they may be motivated to take preventive action.
Our aim was to determine the effectiveness of a tailored, computer-based, interactive personal action planning session to initiate PA among a group of sedentary cardiac patients following exercise treadmill testing (ETT).
This study was part of a larger 2x2 randomized controlled trial to determine the impact of environmental and social-cognitive intervention approaches on the initiation and maintenance of weekly PA for patients post ETT. Participants who were referred to an ETT center but had a negative-test (ie, stress tests results indicated no apparent cardiac issues) were randomized to one of four treatment arms: (1) increased environmental accessibility to PA resources via the provision of a free voucher to a fitness facility in close proximity to their home or workplace (ENV), (2) a tailored social cognitive intervention (SC) using a “5 As”-based (ask, advise, assess, assist, and arrange) personal action planning tool, (3) combined intervention of both ENV and SC approaches (COMBO), or (4) a matched contact nutrition control (CON). Each intervention was delivered using a computer-based interactive session. A general linear model for repeated measures was conducted with change in PA behavior from baseline to 1-month post interactive computer session as the primary outcome.
Sedentary participants (n=452; 34.7% participation rate) without a gym membership (mean age 58.57 years; 59% female, 78% white, 12% black, 11% Hispanic) completed a baseline assessment and an interactive computer session. PA increased across the study sample (F 1,441=30.03, P<.001). However, a time by condition interaction (F 3,441=8.33, P<.001) followed by post hoc analyses indicated that SC participants exhibited a significant increase in weekly PA participation (mean 45.1, SD 10.2) compared to CON (mean -2.5, SD 10.8, P=.004) and ENV (mean 8.3, SD 8.1, P<.05). Additionally, COMBO participants exhibited a significant increase in weekly PA participation (mean 53.4, SD 8.9) compared to CON (P<.001) and ENV (P=.003) participants. There were no significant differences between ENV and CON or between SC and COMBO.
A brief, computer-based, interactive personal action planning session may be an effective tool to initiate PA within a health care setting, in particular as part of the ETT system.
Trial Registration NCT00432133, (Archived by WebCite at
PMCID: PMC4642390  PMID: 26303347
exercise, physical; treadmill test; human computer interaction; behavioral research; cardiovascular diseases; interactive media
3.  Estimating the effects of time-varying exposures in observational studies using Cox models with stabilized weights adjustment 
Assessing the safety and effectiveness of medical products with observational electronic medical record data is challenging when the treatment is time-varying. The objective of this paper is to develop a Cox model stratified by event times with stabilized weights adjustment to examine the effect of time-varying treatment in observational studies.
Time-varying stabilized weights are calculated at unique event times and are used in a Cox model stratified by event times to estimate the effect of time-varying treatment. We applied this method in examining the effect of an anti-platelet agent, clopidogrel, on events, including bleeding, myocardial infarction (MI), and death after a Drug-Eluting Stent was implanted in coronary artery. Clopidogrel use may change over time based on patients' behavior (e.g., non-adherence) and physicians' recommendations (e.g., end of duration of therapy). We also compared the results to those from a Cox model for counting processes adjusting for all covariates used in creating stabilized weights.
We demonstrate that 1) results from the stratified Cox model without stabilized weights adjustment and the Cox model for counting processes without covariate adjustment are identical in analyzing the clopidogrel data; and 2) effects of clopidogrel on bleeding, MI and death are larger in the stratified Cox model with stabilized weights adjustment compared to those from the Cox model for counting processes with covariate adjustment.
The Cox model stratified by event times with time-varying stabilized weights adjustment is useful in estimating the effect of time-varying treatments in observational studies while balancing for known confounders.
PMCID: PMC4351798  PMID: 24596337
Drug-Eluting Stent; Clopidogrel; Time-varying exposure; Cox model; Stabilized weights
4.  Use of stabilized inverse propensity scores as weights to directly estimate relative risk and its confidence intervals 
Inverse probability of treatment weighting (IPTW) has been used in observational studies to reduce selection bias. To obtain estimates of the main effects, a pseudo data set is created by weighting each subject by IPTW and analyzed with conventional regression models. Currently variance estimation requires additional work depending on type of outcomes. Our goal is to demonstrate a statistical approach to directly obtain appropriate estimates of variance of the main effects in regression models.
