Color coding corresponds to retail prices of the combination of digoxin, 0.25 mg/d; lisinopril, 40 mg/d; and carvedilol, 25 mg twice daily.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Hauptman PJ, Goff ZD, Vidic A, Chibnall JT, Bleske BE. Variability in Retail Pricing of Generic Drugs for Heart Failure. JAMA Intern Med. 2017;177(1):126–128. doi:10.1001/jamainternmed.2016.6955
Copyright 2017 American Medical Association. All Rights Reserved.
Generic medications may lessen patients’ financial burden and improve adherence.1 Recent increases in generic drug costs2 raise concerns about the effect on uninsured and underinsured patients whose options may be restricted to retail pharmacies within a limited geographic area. An estimated 7.3 million Americans with cardiovascular disease are in the uninsured group.3 Therefore, we evaluated US retail pharmacy pricing for generic guideline-directed heart failure (HF) drugs in a metropolitan area as a function of dose, supply, pharmacy type and zip code, and zip code median annual income.
Pharmacies were identified across 55 zip codes in a 2-state region using a government website.4 In zip codes with more than 4 pharmacies, the nonrandomized convenience sample included at least half of these stores. Pharmacies were queried by phone during a 3-week period in May 2016 regarding cost, without insurance, of digoxin (0.125 mg/d and 0.25 mg/d), lisinopril (10 mg/d and 40 mg/d), and carvedilol (6.25 mg and 25 mg twice daily) for 30- and 90-day supplies. Median annual income by zip code was determined from US Census Bureau data. Pricing between groups was compared using Kruskal-Wallis and Mann-Whitney tests; associations were assessed with Spearman ρ correlations. Cluster analysis was used to create groups based on zip code and zip code–based median annual income. The TwoStep cluster procedure (SPSS-22) was used to determine the optimal number of natural groupings based on a set of variables such that similarity of cases within a cluster and differences between clusters were maximized.
Data were provided by 153 chain and 22 independent pharmacies (Figure). Median annual income within pharmacy zip codes was $53 122 (range, $10 491-$112 017). The number of manufacturers varied by drug (digoxin, 7; lisinopril, 9; carvedilol, 8). The price range was wide (Table); for example, the cost of a 30-day supply of digoxin (0.25 mg) plus higher-dose lisinopril and carvedilol varied from $12.00 to $397.58 (median price, $70.68). Small percentages of pharmacies charged less than $25 for 30-day supplies and less than $100 for 90-day supplies for all 3 drugs (1.7% [n = 3] and 5.3% [n = 9], respectively). Digoxin was consistently the most expensive drug. Only 1 chain had consistent pricing across its stores.
Pharmacy type was not significantly associated with pricing for drug, dose, supply, or combinations, with Mann-Whitney z scores ranging from −0.03 (P = .97) to −1.58 (P = .12). Correlations of pricing with pharmacy zip code median annual income were not significant: ρ ranged from ±0.01 (P = .99) to ±0.09 (P = .25). Pricing did not vary significantly by state in a cluster analysis by zip code yielding clusters for Missouri (n = 128) and Illinois (n = 47) pharmacies: z scores ranged from −0.26 (P = .79) to −1.31 (P = .19). In addition, pricing did not vary by median annual income clusters corresponding to high (n = 27, $85 883), moderate (n = 79, $59 347), and lower income (n = 69, $31 032) regions: Kruskal-Wallis χ2 ranged from 0.04 (P = .98) to 0.98 (P = .61). Finally, zip code and annual income clusters representing Illinois moderate income (n = 47, $53 122), North St Louis lower income (n = 55, $31 032), South St Louis moderate income (n = 49, $60 626), and West St Louis higher income (n = 24, $89 250) did not demonstrate pricing differences: Kruskal-Wallis χ2 ranged from 0.25 (P = .97) to 2.45 (P = .49).
Median prices for commonly prescribed generic medications used to treat HF were high and variable in a large sample of retail pharmacies. The retail pharmacy—rather than drug dose, duration of therapy, pharmacy ownership, or pharmacy location—was the primary cost driver. Variability remained when all 3 HF drugs were bundled in a single purchase within a given pharmacy. Digoxin, the oldest cardiovascular medication available, was paradoxically the most expensive.
Previously, Erikson and Lin5 demonstrated that pricing of generic levothyroxine, but not branded atorvastatin, was dependent on zip code. Our findings suggest that location does not matter. However, patients in low-income areas, who may be at highest risk of readmissions for and complications from HF,6 may not have convenient access to multiple pharmacies with competitive pricing, increasing vulnerability to lapses in adherence.
We limited our sample to 3 drugs, excluding other generic drugs with guideline-derived indications for HF. Data on pharmacy volumes could not be obtained for the individual retail stores because sales figures are proprietary. Mail-order pharmacies were not included, by design, because most uninsured and underinsured patients do not have access to this option.
In conclusion, generic drugs for HF show wide variability in pricing at the retail pharmacy level. The precise reasons for this, and the implications for adherence and subsequent clinical outcomes, require further study from both scientific and policy standpoints.
Corresponding Author: Paul J. Hauptman, MD, Department of Medicine, Saint Louis University Hospital, 3635 Vista Ave, St Louis, MO 63110 (firstname.lastname@example.org).
Published Online: November 15, 2016. doi:10.1001/jamainternmed.2016.6955
Author Contributions: All authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Hauptman, Vidic.
Acquisition, analysis, or interpretation of data: All Authors.
Drafting of the manuscript: Hauptman, Vidic, Chibnall.
Critical revision of the manuscript for important intellectual content: Hauptman, Goff, Chibnall, Bleske.
Statistical analysis: Chibnall, Bleske.
Administrative, technical, or material support: Hauptman, Chibnall.
Figure design: Goff.
Conflict of Interest Disclosures: Dr Hauptman has no conflicts of interest relevant to the topic of this article, but he reports serving as a consultant (St Jude Medical, Sensible Medical, Relypsa, Amgen, Corvia), participating in speaker bureaus (Relypsa, Amgen, Otsuka), and receiving research grants (Alnylam, National Heart, Lung, and Blood Institute). No other disclosures are reported.
Previous Presentation: This article was presented at the American Heart Association Scientific Sessions 2016; November 15, 2016; New Orleans, Louisiana.
Create a personal account or sign in to: