In this analysis of more than 97 million commercially insured individuals, investigators found that the COVID-19 pandemic induced a spending shock in 2020 and that health care spending did not recover to baseline until mid-2021.
Objectives: To describe trends in US health care spending in a large, national, and commercially insured population during the COVID-19 pandemic.
Study Design: Cross-sectional study of commercially insured members enrolled between May 1, 2018, and December 31, 2021.
Methods: The study utilized a population-based sample of continuously enrolled members in a geographically diverse federation of Blue Cross Blue Shield plans across the United States. Our sample excluded Medicare and Medicare Advantage beneficiaries. The COVID-19 exposure period was defined as 2020-2021; 2018-2019 were pre–COVID-19 years. We defined 4 post–COVID-19 periods: March 1 to April 30, 2020; May 1 to December 31, 2020; January 1 to March 31, 2021; and April 1 to December 31, 2021. The primary outcome was inflation-adjusted overall per-member per-month (PMPM) medical spending adjusted for age, sex, Elixhauser comorbidities, area-level racial composition, income, and education.
Results: Our sample included 97,319,130 individuals. Mean PMPM medical spending decreased from $370.92 in January-February 2020 to $281.00 in March-April 2020. Between May and December 2020, mean PMPM medical spending recovered to—but did not exceed—prepandemic levels. Mean PMPM medical spending stayed below prepandemic levels between January and March 2021, rose above prepandemic baselines between April and June 2021, and decreased below baseline between July and December 2021.
Conclusions: The COVID-19 pandemic induced a spending shock in 2020, and health care spending did not recover to near baseline until mid-2021, with some emerging evidence of pent-up demand. The observed spending below baseline through the end of 2021 will pose challenges to setting spending benchmarks for alternative payment and shared savings models.
Am J Manag Care. 2023;29(10):517-521. https://doi.org/10.37765/ajmc.2023.89440
Total health spending declined in the immediate post–COVID-19 period, returned to prepandemic levels in late 2020, and exceeded prepandemic levels.
The COVID-19 pandemic introduced major shifts in patterns of health care utilization and spending for the majority of the US population, even patients who did not have COVID-19. Due to the impact of stay-at-home orders and patients’ preferring to avoid discretionary outpatient visits, the early phase of the COVID-19 pandemic witnessed large declines in US health care utilization and spending.1-4 It is unclear whether and when health care utilization and spending during the latter phase of the pandemic returned to baseline trends, particularly because of recurring waves of COVID-19. Findings from prior studies based on Medicare data have suggested that national health expenditures may resume prepandemic trends only in 2024 and that projected growth may be slower in 2025-2030 due to the pandemic accelerating shifts away from inpatient and acute care settings.5 However, these projections were not based on full 2021 data, and it is unclear how they apply to the commercially insured population, which consists of 160 million Americans.6
Understanding patterns of health care spending among commercially insured beneficiaries into 2021 may inform expectations around pent-up demand for health care utilization. Furthermore, understanding spending in 2021 may facilitate commercial payers setting spending benchmarks for future years and may inform policies on reducing cost sharing, such as premium reductions or co-payment/coinsurance waivers that were present in the early phase of the pandemic, and future premiums.7 Herein, we describe trends in US health care spending for a large, national commercially insured population from 2018 to 2021.
In this cross-sectional study, we analyzed paid professional, facility, and pharmacy claims between May 1, 2018, and December 31, 2021, from a federation of 35 separate Blue Cross Blue Shield plans that provide health insurance to more than 111 million members in all 50 states, the District of Columbia, and Puerto Rico. To characterize comorbidities, members with less than 6 months of continuous eligibility prior to the start of each study period (16.78%) were excluded. Medicare Advantage members and members enrolled in 2 specific commercial plans with incomplete data (7.47%) were excluded. Members without claims were included in this analysis. The University of Pennsylvania Institutional Review Board declared the study exempt with a waiver of consent. Our study followed the Strengthening the Reporting of Observational Studies in Epidemiology, or STROBE, guidelines.
Overall spending, inflation indexed using the medical care component of the US consumer price index to December 2020, was reported on a per-member per-month (PMPM) basis for each period to allow for monthly changes in enrollment. Spending was apportioned into inpatient, outpatient, laboratory, pharmacy, out-of-pocket, and other spending and analyzed separately. Medical spending (inpatient, outpatient, laboratory, and other) was calculated for all members. Pharmacy spending was calculated only among members with pharmacy benefits with the commercial insurer; individuals who opted out of pharmacy coverage or who obtained prescription drug benefits from a plan other than their medical plan were excluded from the analysis of pharmacy spend. Out-of-pocket spending was calculated by aggregating patient co-pay, coinsurance, and deductible amounts. The unit of analysis was the mean PMPM spending at the hospital service area (HSA) level (n = 3437) in each month.
