Long-Term Health Care Costs Following COVID-19: Implications for Pandemic Preparedness

ABSTRACT

Objectives: To estimate excess health care costs in the 12 months following COVID-19 diagnosis.

Study Design: Retrospective cohort study using Blue Cross Blue Shield of Rhode Island claims incurred from January 1, 2019, to March 31, 2022, among commercial and Medicare Advantage members.

Methods: Difference-in-differences analyses were used to compare changes in health care spend between the 12 months before (baseline period) and the 12 months after (post period) COVID-19 diagnosis for COVID-19 cases and contemporaneous matched controls without COVID-19.

Results: Overall, there were 7224 commercial and 1630 Medicare Advantage members with a COVID-19 diagnosis on/before March 31, 2021, each with a matched control, yielding a sample of 14,448 commercial and 3260 Medicare Advantage members. Among commercial members, 51.9% were aged 25 to 54 years and 54.0% were female. Among Medicare Advantage members, 94.2% were 65 years or older and 62.0% were female. Among commercial members, from the baseline period to the post period, total health care spend increased $41.61 (7.7%) per member per month (PMPM) more among COVID-19 cases compared with their matched controls. Among Medicare Advantage members, the difference-in-differences was greater, with spend increasing $97.30 (13.1%) PMPM more among cases compared with controls. The difference-in-differences was greatest for outpatient and professional services (both populations) and prescription services (Medicare Advantage only).

Conclusions: COVID-19 diagnosis was associated with excess health care spend PMPM over the subsequent 12 months, highlighting the importance of societal preparations to support individuals’ long-term health care needs following COVID-19 and as a part of future pandemic preparedness.

Am J Manag Care. 2023;29(11):566-572. https://doi.org/10.37765/ajmc.2023.89452

Takeaway Points

The long-term health effects of COVID-19 are expected to have enormous costs. This study used a difference-in-differences analysis to estimate excess health care costs in the 12 months following COVID-19 diagnosis.

• Individuals with COVID-19 had a greater increase in health care spend in the 12 months following COVID-19 compared with controls, particularly for outpatient and professional services.
• It is essential that health care payers prepare for the costs and system capacity needed to support individuals’ long-term health care needs after COVID-19 and as a part of pandemic preparedness.
• Efforts to prevent COVID-19 and postacute sequelae of COVID-19, including through vaccination, remain critical.

Individuals with COVID-19 may experience adverse outcomes following acute infection. For example, 1 in 3 adults who had COVID-19 in the United States experienced new-onset or persistent symptoms after infection, known as postacute sequelae of COVID-19.^1-3 Individuals may also experience long-term effects of post–intensive care syndrome and/or major life disruptions during infection.^4-7

These long-term effects of COVID-19 are expected to have enormous medical costs.⁸ In 1 study, patients in the United States attended more health care visits and had greater health care costs in the 6 months following diagnosis than before diagnosis.⁹ However, this pre-/post design is limited by underlying temporal trends due to the pandemic (eg, health care–seeking behavior).^10-12 Longer follow-up is also needed to fully understand the long-term costs associated with COVID-19, including when costs may return to baseline levels.

The primary objective of this study was to estimate excess health care costs in the 12 months following COVID-19 by comparing change over time among individuals with diagnosed COVID-19 with similar contemporaneous controls. A secondary objective was to estimate excess diagnosis of key symptoms and conditions in the 12 months following COVID-19. We hypothesized that individuals with COVID-19 would have a greater increase in health care costs and key diagnoses than controls.