We carried out theoretical and simulation studies to show that the variance of the main effects estimated directly from regressions using IPTW is underestimated, and that the type I error rate is higher due to the inflated sample size in the pseudo data. The robust variance estimator using IPTW often slightly overestimates the variance of the main effects. We propose to use the stabilized weights to directly estimate both the main effect and its variance from conventional regression models.
We applied the approach to a study examining the effectiveness of serum potassium monitoring in reducing hyperkalemia-associated adverse events among 27,355 diabetic patients newly-prescribed a renin-angiotensin-aldosterone system (RAAS) inhibitor. The incidence rate ratio (with monitoring versus without monitoring) and confidence intervals were 0.46 (0.34, 0.61) using the stabilized weights compared to 0.46 (0.38, 0.55) using typical inverse probability of treatment weighting.
Our theoretical, simulation results and real data example demonstrate that the use of the stabilized weights in the pseudo data preserves the sample size of the original data, produces appropriate estimation of the variance of main effect, and maintains an appropriate type I error rate.
PMCID: PMC4351790  PMID: 19912596
Inverse probability of treatment weighting; stabilized weights; type I error rates; incidence rate ratio; confidence intervals
5.  Work productivity loss from depression: evidence from an employer survey 
National working groups identify the need for return on investment research conducted from the purchaser perspective; however, the field has not developed standardized methods for measuring the basic components of return on investment, including costing out the value of work productivity loss due to illness. Recent literature is divided on whether the most commonly used method underestimates or overestimates this loss. The goal of this manuscript is to characterize between and within variation in the cost of work productivity loss from illness estimated by the most commonly used method and its two refinements.
One senior health benefit specialist from each of 325 companies employing 100+ workers completed a cross-sectional survey describing their company size, industry and policies/practices regarding work loss which allowed the research team to derive the variables needed to estimate work productivity loss from illness using three methods. Compensation estimates were derived by multiplying lost work hours from presenteeism and absenteeism by wage/fringe. Disruption correction adjusted this estimate to account for co-worker disruption, while friction correction accounted for labor substitution. The analysis compared bootstrapped means and medians between and within these three methods.
The average company realized an annual $617 (SD = $75) per capita loss from depression by compensation methods and a $649 (SD = $78) loss by disruption correction, compared to a $316 (SD = $58) loss by friction correction (p < .0001). Agreement across estimates was 0.92 (95% CI 0.90, 0.93).
Although the methods identify similar companies with high costs from lost productivity, friction correction reduces the size of compensation estimates of productivity loss by one half. In analyzing the potential consequences of method selection for the dissemination of interventions to employers, intervention developers are encouraged to include friction methods in their estimate of the economic value of interventions designed to improve absenteeism and presenteeism. Business leaders in industries where labor substitution is common are encouraged to seek friction corrected estimates of return on investment. Health policy analysts are encouraged to target the dissemination of productivity enhancing interventions to employers with high losses rather than all employers.
Trial registration
Clinical trials registration number: NCT01013220.
PMCID: PMC4307989  PMID: 25519705
Return on investment; Work; Productivity; Depression; Health promotion
6.  Intervention impact on depression product appraisal and purchasing behavior by employers: a randomized trial 
Employers can purchase high quality depression products that provide the type, intensity and duration of depression care management shown to improve work outcomes sufficiently for many employers to achieve a return on investment. The purpose of this randomized controlled trial was to test an intervention to encourage employers to purchase a high quality depression product for their workforce.
Twenty nine organizations recruited senior health benefit professional members representing public or private employers who had not yet purchased a depression product for all 100+ workers in their company. The research team used randomization blocked by company size to assign eligible employers to: (1) a presentation encouraging employers to purchase a high quality depression product accompanied by a scientifically-derived return on investment estimate, or (2) a presentation encouraging employers to work with their most subscribed health plan to improve depression treatment quality indicators. Two hundred ninety three employers (82.3% of 356) completed baseline data immediately before learning that 140 employers had been randomized to the evidence-based (EB) depression product presentation and 153 had been randomized to the usual care (UC) depression treatment quality indicator presentation. Analysis of 250 (85.3% of 293) employers who completed web-based interviews at 12 and/or 24 months was conducted to determine presentation impact on depression product appraisal and purchasing behavior.
The intervention had no impact on depression product appraisal in 232 subjects (F = 2.36, p = .07) or depression product purchasing (chisquare = 1.82, p = .44) in 250 subjects. Depression product appraisal increased in companies with greater health benefit generosity whose benefit professionals were male. Depression product purchasing behavior increased in small companies compared to large companies, companies who knew a vendor that sold depression products at baseline, companies with greater health benefit risk taking, and companies with less politicalization of health care benefit decision making.
