Susan Shetterly

Medication nonadherence is associated with a broad range of adverse outcomes in patients with coronary artery disease

BACKGROUND Little is known about the effect of nonadherence among patients with coronary artery disease (CAD) on a broad spectrum of outcomes including cardiovascular mortality, cardiovascular hospitalizations, and revascularization procedures. METHODS This was a retrospective cohort study of 15,767 patients with CAD. Medication adherence was calculated as proportion of days covered for filled prescriptions of beta-blockers, angiotensin-converting enzyme (ACE) inhibitors, and statin medications. Multivariable Cox regression assessed the association between medication nonadherence as a time-varying covariate and a broad range of outcomes, adjusting for demographics and clinical characteristics. Median follow-up was 4.1 years. RESULTS Rates of medication nonadherence were 28.8% for beta-blockers, 21.6% for ACE inhibitors, and 26.0% for statins. In unadjusted analysis, nonadherence to each class of medication was associated with higher all-cause and cardiovascular mortality. In multivariable analysis, nonadherence remained significantly associated with increased all-cause mortality risk for beta-blockers (hazard ratio [HR] 1.50, 95% CI 1.33-1.71), ACE inhibitors (HR 1.74, 95% CI 1.52-1.98), and statins (HR 1.85, 95% CI 1.63-2.09). In addition, nonadherence remained significantly associated with higher risk of cardiovascular mortality for beta-blockers (HR 1.53, 95% CI 1.16-2.01), ACE inhibitors (HR 1.66, 95% CI 1.26-2.20), and statins (HR 1.62, 95% CI 1.124-2.13). The findings of increased risk associated with nonadherence were consistent for cardiovascular hospitalization and revascularization procedures. CONCLUSIONS Nonadherence to cardioprotective medications is common in clinical practice and associated with a broad range of adverse outcomes. These findings suggest that medication nonadherence should be a target for quality improvement interventions to maximize the outcomes of patients with CAD.

Executive Cognitive Abilities and Functional Status Among Community‐Dwelling Older Persons in the San Luis Valley Health and Aging Study

OBJECTIVES: The purpose of this study was to evaluate the contribution of the executive cognitive functions to self‐reported and observed performance of activities of daily living and instrumental activities of daily living.

Health Literacy and Outcomes Among Patients With Heart Failure

CONTEXT Little is known about the effects of low health literacy among patients with heart failure, a condition that requires self-management and frequent interactions with the health care system. OBJECTIVE To evaluate the association between low health literacy and all-cause mortality and hospitalization among outpatients with heart failure. DESIGN, SETTING, AND PATIENTS Retrospective cohort study conducted at Kaiser Permanente Colorado, an integrated managed care organization. Outpatients with heart failure were identified between January 2001 and May 2008, were surveyed by mail, and underwent follow-up for a median of 1.2 years. Health literacy was assessed using 3 established screening questions and categorized as adequate or low. Responders were excluded if they did not complete at least 1 health literacy question or if they did not have at least 1 year of enrollment prior to the survey date. MAIN OUTCOME MEASURES All-cause mortality and all-cause hospitalization. RESULTS Of the 2156 patients surveyed, 1547 responded (72% response rate). Of 1494 included responders, 262 (17.5%) had low health literacy. Patients with low health literacy were older, of lower socioeconomic status, less likely to have at least a high school education, and had higher rates of coexisting illnesses. In multivariable Cox regression, low health literacy was independently associated with higher mortality (unadjusted rate, 17.6% vs 6.3%; adjusted hazard ratio, 1.97 [95% confidence interval, 1.3-2.97]; P = .001) but not hospitalization (unadjusted rate, 30.5% vs 23.2%; adjusted hazard ratio, 1.05 [95% confidence interval, 0.8-1.37]; P = .73). CONCLUSION Among patients with heart failure in an integrated managed care organization, low health literacy was significantly associated with higher all-cause mortality.

Dietary Fat Predicts Conversion From Impaired Glucose Tolerance to NIDDM: The San Luis Valley Diabetes Study

OBJECTIVE To determine if dietary fat intake measured at a baseline exam in subjects with impaired glucose tolerance (IGT) predicted the subsequent development of non-insulin-dependent diabetes mellitus (NIDDM). RESEARCH DESIGN AND METHODS Based on an oral glucose tolerance test (OGTT) (World Health Organization criteria), we identified 134 eligible subjects with IGT from a geographically based sample of subjects with no prior history of diabetes. One to three years after the baseline exam, 123 subjects (92%) had a repeat OGTT. Diet was assessed by a 24-h diet recall reported before the baseline OGTT. RESULTS The mean percentage of energy eaten as fat was 43.4% in 20 people subsequently developing NIDDM compared with 40.6% in 43 people remaining IGT and 38.9% in 60 subjects who subsequently reverted to normal glucose tolerance. In comparing the 20 subjects who developed NIDDM with the 103 who remained IGT or normal, an increase in fat intake of 40 g/day was associated with an increase in risk of NIDDM of 3.4-fold (95% confidence interval [CI] 0.8–13.6) adjusted for energy intake, age, sex, ethnicity, and obesity. The odds ratio increased to sixfold (95% CI 1.2–29.8) after adjustment for fasting glucose, insulin, and 1-h insulin. CONCLUSIONS Fat consumption significantly predicts NIDDM risk in subjects with IGT after controlling for obesity and markers of glucose metabolism.