METHODS

Study Design, Sample, and Data

This was a retrospective cohort study using claims data to conduct a difference-in-differences analysis among Blue Cross Blue Shield of Rhode Island (BCBSRI) insurance plan members. The study included commercial and Medicare Advantage members enrolled as of June 30, 2022, and considered claims incurred from January 1, 2019, to March 31, 2022, and paid through June 30, 2022. The study period was from January 1, 2019, through March 31, 2022. Changes in health care spend and key diagnoses between the 12 months before (baseline period) and 12 months after (post period) COVID-19 diagnosis were compared for COVID-19 cases and matched controls. The calendar month of COVID-19 diagnosis was considered a washout period of the acute COVID-19 disease and excluded from the baseline and post periods. Members included in this analysis were required to have at least 1 month of enrollment in each period. Analyses were completed by BCBSRI staff to inform internal decision-making and, therefore, did not meet criteria for human subjects research review.

Cases. Members with a COVID-19 diagnosis (International Statistical Classification of Diseases, Tenth Revision [ICD-10] code U07.1, B97.29, B34.2, or B97.21) anywhere on a medical claim on/before March 31, 2021, were identified as COVID-19 cases. For members with multiple COVID-19 diagnoses during the study period, the first COVID-19 diagnosis was used as the index date, and subsequent claims with a COVID-19 diagnosis were excluded from the post period.

Controls. Each COVID-19 case was matched to a similar control member who did not have a COVID-19 diagnosis through March 31, 2022, using propensity score matching. The COVID-19 diagnosis date of their matched case was considered the control’s index date. Members who did not have a COVID-19 diagnosis but had a post–COVID-19 condition diagnosis (ICD-10 code U09.9, B94.8, or Z86.16) were excluded.

Measures

Matching variables. Propensity score matching was based on patients’ sociodemographic characteristics, number and type of chronic conditions in 2019, and health care utilization and costs in 2019. Sociodemographic measures included age, sex assigned at birth, insurance plan type, Johns Hopkins ACG System¹³ resource utilization band in 2019 (0 to 5; higher bands indicate more frequent health service utilization), number of member months in 2019, and number of pharmacy member months in 2019. Chronic conditions were based on the ACG System¹³ and included attention-deficit/hyperactivity disorder, anxiety, asthma, bipolar disorder, cancer, congestive heart failure, chronic obstructive pulmonary disease, dementia, depression, diabetes, end-stage renal disease, HIV, high cholesterol, high blood pressure, ischemic heart disease, joint disease, low back pain, obesity, personality disorder, schizophrenia, and substance use disorder. The corresponding diagnosis code was required to be present on more than 1 claim in 2019. Health care utilization and cost measures were continuous and included spend allowed in 2019 for each of inpatient, outpatient, professional (claims paid to credentialed clinicians), pharmacy, high-cost (claim payments greater than $50,000), inpatient medical, emergency department, behavioral health, and office visit services.

Study outcomes. The study outcomes were change in (1) health care spend and (2) utilization of specific symptom/condition diagnosis codes in the post period compared with the baseline period. Health care spend was calculated overall and by area of spend (inpatient, outpatient, professional, and prescription health care services). High-cost claims were excluded because of extreme variability induced by these claims on small samples. Monthly additional spend per member per month (PMPM) was calculated as mean spend per case enrolled that month minus mean spend per control enrolled that month.

Utilization of specific symptom/condition ICD-10 codes as a primary diagnosis was measured per 1000 members. Symptoms/conditions evaluated included chest pain (ICD-10 code R07), cough (R05), dyspnea (R06), fatigue (R53.8), elevated blood pressure reading (R03.0), high cholesterol (E78.0), joint pain (M25.50), and thrombotic complication (I82.4, I82.6, M62.2, K55.0, I26, I63, or G45). These symptoms/conditions were selected based on common post–COVID-19 symptoms as of July 2021.¹⁴

Analyses

Analyses were conducted in SAS/STAT (SAS Institute) and Microsoft Excel (Microsoft). Propensity score matching was conducted using the PSMATCH procedure in SAS with the caliper set to 0.2, and a requirement of an exact match on specific chronic conditions.

Difference-in-differences analysis was used to compare change in health care spend and utilization of symptom/condition diagnosis codes among members with diagnosed COVID-19 to that among their matched counterparts between the baseline period and post period. Health care spend and diagnosis code utilization from the month of their diagnosis/index date was excluded from both periods. Monthly additional health care spend PMPM was also summarized to identify when most excess spend accrues after COVID-19 diagnosis. Analyses were stratified by insurance plan type.