Policy makers need to build innovative bridges to the employer community to convince them to purchase evidence-based benefits, even when benefits offer potential financial savings.
Trial registration
Clinical Trials Registration Number: NCT01013220.
PMCID: PMC4263121  PMID: 25248854
Depression care management; Employers; Return on investment; Academic detailing; Implementation science; Randomized trial
7.  Prescription Medication Burden in Patients with Newly-Diagnosed Diabetes: A SUrveillance, PREvention, and ManagEment of Diabetes Mellitus (SUPREME-DM) Study 
To understand the burden of medication use for newly-diagnosed diabetes patients both before and after diabetes diagnosis, and to identify subpopulations of newly-diagnosed diabetes patients who face a relatively high drug burden.
Retrospective cohort.
Eleven U.S. integrated health systems.
196,654 insured adults aged ≥20 diagnosed with newly-diagnosed diabetes from 1/1/2005 – 12/31/2009.
Main Outcome Measure
Number of unique therapeutic classes of drugs dispensed in the 12 months prior to, and 12 months post, the diagnosis of diabetes in 5 categories: overall, antihypertensive, antihyperlipidemic, mental health, and antihyperglycemic (post-period only).
The mean number of drug classes used by newly-diagnosed diabetes patients is high before diagnosis (5.0), and increases significantly afterwards (6.6, p<.001). Eighty-one percent of this increase is due to antihyperglycemic initiation and increased use of medications to control hypertension and lipid levels. Multivariate analyses showed that overall drug burden after diabetes diagnosis was higher in female, older, white, and obese patients, as well as among those with higher A1cs and comorbidity levels (p<.001 for all comparisons). The overall number of drug classes used by newly-diagnosed diabetes patients after diagnosis decreased slightly between 2005 and 2009 (p<.001).
Diabetes patients face significant drug burden to control diabetes and other comorbidities, and our data indicate an increased focus on cardiovascular disease risk factor control after diabetes diagnosis. However, total drug burden may be slightly decreasing over time. This information can be valuable to pharmacists working with newly-diagnosed diabetes patients to address their increasing drug regimen complexity.
PMCID: PMC4161641  PMID: 24860866
diabetes; medication burden; surveillance
8.  A scan statistic for identifying optimal risk windows in vaccine safety studies using self-controlled case series design 
Statistics in medicine  2013;32(19):3290-3299.
In examining the association between vaccines and rare adverse events after vaccination in post-licensure observational studies, it is challenging to define appropriate risk windows because pre-licensure randomized clinical trials provide little insight on the timing of specific adverse events. Past vaccine safety studies have often used pre-specified risk windows based on prior publications, biological understanding of the vaccine, and expert opinion. Recently, a data driven approach was developed to identify appropriate risk windows for vaccine safety studies that use the self-controlled case series design. This approach employs both the maximum incidence rate ratio and the linear relation between the estimated incidence rate ratio and the inverse of average person time at risk, given a specified risk window. In this paper, we present a scan statistic that can identify appropriate risk windows in vaccine safety studies using the self-controlled case series design while taking into account the dependence of time intervals within an individual and while adjusting for time-varying covariates such as age and seasonality. This approach uses the maximum likelihood ratio test based on fixed effects models, which has been used for analyzing data from self-controlled case series design in addition to conditional Poisson models.