Importance of Therapy Intensification and Medication Nonadherence for Blood Pressure Control in Patients With Coronary Disease

BACKGROUND Despite the importance of blood pressure (BP) control in secondary prevention, a significant proportion of patients with coronary disease have uncontrolled BP. METHODS This retrospective cohort study of patients with coronary disease (N = 10 447) evaluated the impact of medication nonadherence and therapy intensification on reaching target BP goals. Medication adherence was calculated as the proportion of days covered for filled prescriptions of antihypertensive medications. Therapy intensification included dosage increase or increase in number of antihypertensive medications. The primary outcome was uncontrolled systolic BP (SBP) over time, using a latent class model that incorporated longitudinal SBP data and assigned patients to SBP trajectory groups. Multivariable regression evaluated the association between medication nonadherence (ie, proportion of days covered, <0.80) and therapy intensification with SBP control over time, with adjustment for demographics and clinical characteristics. RESULTS Three SBP trajectory groups were identified: (1) patients with BP that remained controlled (ie, SBP, <or=140 mm Hg) over time (n = 9114 [87.2%]); (2) patients with high BP that became controlled (n = 779 [7.5%]); and (3) patients with BP that remained high over time (n = 554 [5.3%]). In multivariable analyses, therapy intensification (odds ratio, 1.31; 95% confidence interval, 1.01-1.70) and medication nonadherence (odds ratio, 1.73; 95% confidence interval, 1.34-2.24) were associated with uncontrolled BP compared with high SBP that became controlled over time. CONCLUSIONS These findings suggest that medication nonadherence can help explain why BP levels remained elevated despite intensification of antihypertensive medications. Successful BP control is seen with a combination of intensification and adherence, suggesting that therapy intensification must be coupled with interventions to enhance medication adherence.

Effectiveness of the 2003–2004 Influenza Vaccine Among Children 6 Months to 8 Years of Age, With 1 vs 2 Doses

Objective. To evaluate the effectiveness of 1 and 2 doses of the 2003–2004 influenza vaccine in preventing medically attended influenza-like illness (ILI) among children 6 to 23 months and 6 months to 8 years of age. Design and Methods. Outpatient and emergency department visits and immunization records were used to conduct a retrospective cohort study among children 6 months to 8 years of age. ILI and pneumonia and influenza (P&I) outcomes were defined on the basis of International Classification of Diseases, Ninth Revision, codes. Influenza vaccine effectiveness (VE) was calculated as (1 − hazard rate ratio) × 100. Results. A total of 29726 children were included in the analyses; 17.3% were 6 to 23 months of age. By November 19, 2003, the start of peak influenza activity, 7.5% and 9.9% of children 6 months to 8 years were fully or partially vaccinated against influenza, respectively. For fully vaccinated children 6 to 23 months of age, VE against ILI and P&I was 25% and 49%, respectively. No statistically significant reduction in ILI or P&I rates was observed for partially vaccinated children 6 to 23 months of age (−3% and 22%, respectively). For fully vaccinated children 6 months to 8 years of age, VE against ILI and P&I was 23% and 51%, respectively. For partial vaccination, VE was significant only for P&I (23%). Conclusions. Despite a suboptimal match between the influenza vaccine and predominant circulating strains, influenza vaccination provided substantial protection for fully vaccinated children and possibly some protection for partially vaccinated children <9 years of age. These findings support vaccinating targeted children even when the vaccine match is suboptimal, and they highlight the need to vaccinate previously unvaccinated children with 2 doses for optimal protection.

