Currently Viewing:
The American Journal of Managed Care August 2010
Clinical and Economic Outcomes After Introduction of Drug-Eluting Stents
Charanjit S. Rihal, MD, MBA; James L. Ryan, MHA; Mandeep Singh, MBBS; Ryan J. Lennon, MS; John F. Bresnahan, MD; Juliette T. Liesinger, BA; Bernard J. Gersh, MBChB, DPhil; Henry H. Ting, MD, MBA; David R. Holmes, Jr, MD; and Kirsten Hall Long, PhD
Antihypertensive Medication Adherence and Subsequent Healthcare Utilization and Costs
Donald G. Pittman, PharmD; Zhuliang Tao, MPH; William Chen, PhD; and Glen D. Stettin, MD
Statin Therapy for Elevated hsCRP: What Are the Public Health Implications?
Paul M. Ridker, MD
Statin Therapy for Elevated hsCRP: More Evidence Is Needed
Ogochukwu C. Molokwu, PharmD, MScMed
Opening and Continuing the Discussion on Influenza Vaccination Timing
Kellie J. Ryan, MPH Matthew D. Rousculp, PhD, MPH. Reply by Bruce Y. Lee, MD, MBA Julie H. Y. Tai, MD Rachel R. Bailey, MPH
Currently Reading
Cost Sharing, Adherence, and Health Outcomes in Patients With Diabetes
Teresa B. Gibson, PhD; Xue Song, PhD; Berhanu Alemayehu, DrPH; Sara S. Wang, PhD; Jessica L. Waddell, MPH; Jonathan R. Bouchard, MS, RPh; and Felicia Forma, BSc
Value and the Medical Home: Effects of Transformed Primary Care
Richard J. Gilfillan, MD; Janet Tomcavage, RN, MSN; Meredith B. Rosenthal, PhD; Duane E. Davis, MD; Jove Graham, PhD; Jason A. Roy, PhD; Steven B. Pierdon, MD; Frederick J. Bloom Jr, MD, MMM; Thomas R. Graf, MD; Roy Goldman, PhD, FSA; Karena M. Weikel, BA; Bruce H. Hamory, MD; Ronald A. Paulus, MD, MBA; and Glenn D. Steele Jr, MD, PhD
Insomnia Risks and Costs: Health, Safety, and Quality of Life
Mark R. Rosekind, PhD; and Kevin B. Gregory, BS
Effect of Inadequate Response to Treatment in Patients With Depression
Russell L. Knoth, PhD; Susan C. Bolge, PhD; Edward Kim, MD, MBA; and Quynh-Van Tran, PharmD
Using Search Engine Query Data to Track Pharmaceutical Utilization: A Study of Statins
Nathaniel M. Schuster, BS; Mary A.M. Rogers, PhD, MS; and Laurence F. McMahon Jr, MD, MPH
Employees' Willingness to Pay to Prevent Influenza
Stephen S. Johnston, MA; Matthew D. Rousculp, PhD, MPH; Liisa A. Palmer, PhD; Bong-Chul Chu, PhD; Parthiv J. Mahadevia, MD, MPH; and Kristin L. Nichol, MD, MPH, MBA
Anticonvulsant Use After Formulary Status Change for Brand-Name Second-Generation Anticonvulsants
Hemal Patel, PharmD, MBA; Diana C. Toe, PharmD; Shawn Burke, RPh; and Rafia S. Rasu, PhD

Cost Sharing, Adherence, and Health Outcomes in Patients With Diabetes

Teresa B. Gibson, PhD; Xue Song, PhD; Berhanu Alemayehu, DrPH; Sara S. Wang, PhD; Jessica L. Waddell, MPH; Jonathan R. Bouchard, MS, RPh; and Felicia Forma, BSc

Higher cost-sharing levels reduced adherence to antidiabetic medications in patients with type 2 diabetes.

A 2-stage model was used to estimate the relationship between antidiabetic medication (OAD with or without insulin) adherence and health outcomes.

First Stage: Cost Sharing and Adherence. A logistic regression model was estimated for adherence (PDC >80%) in the 18-month time period (January 2003 through June 2004). Explanatory variables included sociodemographic variables, employee status, health plan type, physician specialist visit, health status, and cost sharing.