PMCID: PMC4004030  PMID: 23303643
self-controlled case series; adverse events after vaccination; fixed effects model; scan statistics; maximum likelihood ratio test
9.  Use of Fixed Effects Models to Analyze Self-Controlled Case Series Data in Vaccine Safety Studies 
Conditional Poisson models have been used to analyze vaccine safety data from self-controlled case series (SCCS) design. In this paper, we derived the likelihood function of fixed effects models in analyzing SCCS data and showed that the likelihoods from fixed effects models and conditional Poisson models were proportional. Thus, the maximum likelihood estimates (MLEs) of time-varying variables including vaccination effect from fixed effects model and conditional Poisson model were equal. We performed a simulation study to compare empirical type I errors, means and standard errors of vaccination effect coefficient, and empirical powers among conditional Poisson models, fixed effects models, and generalized estimating equations (GEE), which has been commonly used for analyzing longitudinal data. Simulation study showed that both fixed effect models and conditional Poisson models generated the same estimates and standard errors for time-varying variables while GEE approach produced different results for some data sets. We also analyzed SCCS data from a vaccine safety study examining the association between measles mumps-rubella (MMR) vaccination and idiopathic thrombocytopenic purpura (ITP). In analyzing MMR-ITP data, likelihood-based statistical tests were employed to test the impact of time-invariant variable on vaccination effect. In addition a complex semi-parametric model was fitted by simply treating unique event days as indicator variables in the fixed effects model. We conclude that theoretically fixed effects models provide identical MLEs as conditional Poisson models. Because fixed effect models are likelihood based, they have potentials to address methodological issues in vaccine safety studies such as how to identify optimal risk window and how to analyze SCCS data with misclassification of adverse events
PMCID: PMC3976179  PMID: 24707443
Self-controlled case series; Adverse events after immunization; Fixed effects model; Longitudinal data; Conditional Poisson model
10.  Extension of Kaplan-Meier methods in observational studies with time-varying treatment 
Value in Health  2011;15(1):167-174.
Inverse probability of treatment weighted (IPTW) Kaplan-Meier estimates have been developed to compare two treatments in the presence of confounders in observational studies. Recently, stabilized weights were developed to reduce the influence of extreme IPTW weights in estimating treatment effects. The objective of this paper was to use adjusted Kaplan-Meier estimates and modified log-rank and Wilcoxon tests to examine the effect of a treatment which varies over time in an observational study.
In this paper, we propose stabilized weight (SW) adjusted Kaplan-Meier estimates and modified log-rank and Wilcoxon tests when the treatment is time-varying over the follow-up period. We applied these new methods in examining the effect of an anti-platelet agent, clopidogrel, on subsequent events, including bleeding, myocardial infarction, and death after a Drug-Eluting Stent was implanted into a coronary artery. In this population, clopidogrel use may change over time based on patients' behavior (e.g., non-adherence) and physicians' recommendations (e.g., end of duration of therapy). Consequently, clopidogrel use was treated as a time-varying variable.
We demonstrate that 1) the sample sizes at three chosen time points are almost identical in the original and weighted datasets, and 2) the covariates between patients on and off clopidogrel were well balanced after SWs were applied to the original samples.
The SW-adjusted Kaplan-Meier estimates and modified log-rank and Wilcoxon tests are useful in presenting and comparing survival functions for time-varying treatments in observational studies while adjusting for known confounders.
PMCID: PMC3267428  PMID: 22264985
Observational study; Kaplan Meier estimates; Stabilized weights; Time-varying treatment; Stents
11.  The Positive Predictive Value of a Hyperkalemia Diagnosis in Automated Health Care Data 
Pharmacoepidemiology and drug safety  2010;19(11):1204-1208.
Our objectives were to determine performance of coded hyperkalemia diagnosis at identifying 1) clinically-evident hyperkalemia and 2) serum potassium ≥ 6 mmol/liter.
This retrospective observational study included 8,722 patients with diabetes within an integrated healthcare system who newly-initiated an angiotensin converting enzyme inhibitor, angiotensin receptor blocker, or spironolactone. The primary outcome was first hyperkalemia-associated event (hospitalization, emergency department visit or death within 24 hours of coded diagnosis and/or potassium ≥ 6 mmol/liter) during the first year of therapy. Medical records were reviewed.
Among a random sample of 99 patients not coded as having hyperkalemia, none had hyperkalemia upon record review. Among all 64 patients identified as having hyperkalemia, all had hospitalization or emergency department visit associated with coded diagnosis or elevated potassium. Of 55 with coded diagnosis, 42 (PPV 76%) had clinically-evident hyperkalemia; 32 (PPV 58%) had potassium ≥ 6. Of 9 identified using only potassium ≥ 6, 7 (PPV 78%) had clinically-evident hyperkalemia.
Nearly one-fourth of patients with coded diagnosis do not have clinically-evident hyperkalemia and nearly one-half do not have potassium ≥ 6. Because both false positives and negatives occur with coded diagnoses, medical record validation of hyperkalemia-associated outcomes is necessary.
PMCID: PMC2996391  PMID: 20878650
Hyperkalemia; positive predictive value; sensitivity; specificity; ACEi; ARB
12.  Diabetes and Drug-Associated Hyperkalemia: Effect of Potassium Monitoring 
Renin-angiotensin-aldosterone system (RAAS) inhibitors are associated with hyperkalemia, but there is little evidence demonstrating patients who receive potassium monitoring have a lower rate of hyperkalemia.