The association of comorbidities, utilization and costs for patients identified with low back pain

BackgroundExisting studies have examined the high prevalence of LBP along with the high treatment costs of patients with low back pain (LBP). Various factors have been shown to be correlated or predictive of chronic or episodic LBP including the characteristics of the initial episode, pain, comorbid conditions, psychosocial issues, and opiate use. This study replicates and extends earlier studies by examining the association of patient characteristics including baseline comorbidities with patterns of healthcare service use and cost.MethodsThis is a retrospective analysis of measures of comorbidities, healthcare use, and cost for patients identified with LBP, stratified by the number of LBP episodes. Administrative data associated with outpatient and hospital based care for the years 1996 through 2001, were used to identify adult patients with LBP. LBP patients continuously enrolled for 12 months prior and 24 months after their initial LBP event were included in the study. A LBP episode was identified as the number of 30-day periods where a patient had one or more healthcare events with a diagnosis consistent with LBP. Chi-square and multivariate regression analyses were employed to estimate the variation in utilization and costs.ResultsOf 16,567 patients enrolled, 67% were identified with only one LBP episode and 4.5% had ≥6. The prevalence of comorbidities, analgesic use, and healthcare service use, varied by the number of back pain episodes. Diabetes, rheumatoid arthritis, anxiety, psychotic illness, depression, use of opiates and NSAIDs were associated with significant incremental increases in costs (P < .003).ConclusionPhysical and mental health co-morbidities and measures of analgesic use were associated with chronicity, healthcare utilization and costs. Given the association of comorbidities and cost for patients with LBP, management approaches that are effective across chronic illnesses may prove to be beneficial for high cost patients identified with LBP.

Relationship Between Habitual Physical Activity and Insulin Levels Among Nondiabetic Men and Women: San Luis Valley Diabetes Study

Objective To determine whether higher levels of physical activity would be associated with lower fasting insulin and C-peptide levels in a free-living nondiabetic population. Research Design and Methods A cross-sectional study was conducted with a Hispanic and non-Hispanic white population of 442 men and 489 women with normal glucose tolerance (by World Health Organization criteria) in two rural Colorado counties. Total physical activity was assessed by a 7-day physical activity recall from which metabolic equivalents were estimated. Relationships between metabolic equivalents and fasting insulin and C-peptide were assessed while considering obesity, age, and other risk factors known to influence fasting insulin levels. Results Among all subjects, univariate analyses showed that higher activity levels were associated with lower mean fasting insulin and C-peptide levels (P < or equal to 0.05). Multiple linear regression showed that higher activity was significantly associated with lower values of log fasting insulin and C-peptide levels in men only (P < 0.001) independent of obesity, fat distribution, and age. Men in the highest tertile of activity had an adjusted mean fasting insulin level of 59.2 pM and fasting C-peptide level of 0.5 nM compared with a fasting insulin level of 72.7 pM and fasting C-peptide level of 0.6 mM for men in the lowest tertile of activity. The magnitude of the inverse association between activity and insulin was greatest in older rather than younger men. Physical activity was not associated with fasting insulin or C-peptide levels in women in the multivariate analyses. Conclusions Based on cross-sectional data, we conclude that higher levels of habitual physical activity are associated with lower fasting insulin and C-peptide levels in Hispanic and non-Hispanic white men.

Use of Stabilized Inverse Propensity Scores as Weights to Directly Estimate Relative Risk and Its Confidence Intervals

OBJECTIVES Inverse probability of treatment weighting (IPTW) has been used in observational studies to reduce selection bias. For estimates of the main effects to be obtained, 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. METHODS 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 because of 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. RESULTS 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 with a renin-angiotensin-aldosterone system inhibitor. The incidence rate ratio (with monitoring vs. without monitoring) and confidence intervals were 0.46 (0.34, 0.61) using the stabilized weights compared with 0.46 (0.38, 0.55) using typical IPTW. CONCLUSIONS 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.

Patterns and predictors of hypertension incidence among Hispanics and non-Hispanic whites: the San Luis Valley Diabetes Study.

Objectives: To determine whether Hispanics are at lower risk for the development of hypertension than non-Hispanic Whites. We also examined selected predictors of hypertension incidence and explored the role of markers of insulin resistance in the development of hypertension. Design A cohort study of a geographically-based sample of Hispanic and non-Hispanic white southern Colorado residents who were re-examined an average of 4 years after their baseline examination. Methods: These analyses included 664 participants who were normotensive and confirmed nondiabetic by an oral glucose tolerance test at their baseline examination. Hypertension was defined as systolic blood pressure ± 140 mmHg or diastolic blood pressure ± 90 mmHg or use of antihypertensive medication. Results: Hispanics and non-Hispanic Whites had similar hypertension incidence rates. The strongest predictors of hypertension incidence were baseline blood pressure and age. Higher baseline heart rates and higher body mass index also predicted hypertension. Increased fasting insulin levels were associated with hypertension incidence among lean participants, though the association disappeared once baseline blood pressure levels were added to the models. Models investigating change in systolic or diastolic blood pressure levels found higher baseline levels of insulin area under the glucose tolerance curve predicted greater increases in systolic blood pressure in non-Hispanic Whites only. Conclusions: Hypertension incidence rates were similar in Hispanics and non-Hispanic Whites. Higher levels of insulin area were associated with larger increases in systolic blood pressure among non-Hispanic Whites only.