Second Stage: Adherence and Health Outcomes. In the second stage, generalized linear models estimated events between July 2004 through June 2006. All models included the first-stage explanatory variables, except for the cost-sharing variables. The cost-sharing variables (cost-sharing index, coinsurance flag, office visit cost sharing, and previous  mail-order use) served as instruments in the first-stage equation and thus were not included in the second stage. Using a 2-stage residual inclusion model, actual patient   adherence and the residual from the first-stage model also were included in the second-stage model. The residual was included to produce consistent estimates in the nonlinear second stage and to adjust for any unobservable confounding in the second stage.20 The coefficient of the residual term in the second stage also allowed an indication of   endogeneity. If the coefficient of the residual is statistically significant, the inclusion of the residual term aids in reducing selection effects, making the estimates consistent. If the coefficient of the residual term is not statistically significant, consistent estimates are produced, possibly with a loss of efficiency.

Each complication model was estimated using a logit link and a binomial family for the binary outcomes. Using 2 versions of the complication models to focus on the development of complications in the measurement period, complication models were estimated (1) controlling for previous complications by including indicator variables for the presence of each complication in 2002 and (2) for the subset of patients with no complications in 2002. The utilization and productivity models were estimated using a log link and a negative binomial model to reflect outcomes expressed as counts.

We also estimated the reduced form of each of the outcome equations to determine the strength and direction of the association between copayments and each outcome.

RESULTS

The characteristics of the full and OAD-only samples are presented in Table 2. The full sample was 46% female and had an average age of 52.2 years. The majority of the full sample lived in the South, and three-fourths lived in urban areas. The median income in the patients’ area of residence averaged about $44,000. More than two-thirds of patients were employees (67.5%), and the majority of patients were enrollees in a preferred provider organization. The mean cost-sharing index amount was $11.45. Seventeen percent of individuals were subject to coinsurance, more than one-quarter (28.5%) had used a mail-order pharmacy in the past year, and the average physician office visit copayment was  about $20. Similar demographics were observed among OAD-only users.

Almost three-quarters of patients were adherent (PDC >80%) from January 2003 through June 2004. Complication rates for OAD-only users ranged from 1.4% for an acute myocardial infarction (AMI) to 7.3% for retinopathy. Complication rates were higher (2.0% for an AMI and 13.5% for retinopathy) for all OAD users. Patients visited the ED on average less than once and had inpatient admissions less frequently than ED visits in the 2-year time frame. Employees reported slightly more than 30 days absent. OAD-only users had  13.3 short-term disability days, and the full sample had 17.4 short-term disability days.

Table 3 presents results from the first-stage logistic regression estimates for adherence. Notably, as the level of prescription drug cost sharing (measured as the cost-sharing index) increased, adherence decreased. For all OAD users, the odds ratio (OR) was 0.974 (95% confidence interval [CI] = 0.970, 0.978). For OAD-only users, the OR was 0.978 (95% CI = 0.973, 0.984). Higher cost sharing for physician visits also was associated with lower levels of adherence, although the effects were not as large. For all OAD users, the OR was 0.996 (95% CI = 0.994, 0.997). For OAD-only users, the OR was 0.995 (95% CI = 0.993, 0.996). Patients with coinsurance had higher levels of adherence, as did patients who had previous mail-order use. All instruments were individually and jointly statistically significant in the first-stage (adherence) models (P <.01).

Figure 1 shows the adjusted relationship between cost-sharing index amounts ranging from $10 to $30 and the predicted probability of adherence. An increase from $10 to $20 in the cost-sharing index resulted in an average 4.2 percentage point reduction in the probability of being 80% adherent for OAD-only users and a 4.8 percentage point reduction for all OAD users.

Consistent with previous studies, the effects of selected explanatory variables on adherence were in the expected direction.8,21 For example, females had lower levels of adherence, and as age increased, adherence also increased. Patients in higher-income areas exhibited higher levels of adherence.8 Patients with a greater number of comorbidities (as indicated by the Charlson Comorbidity Index) were more adherent, which is consistent with descriptive findings in the article by Encinosa et al,5 who reported that patients with more chronic conditions were more adherent. Chronic out-of-pocket payments were not associated with adherence, although thechronic out-of-pocket-payment variable was a lagged variable (measured during the year before the adherence time frame), so it was intended to represent the amount of copayment burden. Because of its construction, it is likely to be   a proxy for health status. Perhaps the time separation in measurement affected the strength of the association.