Characteristics of cases with vs without suitable controls were compared to understand potential biases in the cases included in the matched analyses. Additionally, sensitivity analyses were conducted to understand the impact of 5 major assumptions. Our primary analyses were repeated, including high-cost claims, any claims with a COVID-19 diagnosis in the post period, and inpatient admissions claims with a COVID-19 diagnosis in the post period, as well as restricting the post period to the 4 to 12 months following the diagnosis/index date and restricting to continuously enrolled members.

RESULTS

As of June 30, 2022, there were 359,907 individuals enrolled in a BCBSRI insurance plan, including 122,090 commercial and 65,595 Medicare Advantage members. Of those, 20,213 commercial (16.6%) and 4683 Medicare Advantage (7.1%) members had a COVID-19 diagnosis on or before March 31, 2021. Suitable controls without a COVID-19 diagnosis were identified via propensity score matching for a subset of the members with COVID-19, including 7224 commercial (35.7%) and 1630 Medicare Advantage (34.8%) members, yielding a final study sample of 14,448 commercial (7224 cases, 7224 controls) and 3260 Medicare Advantage (1630 cases, 1630 controls) members. COVID-19 cases with and without suitable controls were similar (eAppendix Table 1) (eAppendix available at ajmc.com).

Characteristics of the Sample

Propensity score matching yielded a sample of COVID-19 cases and matched controls with similar characteristics (eAppendix Table 2). Overall, commercial members in the study sample had a broad age distribution, with just over half (51.9%) aged 25 to 54 years (Table 1). Fifty-four percent of commercial members were female. About 45.0% of commercial members had unknown race/ethnicity, whereas 43.8% were non-Hispanic White. The majority (73.2%) of commercial members had moderate health service utilization in 2019 (resource utilization band 2 or 3). Common underlying health conditions among commercial members were obesity (19.9%) and high cholesterol (18.2%).

By contrast, most Medicare Advantage members (94.2%) were 65 years or older and 62.0% were female. A smaller percentage (22.9%) of Medicare Advantage members had unknown race/ethnicity, and 65.2% were non-Hispanic White. Most Medicare Advantage members (82.6%) had moderate to high health service utilization in 2019 (resource utilization band 3 or 4), and common underlying health conditions were high cholesterol (69.3%) and high blood pressure (64.9%).

Outcomes

Health care spend. Among commercial members, the median (IQR) change in total health care spend from the 12-month baseline period to the 12-month post period was $18.23 (–$65.31 to $175.76) PMPM for COVID-19 cases and $0.00 (–$88.40 to $99.66) PMPM for their matched controls. Among Medicare Advantage members, median change in health care spend was $42.92 (–$93.84 to $305.86) PMPM for COVID-19 cases and $18.06 (–$104.37 to $194.84) PMPM for controls.

Among commercial members, total health care spend among COVID-19 cases was $114.78 greater PMPM in the post period compared with the baseline period (25.0% increase), whereas health care spend for their matched controls was only $73.17 greater PMPM in the post period compared with the baseline period (17.2% increase) (Table 2, Figure 1). Thus, from the baseline period to the post period, health care spend increased $41.61 (7.7%) PMPM more among COVID-19 cases compared with their matched controls. Among Medicare Advantage members, the difference-in-differences was greater, with total health care spend increasing $97.30 (13.1%) PMPM more among COVID-19 cases than their matched controls during the post period compared with the baseline period.

Monthly additional health care spend PMPM among COVID-19 cases compared with matched controls decreased somewhat throughout the 12 months after COVID-19 diagnosis (or the corresponding index date) among commercial and Medicare Advantage members, although there was substantial variability month to month (Table 3). Monthly additional health care spend PMPM did not decrease to $0 by the end of the 12-month post period in either population.