To evaluate the association between potassium monitoring and serious hyperkalemia-associated adverse outcomes among patients with diabetes newly initiating RAAS inhibitor therapy.
Retrospective observational study.
Patients with diabetes without end-stage renal disease initiating RAAS inhibitor therapy between 2001 and 2006 at three integrated health care systems.
Potassium monitoring and first hyperkalemia-associated adverse event during the initial year of therapy. Hyperkalemia-associated adverse events included hospitalizations, emergency department visits or deaths within 24 h of hyperkalemia diagnosis and/or diagnostic potassium ≥6 mmol/l. Incidence rates were calculated in person-years (p-y). We used inverse probability propensity score weighting to adjust for differences between patients with and without monitoring; Poisson regression was used to obtain adjusted relative risks.
A total of 19,391 of 27,355 patients (71%) received potassium monitoring. Serious hyperkalemia-associated events occurred at an incidence rate of 10.2 per 1,000 p-y. Compared to patients without monitoring, adjusted relative risk of hyperkalemia-associated adverse events among all patients with monitoring was 0.50 (0.37, 0.66); in the subset of patients who also had chronic kidney disease (n = 2,176), adjusted relative risk was 0.29 (0.18, 0.46).
Patients prescribed RAAS inhibitors who have both diabetes and chronic kidney disease and receive potassium monitoring are less likely to experience a serious hyperkalemia-associated adverse event compared to similar patients who did not receive potassium monitoring. This evidence supports existing consensus-based guidelines.
PMCID: PMC2842549  PMID: 20087674
hyperkalemia; hyperpotassemia; angiotensin-converting enzyme inhibitor; ACEi; angiotensin receptor blocker; ARB; spironolactone; RAAS inhibitor
13.  Impact of the introduction of pneumococcal conjugate vaccine on immunization coverage among infants 
BMC Pediatrics  2005;5:43.
The introduction of pneumococcal conjugate vaccine (PCV) to the U.S. recommended childhood immunization schedule in the year 2000 added three injections to the number of vaccinations a child is expected to receive during the first year of life. Surveys have suggested that the addition of PCV has led some immunization providers to move other routine childhood vaccinations to later ages, which could increase the possibility of missing these vaccines. The purpose of this study was to evaluate whether introduction of PCV affected immunization coverage for recommended childhood vaccinations among 13-month olds in four large provider groups.
In this retrospective cohort study, we analyzed computerized data on vaccinations for 33,319 children in four large provider groups before and after the introduction of PCV. The primary outcome was whether the child was up to date for all non-PCV recommended vaccinations at 13 months of age. Logistic regression was used to evaluate the association between PCV introduction and the primary outcome. The secondary outcome was the number of days spent underimmunized by 13 months. The association between PCV introduction and the secondary outcome was evaluated using a two-part modelling approach using logistic and negative binomial regression.
Overall, 93% of children were up-to-date at 13 months, and 70% received all non-PCV vaccinations without any delay. Among the entire study population, immunization coverage was maintained or slightly increased from the pre-PCV to post-PCV periods. After multivariate adjustment, children born after PCV entered routine use were less likely to be up-to-date at 13 months in one provider group (Group C: OR = 0.5; 95% CI: 0.3 – 0.8) and were less likely to have received all vaccine doses without any delay in two Groups (Group B: OR = 0.4, 95% CI: 0.3 – 0.6; Group C: OR = 0.5, 95% CI: 0.4 – 0.7). This represented 3% fewer children in Group C who were up-to-date and 14% (Group C) to 16% (Group B) fewer children who spent no time underimmunized at 13 months after PCV entered routine use compared to the pre-PCV baseline. Some disruptions in immunization delivery were also observed concurrent with temporary recommendations to suspend the birth dose of hepatitis B vaccine, preceding the introduction of PCV.
These findings suggest that the introduction of PCV did not harm overall immunization coverage rates in populations with good access to primary care. However, we did observe some disruptions in the timely delivery of other vaccines coincident with the introduction of PCV and the suspension of the birth dose of hepatitis B vaccine. This study highlights the need for continued vigilance in coming years as the U.S. introduces new childhood vaccines and policies that may change the timing of existing vaccines.
PMCID: PMC1314888  PMID: 16313673

Results 1-13 (13)