2017 values were imputed by (1) estimating the ratio of January-to-June values to January-to-December values in 2017 (per outcome) based on the 2006 to 2016 trend in this ratio and (2) multiplying the January-to-June costs from 2017 by the estimated 2017 ratio. Estimates were derived from an autoregressive model with intercept, trend, and change in trend2 terms, adjusting SEs for autocorrelation and removing nonsignificant terms. The eMethods section in the Supplement provides more details.
aCosts and prices were inflation-adjusted to 2017 US dollars according to the Consumer Price Index for US medical care.
bInsulin prices were imputed by calculating the median non-$0 OOP payment toward the annual deductible per 30-day equivalent insulin fill among members with a high-deductible health plan with a health savings account (HDHP-HSA). The percentage paid was the mean annual insulin OOP cost divided by the imputed price of insulin per given year. The eMethods section in the Supplement provides more details.
cOOP costs were derived by using 30-day equivalent insulin fills.
dOOP costs were calculated at the per member per month level.
eHDHP-HSA plans have federally regulated deductible minimums (≥$1050 to $13 100 over this period) and allow tax-exempt HSA contributions, but most services (including medications) are subject to a deductible.
fNo account and health reimbursement arrangement (HRA) are plans with any deductible amount but without any type of HSA.
eFigure. CONSORT Diagram
Customize your JAMA Network experience by selecting one or more topics from the list below.
Meiri A, Zhang F, Ross-Degnan D, Wharam JF. Trends in Insulin Out-of-Pocket Costs and Reimbursement Price Among US Patients With Private Health Insurance, 2006-2017. JAMA Intern Med. 2020;180(7):1010–1012. doi:10.1001/jamainternmed.2020.1302
In the United States, increasing insulin out-of-pocket costs1 have been a concern, and the potential for rationing2 has prompted states to consider capping these costs.3
Patients with a high-deductible health plan with a health savings account (HDHP-HSA) have high medication costs because these plans require payment of the full reimbursement price until the annual deductible is reached. In contrast, patients with a health reimbursement arrangement (HRA) or a no-account plan (without an HSA) typically have tiered drug copayments. Plan members with an HRA can use their reimbursement account to pay for medical expenses. In this cohort study, we examined insulin out-of-pocket costs and reimbursement price from 2006 to 2017 among patients with commercial health insurance.
We used Optum’s deidentified Clinformatics Data Mart Database4 to obtain medical and pharmacy claims data for privately insured members from all 50 states. We applied the HEDIS (Healthcare Effectiveness Data and Information Set) claims-based definition for diabetes5 and identified patients with diabetes who had insulin fills. The Harvard Pilgrim Health Care Institutional Review Board approved this study. Informed consent was waived by this institutional review board because the study used deidentified data.
We classified health insurance benefit types as HDHP-HSA, HRA, or no account. We imputed insulin reimbursement price per calendar year by calculating the median non-$0 deductible payment per 30-day equivalent fill of insulin prescriptions among members with an HDHP-HSA. These members pay the reimbursement price negotiated by the pharmacy benefit manager or insurer until they reach their deductible amount; this value approximates the reimbursement price while minimizing outlier price effects.
In addition, we calculated insulin out-of-pocket costs per 30-day fill and per member per month to characterize the actual monthly patient costs. Using an annual time-series design, we estimated the 2006 to 2017 inflation-adjusted (2017 US dollars) median reimbursement prices and mean out-of-pocket costs. We stratified out-of-pocket cost per member per month trends by health insurance benefit type. Because we had only 6 months of 2017 data, we imputed full-year measures on the basis of summarized January-to-June costs to summarized January-to-December costs from 2006 to 2016. The eMethods in the Supplement provides more detail.
All statistical analyses were performed with SAS, version 7.1 (SAS Institute Inc). Data were analyzed from January 1, 2006, to June 30, 2017.
Data from 612 071 unique patients with diabetes and 10 954 436 insulin claims were analyzed. Of these claims, 8 800 930 (80%) originated from members with a no-account plan, 442 662 (4%) from those with an HRA plan, and 774 903 (7%) from those with an HDHP-HSA. In addition, 935 941 (9%) lacked insurance benefit design data and were excluded from the stratified analysis (eFigure in the Supplement). The median insulin 30-day fill reimbursement price increased from $143 (95% CI, $119-$167) in 2006 to $394 (95% CI, $370-$418) in 2017 ($23/y increase; 95% CI, $19-$27). The mean out-of-pocket cost per 30-day insulin fill was $36 (95% CI, $33-$39) in 2006, $45 (95% CI, $43-$46) in 2012, and $38 (95% CI, $36-$41) in 2017. Patients’ share of insulin reimbursement price decreased from 24% (95% CI, 22%-26%) in 2006 to 10% (95% CI, 8.0%-11%) in 2017 (Figure, A and Table).
Among all insulin users, insulin out-of-pocket costs per member per month increased from $52 (95% CI, $48-$57) in 2006 to a peak of $72 (95% CI, $70-$75) in 2013, and then decreased to $64 (95% CI, $60-$68) in 2017. The out-of-pocket costs per member per month for those with an HDHP-HSA increased from $93 (95% CI, $82-$104) in 2006 to a peak of $150 (95% CI, $144-$155) in 2014, and then decreased to $141 (95% CI, $130-$152) in 2017 (Figure, B and Table).
Despite the increasing insulin prices paid by insurers, insulin out-of-pocket costs remained relatively flat from 2006 to 2017 among privately insured patients and began to decrease in 2013. Members with an HDHP-HSA paid substantially higher out-of-pocket amounts per month, which began decreasing in 2014.
This study has some limitations. First, we did not study uninsured or publicly insured patients or those with other private insurance but whose data were not in the Optum database. Second, the imputed insulin price did not account for manufacturer rebates. Third, the claims data used may have had inaccuracies.
Although the findings in this study contradict the view that insulin cost-sharing is excessive for many privately insured patients,2 monthly out-of-pocket payments may be burdensome for low-income individuals. Nevertheless, the data suggest that privately insured patients have been relatively shielded from insulin price increases and that commercial health insurers have accommodated higher insulin prices by increasing premiums or deductibles for all members.6
Accepted for Publication: March 23, 2020.
Corresponding Author: Amir Meiri, MD, MPH, Harvard Pilgrim Health Care Institute, Department of Population Medicine, Harvard Medical School, 401 Park St, Ste 401, Boston, MA 02215 (email@example.com).
Published Online: June 1, 2020. doi:10.1001/jamainternmed.2020.1302
Author Contributions: Drs Meiri and Ross-Degnan had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Meiri, Ross-Degnan, Wharam.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Meiri.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Meiri, Zhang, Wharam.
Obtained funding: Ross-Degnan, Wharam.
Administrative, technical, or material support: Wharam.
Supervision: Ross-Degnan, Wharam.
Conflict of Interest Disclosures: Drs Meiri, Zhang, Ross-Degnan, and Wharam reported receiving grants from the Centers for Disease Control and Prevention and the National Institute of Diabetes and Digestive and Kidney Diseases during the conduct of the study.
Funding/Support: This work was funded by grant U18DP006122 from the CDC/NIDDK and grant 1P30-DK092924 from the NIDDK Health Delivery Systems Center for Diabetes Translational Research (Drs Zhang, Ross-Degnan, and Wharam). Dr Meiri was supported by the General Internal Medicine Fellowship from the Harvard Medical School Department of Population Medicine.
Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.