Among commercial members, increases in health care spend PMPM for inpatient ($5.39 difference-in-differences), outpatient ($22.65), and professional ($21.26) services during the post period compared with the baseline period were greater among COVID-19 cases than among matched controls (eAppendix Table 3). In contrast, COVID-19 cases experienced a lesser increase in health care spend for prescription services compared with matched controls (–$7.69). Among Medicare Advantage members, increases in health care spend for outpatient ($59.80), professional ($25.89), and prescription ($27.73) services during the post period compared with the baseline period were greater among COVID-19 cases than among matched controls, whereas they experienced a lesser increase in health care spend for inpatient services than controls (–$16.13).

Utilization of symptom/condition diagnosis codes. Among commercial members, from the baseline period to the post period, the difference-in-differences for utilization of certain diagnosis codes per 1000 members was greater among COVID-19 cases compared with their matched controls, including codes for dyspnea (93.7), chest pain (68.9), fatigue (30.8), high cholesterol (12.0), and high blood pressure (6.8) (Figure 2 and eAppendix Table 4). Utilization of codes for cough decreased similarly, utilization of codes for thrombotic complications increased similarly, and utilization of codes for joint pain remained relatively stable among COVID-19 cases and matched controls from the baseline period to the post period. Among Medicare Advantage members, from the baseline period to the post period, the difference-in-differences for utilization of diagnosis codes per 1000 members for dyspnea (80.9), thrombotic complications (69.7), and fatigue (28.2) were greater among COVID-19 cases compared with their matched controls. Utilization of diagnosis codes for cough decreased from baseline to the post period for both COVD-19 cases and controls, although it decreased to a lesser extent among cases. In contrast, utilization of codes for chest pain decreased similarly in both groups from the baseline period to the post period, whereas utilization of codes for high blood pressure and joint pain remained relatively stable in both groups and utilization of codes for high cholesterol increased in both groups, although to a greater extent among controls.

Sensitivity Analyses

Our findings were generally similar in sensitivity analyses (eAppendix Tables 5 and 6 and eAppendix Figures 1, 2, and 3). Of note, among commercial and Medicare Advantage members, many difference-in-differences for health care spend and diagnosis utilization were somewhat attenuated in analyses restricting the post period to the 4 to 12 months following the diagnosis/index date, consistent with the greater monthly additional spend in the months immediately following COVID-19.

DISCUSSION

In this study, commercial and Medicare Advantage BCBSRI members with diagnosed COVID-19 experienced a greater increase in health care spend in the 12 months post diagnosis compared with the 12 months before diagnosis than matched control members. Monthly additional health care spend was greatest in the months immediately following COVID-19 diagnosis, although excess monthly spend persisted through the 12-month post period in both populations. Excess spend in the 12 months after COVID-19 was most apparent for outpatient and professional services among commercial members and for outpatient, professional, and prescription services among Medicare Advantage members. Excess diagnosis code utilization in the 12 months after COVID-19 was greatest for dyspnea, chest pain, and fatigue among commercial members and for dyspnea, thrombotic complications, and fatigue among Medicare Advantage members.

Our overall findings are generally consistent with a previous study by Koumpias and colleagues that compared nonpulmonary health care utilization and costs in the 6 months before vs after COVID-19 diagnosis.⁹ They similarly identified an increase in health care costs following COVID-19, with the largest and most persistent excess costs among older adults. Both studies also found that excess costs decreased each month following COVID-19 without returning to baseline levels over 6 and 12 months, respectively. However, findings by health service type differed. In our study, excess costs following COVID-19 were most notable for outpatient and professional services, although older adults also had excess costs for prescription services; there was no evidence of excess costs for inpatient services without a COVID-19 diagnosis. In contrast, Koumpias and colleagues noted the sharpest rise in costs for inpatient visits in the 6 months following COVID-19, although those analyses were inclusive of the first 30 days following diagnosis and may be largely reflective of inpatient acute COVID-19 care. We excluded the month of COVID-19 diagnosis from our analyses, as well as subsequent claims with a COVID-19 diagnosis; therefore, our results are likely reflective of ongoing health care utilization unrelated to acute COVID-19.