The relationship between adherence and events is detailed in Table 4A and Table 4B. The full sample of OAD users who were adherent had a lower likelihood of all complications. OAD-only users who were adherent had a lower likelihood of amputation/ulcers, AMI, neuropathy, renal events, and retinopathy. Among OAD-only users, cerebrovascular disease and peripheral vascular disease rates were not significantly different for adherent patients when using the 2-stage residual inclusion approach. However, in both cases, the  coefficient of the residual indicated that there was little likelihood of endogeneity, so we reestimated the models with a 1-stage approach and found significantly lower rates of  cerebrovascular disease for adherent patients (OR = 0.900; 95% CI = 0.829, 0.977) and continued to find no significant difference in peripheral vascular disease rates.

Figure 2 shows the predicted mean effect size of each of these complications for adherent and nonadherent patients in terms of the likelihood of each complication. For example,   1.8% of patients adherent to antidiabetic medications in January 2003 through June 2004 were likely to have an AMI in the next 2 years, and more than twice as many (4%)   nonadherent patients were likely to have an AMI.

The number of ED visits was significantly lower among adherent patients, while the number of physician visits was higher among adherent patients. Inpatient admission rates were no different for adherent and nonadherent OAD-only users, but were lower for all OAD users. As with cerebrovascular disease, because there was no indication of endogeneity in the 2-stage residual inclusion inpatient admission models for OAD-only users, we reestimated the inpatient admission outcome model with a 1-stage model and found that adherence was associated with lower admission rates (incidence rate ratio = 0.808; 95% CI = 0.769, 0.850). The number of days of short-term disability was significantly lower for adherent patients (P <.01 for both the full sample and the OAD-only users). When calculating the predicted mean effect size for short-term disability, nonadherent patients in the full sample received an average of 18 more days of short-term disability (40 days nonadherent, 22 days adherent) (not shown). Nonadherent OAD-only users received an average of 8 more days of short-term disability (24 days) than adherent patients (16 days) (not shown).

Absence effects were mixed, with slightly higher rates of absence among patients who were adherent. However, OAD-only users who were adherent had the same rates of absence as those who were nonadherent.

Table 5 summarizes the direction and strength of the reduced-form coefficients of the prescription drug and physician visit copayment variables on each set of outcomes. In about three-quarters of the reduced-form models, higher copayments were associated with higher levels of each adverse outcome, although not all estimates were statistically significant.

DISCUSSION

In this large cohort of patients with type 2 diabetes who were users of OAD and enrolled in employer-sponsored plans, we found a wealth of positive effects of adherence to  antidiabetic medications. Importantly, this study found that adherent patients had lower rates of the diabetes-related complications examined, a finding that extends previous work  on adherence and diabetes-related outcomes to include complications. Ho and colleagues found that nonadherence to OAD, antihypertensives, or statins was related to increased A1C levels, higher systolic and diastolic blood pressure, and higher low-density lipoprotein cholesterol levels.7 Further, Hunt and colleagues linked higher levels of patient cost  sharing with inadequate adherence and ultimately increased A1C levels.9

This study also found that adherence to OAD among the full patient sample was associated with lower rates of ED visits and inpatient admissions. Existing literature shows an  inverse relationship between medication adherence and hospitalizations or ED visits.4-7,17,22,23 Our results corroborate these findings among the commercially insured in  employer-sponsored plans.

We also found that adherence was associated with productivity benefits in terms of fewer short-term disability days, revealing that the benefits of adherence affect  productivity.However, in the overall cohort of OAD users (OAD with or without insulin), absence rates were higher among adherent compared with nonadherent employees, which  may correspond to the length of illness. For instance, nonadherent patients may require longer spells of absence, constituting short-term disability, whereas adherent patients may  incur short-term absences covered by vacation or sick days.

We also found that prescription drug and physician copayments were inversely associated with adherence. In the first-stage models, we found that an increase from $10 to $20 in  the cost-sharing index was associated with an average 4.2 percentage point reduction in the probability of being 80% adherent for OAD-only users and a 4.8 percentage point reduction for the full sample. This association translates to a price elasticity in the OAD-only group of −0.054 and of −0.062 among all OAD users.

 
Copyright AJMC 2006-2019 Clinical Care Targeted Communications Group, LLC. All Rights Reserved.
x
Welcome the the new and improved AJMC.com, the premier managed market network. Tell us about yourself so that we can serve you better.
Sign Up