We analyzed COVID-19–associated changes in calendar year 2020 PMPM spending using an interrupted time series analysis and ordinary least squares models, controlling for HSA-level means for age, sex, number of Elixhauser comorbidities8 (assessed in the 6 months prior to the index month), and area-level racial composition (% minority [non-White] race), income, and education. As a sensitivity analysis, we ran similar models using generalized estimating equations with an exchangeable correlation structure. Area-level covariates were defined at the zip code level and derived from the US Census Bureau American Community Survey. We also included a market fixed effect to control for market-level price differences and a month fixed effect to control for seasonality. The years 2018 and 2019 were considered pre–COVID-19, and 2020 and 2021 were the COVID-19 exposure years. To characterize immediate vs longer-term changes in spending, we defined 4 post–COVID-19 periods: March 1 to April 30, 2020 (immediate COVID-19, directly after most state stay-at-home orders); May 1 to December 31, 2020; January 1 to March 31, 2021; and April 1 to December 31, 2021. Statistical significance was defined as a 2-sided P < .05. The analyses were conducted using Stata SE 16.0 (StataCorp LLC).
Our sample included 97,319,130 individuals; 44,021,902 (45.2%) had pharmacy coverage through an employer, and of those, 42,519,361 individuals had pharmacy claims. There were no differences in demographic characteristics across time periods (Table 1). Mean PMPM medical spending decreased from $370.92 in January-February 2020 to $281.00 in March-April 2020 (adjusted difference-in-differences [aDID], –$147.10; 95% CI, –$151.21 to –$143.00). Between May and December 2020, mean PMPM medical spending was $383.52 and recovered to but did not exceed prepandemic levels (aDID, –$30.35; 95% CI, –$33.77 to –$26.92). Mean PMPM medical spending stayed below prepandemic levels between January and March 2021 ($368.87; aDID, –$12.40; 95% CI, –$15.54 to –$9.26), rose above prepandemic baselines between April and June 2021 ($430.18; aDID, $10.81; 95% CI, $6.96-$14.66), and decreased below baseline between July and December 2021 ($399.16; aDID, –$10.69; 95% CI, –$14.21 to –$7.17) (Figure [A]). Pharmacy spending remained consistent across the post–COVID-19 period (Figure [B]). Inpatient spending fell in April 2020 but recovered to baseline in the May to December 2020 period, whereas out-of-pocket spending remained below prepandemic baseline through April to June 2021 (Table 28). Outpatient spending fell in March to April 2020 (aDID, –$104.18; 95% CI, –$106.05 to –$102.32) and remained persistently below prepandemic baselines in the May to December 2020 (aDID, –$27.45; 95% CI, –$29.01 to –$25.90) and January to March 2021 (aDID, –$13.47; 95% CI, –$14.90 to –$12.05) periods, but it rose to above prepandemic baselines in April to June 2021 (aDID, $7.69; 95% CI, $5.94-$9.44) and then decreased to below baseline in July to December 2021 (aDID, –$9.57; 95% CI, –$11.17 to –$7.97). Results were consistent in the sensitivity analysis (eAppendix Table 1 [eAppendix available at ajmc.com]). Regression results are detailed in the eAppendix.
In this large US cohort of commercially insured members, total health spending declined in the immediate post–COVID-19 period, returned to prepandemic levels in the second half of 2020, exceeded prepandemic levels by mid-2021, but dipped below prepandemic baselines in the latter half of 2021. Lagging spending trends through the end of 2021 foreshadow challenges in setting future spending benchmarks for alternative payment and shared savings models because spending benchmark targets in alternative payment models are based on historical data, usually from the previous year. However, if the shock of COVID-19 led to lower-than-normal spending and if pent-up demand leads to higher spending in 2022, these spending targets may be difficult to achieve.
These findings have 3 implications. First, spending exceeded prepandemic levels only beginning in mid-2021. This may reflect pent-up demand, by which deferred care during the height of the pandemic increased later demand for health services as clinicians began reoffering deferred care and patients were more willing to return to clinics as vaccinations became available. However, given that spending dipped below baseline in the latter half of 2021, it is likely that some pent-up demand may still manifest in 2023 and beyond. Continued deferred services such as cancer screening may contribute to residual pent-up demand.9 The responsiveness of spending to pandemic changes, particularly that the spending can “overshoot” prepandemic levels due to care that may have been delayed, is notable. Still, the overshoot was modest in magnitude, which might give health plan administrators, policy makers, and payment model participants some reassurance. Second, compared with inpatient spending, outpatient spending declined more substantially during the COVID-19 pandemic period, and the relative magnitude of recovery was lower compared with inpatient and other spending segments, suggesting residual deferred outpatient and elective care.10,11 Third, out-of-pocket spending remained lower in 2020 into 2021. Potential reasons for this include payer policies that exempt telemedicine services from patient co-pays and the provision of premium relief or cost-sharing waivers to members that persisted into 2021.11,12
This study has several limitations. First, our population consisted of only commercially insured beneficiaries, although it encompassed a federation of many of the largest commercial plans in the country. Nonetheless, these results may not generalize to Medicare, Medicare Advantage, or Medicaid spending. Furthermore, because of natural variation in penetration of the commercial plans by certain geographic areas of the United States, the extent to which any given area contributes to the overall trend may not correspond to the actual commercial patient population in that area. Second, our data set contained only data going back to 2018; because we required a 6-month lookback window to assess comorbidities, we could not analyze data prior to May 2018. However, our use of a month fixed effect accounts for seasonality in spending.
