Cost-Related Nonadherence and Mortality in Patients With Chronic Disease

A Multiyear Investigation, National Health Interview Survey, 2000-2014

Sarah C. Van Alsten, MPH; Jenine K. Harris, PhD


Prev Chronic Dis. 2020;17(12):e151 

In This Article

Abstract and Introduction


Introduction: Prescription costs are rising, and many patients with chronic illnesses have difficulty paying for prescriptions. Missing or delaying medication because of financial concerns is common; however, the effects of cost-related nonadherence (CRN) on patient outcomes have not been described. Our objective was to determine if CRN is associated with higher all-cause and disease-specific mortality among patients living with diabetes and cardiovascular disease in a representative sample of US adults.

Methods: We ascertained CRN, vital status, and cause of death for 39,571 patients with diabetes, 61,968 patients with cardiovascular disease, and 124,899 patients with hypertension in the 2000 through 2014 releases of the National Health Interview Survey. We used adjusted Cox proportional hazards models to estimate associations between CRN and all-cause mortality and CRN and disease-specific mortality.

Results: On average, 15% of the sample reported CRN in the year before interview. After adjusting for confounders, CRN was associated with 15% to 22% higher all-cause mortality rates for all conditions (diabetes hazard ratio [HR] = 1.18; 95% CI, 1.1–1.3; cardiovascular disease [CVD] HR = 1.15; 95% CI, 1.1–1.2; hypertension HR = 1.22; 95% CI, 1.2–1.3). Relative to no CRN, CRN was associated with 8% to 18% higher disease-specific mortality rates (diabetes HR = 1.18; 95% CI, 1.0–1.4; CVD HR = 1.09; 95% CI, 1.0–1.2; hypertension HR = 1.08; 95% CI, 0.9–1.3).

Conclusion: Relative to full adherence, CRN is associated with higher mortality rates for patients with diabetes, cardiovascular disease, and hypertension, although associations may have weakened since 2011. Policies that increase prescription affordability may decrease mortality for patients experiencing CRN.


The prevalence of diabetes among US adults is 15%, and the prevalence of cardiovascular disease (CVD) is 13%.[1,2] Diabetes is the seventh leading cause of death in the United States and CVD is the first leading cause;[3] both are associated with substantial economic burden. Together, diabetes and CVD accounted for nearly $200 billion in US health care costs in 2013; hypertension treatment accounted for another $83 billion.[4] A substantial portion of these costs are from prescription expenditures, which continue to increase at a rate of 5% to 6% annually.[4]

Among people with diabetes or hypertension, cost is the most common reason for medication nonadherence, with more than two-thirds of patients skipping or delaying medication.[5] Such financially motivated nonadherence behaviors, including not being able to afford needed prescriptions, are collectively known as cost-related nonadherence (CRN);[6] CRN differs from other common forms of nonadherence, such as fear of medication side effects and lack of perceived need, which are determined by material insufficiency rather than psychological factors.[5]

The prevalence of CRN in the US is high, particularly among patients with chronic conditions. In total, 6% to 7% of US adults reported at least 1 form of CRN in 2014.[7] In adults with type 1 or type 2 diabetes, 25% reported rationing insulin in the previous year to manage costs, 3.2% reported rationing insulin on a daily basis, and 40% reported not discussing underuse with their physician.[8,9]

Despite prevalence of CRN, few studies have investigated its implications. Generally, nonadherence is associated with greater risk for hypertension, hypercholesterolemia, and elevated hemoglobin A1c levels in people with diabetes,[9,10] and with greater risk for dyslipidemia and extended hospitalizations in patients with hypertension or CVD.[11] Although studies have documented adverse consequences of medication nonadherence, they have not specified reasons for nonadherence (ie, nonadherence because of economic factors versus psychological factors); it is unclear how CRN contributes to these outcomes. The objective of our study was to determine whether CRN is associated with higher risk of mortality in US adults with diabetes, CVD, or hypertension.