The COVID-19 pandemic induced a spending shock in 2020, and health care spending did not recover to near baseline until mid-2021, with some emerging evidence of pent-up demand.
Author Affiliations: Healthcare Transformation Institute (RBP, EJE, ASN), Department of Medical Ethics and Health Policy (RBP, YZ, DRP, ASN), Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA; Blue Cross Blue Shield Association (SH, DAP), Chicago, IL.
Source of Funding: National Institute for Health Care Management Foundation (to ASN). The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, and approval of the manuscript; or the decision to submit the manuscript for publication.
Author Disclosures: Dr Emanuel has received grants from United Healthcare, Humana, and Hawaii Medical Service Association. Drs Hagen and Pizza are employed by Blue Cross Blue Shield Association. Dr Navathe received support for work on this article from the National Institute of Health Care Management; has received grants from Blue Cross Blue Shield of North Carolina, United Healthcare, Humana, and Hawaii Medical Service Association; and serves on the board for the SCAN Group. The remaining authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.
Authorship Information: Concept and design (RBP, EJE, YZ, SH, DAP, ASN); acquisition of data (RBP, EJE, SH, DAP, ASN); analysis and interpretation of data (RBP, EJE, YZ, DRP, SH, DAP, ASN); drafting of the manuscript (RBP, YZ, DRP); critical revision of the manuscript for important intellectual content (RBP, EJE, YZ, SH, DAP, ASN); statistical analysis (RBP, YZ, SH);administrative, technical, or logistic support (EJE, DRP, SH, DAP, ASN); and supervision (RBP).
Address Correspondence to: Ravi B. Parikh, MD, MPP, Perelman School of Medicine, University of Pennsylvania, 423 Guardian Dr, Blockley 1102, Philadelphia, PA 19104. Email: Ravi.Parikh@pennmedicine.upenn.edu.
1. Rosenbaum L. The untold toll — the pandemic’s effects on patients without COVID-19. N Engl J Med. 2020;382(24):2368-2371. doi:10.1056/NEJMms2009984
2. McWilliams JM, Russo A, Mehrotra A. Implications of early health care spending reductions for expected spending as the COVID-19 pandemic evolves. JAMA Intern Med. 2021;181(1):118-120. doi:10.1001/jamainternmed.2020.5333
3. Sen S, Karaca-Mandic P, Georgiou A. Association of stay-at-home orders with COVID-19 hospitalizations in 4 states. JAMA. 2020;323(24):2522-2524. doi:10.1001/jama.2020.9176
4. Fedewa SA, Star J, Bandi P, et al. Changes in cancer screening in the US during the COVID-19 pandemic. JAMA Netw Open. 2022;5(6):e2215490. doi:10.1001/jamanetworkopen.2022.15490
5. Poisal JA, Sisko AM, Cuckler GA, et al. National health expenditure projections, 2021-30: growth to moderate as COVID-19 impacts wane: study examines national health expenditure projections, 2021-30 and the impact of declining federal supplemental spending related to the COVID-19 pandemic. Health Aff (Millwood). 2022;41(4):474-486. doi:10.1377/hlthaff.2022.00113
6. Federal subsidies for health insurance coverage for people under 65: 2022 to 2032. Congressional Budget Office. June 30, 2022. Accessed October 4, 2023. https://www.cbo.gov/publication/57962
7. Ortaliza J, Rae M, Amin K, McGough M, Cox C. Most private insurers are no longer waiving cost-sharing for COVID-19 treatment. Peterson-KFF Health System Tracker. August 19, 2021. Accessed June 1, 2022.
8. Elixhauser A, Steiner C, Harris DR, Coffey RM. Comorbidity measures for use with administrative data. Med Care. 1998;36(1):8-27. doi:10.1097/00005650-199801000-00004
9. Moynihan R, Sanders S, Michaleff ZA, et al. Impact of COVID-19 pandemic on utilization of healthcare services: a systematic review. BMJ Open. 2021;11(3):e045343. doi:10.1136/bmjopen-2020-045343
10. Ziedan E, Simon KI, Wing C. Effects of state COVID-19 closure policy on non-COVID-19 health care utilization. National Bureau of Economic Research working paper 27621. July 2020. doi:10.3386/w27621
11. Patel SY, Mehrotra A, Huskamp HA, Uscher-Pines L, Ganguli I, Barnett ML. Trends in outpatient care delivery and telemedicine during the COVID-19 pandemic in the US. JAMA Intern Med. 2021;181(3):388-391. doi:10.1001/jamainternmed.2020.5928
12. McDermott D, Kurani N, Ramirez G, Shanosky N, Cox C. 2021 premium changes on ACA exchanges and the impact of COVID-19 on rates. KFF. October 19, 2020. Accessed June 13, 2021. https://www.kff.org/private-insurance/issue-brief/2021-premium-changes-on-aca-exchanges-and-the-impact-of-covid-19-on-rates