In our study, the general population with diagnosed COVID-19 had $41.61 excess health care spend PMPM over the subsequent 12 months ($499.32 per year), whereas older adults had $97.30 excess spend PMPM ($1167.60 per year). It is important to consider what this magnitude of excess spend at the individual level might suggest at a population level. As of October 2022, approximately 50% of the roughly 204 million adults aged 18 to 64 years in the United States reported having had COVID-19,^2,15 and approximately 68% of cases in this age group were likely symptomatic.¹⁶ If each of those symptomatic cases yielded $499.32 excess health care spend over the subsequent year, that would total more than $34 billion. Among the roughly 54 million adults 65 years or older, approximately 34% reported having had COVID-19,2,15 with around 80% of cases symptomatic.¹⁶ This might be expected to yield another $17 billion in excess health care spend, if each symptomatic case experienced $1167.60 excess spend over 1 year. Although these crude calculations may overestimate excess health care spend, given that the COVID-19 diagnoses on medical claims may be more severe than the average symptomatic COVID-19 case, this still suggests that there are massive health care costs following acute COVID-19 at a population level.

Our findings highlight the importance of anticipating long-term health care needs in pandemic preparedness. Health care payers, policy makers, and other health system leaders can anticipate these persistent health care needs and prepare for the associated costs and required system capacity, including from network access and care management perspectives. Additionally, proactive steps to ensure equitable access to services for low-income and undocumented individuals, such as expansion of health insurance coverage, are essential. The present study provides estimates of the long-term cost categories and amounts that can inform these preparations to support individuals’ longer-term health care needs after COVID-19. Our methods can also be adapted for rapid inference in future pandemics.

Our findings also highlight the ongoing importance of preventing COVID-19, not only to reduce the short-term risk of morbidity and mortality but also to minimize longer-term health effects.⁸ Staying up-to-date on COVID-19 vaccinations reduces risk of infection and severe acute illness,¹⁷ as well as risk of persistent symptoms if infected.^18-20 Although some emerging evidence suggests that the risk of postacute sequelae of COVID-19 may be lower for Omicron compared with prior variants,^20,21 findings are mixed²² and uncertainty remains about the incidence and virulence of future strains. Continuation of measures to prevent COVID-19 transmission at the individual and population levels is essential for preventing morbidity and mortality and minimizing costs and demands on the health care system.

This study also sheds light on manifestations of postacute sequelae of COVID-19 that may frequently lead individuals to seek care. Excess utilization of diagnosis codes for dyspnea and fatigue in the year following COVID-19 was noted for the general and older adult populations, in addition to excess utilization of chest pain codes among the general population and thrombotic complication codes among older adults. These findings align with prior work suggesting that fatigue, dyspnea, and chest discomfort are the most common persistent symptoms after COVID-19.²³

Limitations

Our study had important limitations. Some members who had COVID-19 but did not seek clinical care may be included as controls. Additionally, patients’ baseline periods may include deferred care, which would inflate change from the baseline to post period. However, this should affect cases and controls similarly, so a difference-in-differences design remains appropriate. Patients’ baseline periods may also include acute COVID-19–related care that preceded diagnosis; our finding that utilization of diagnosis codes for cough decreased from the baseline to the post period is consistent with this possibility. Inclusion of diagnoses related to acute COVID-19 in the baseline period may have limited our ability to detect an increase in those diagnoses following COVID-19. Additionally, excluding the calendar month of COVID-19 diagnosis from the baseline and post periods led to variability in the exact days excluded relative to their diagnosis date. We matched cases and controls based on health status in 2019; however, their health may have changed prior to COVID-19 diagnosis. Finally, the disproportionate increase in health care spend observed among COVID-19 cases may be driven by a relatively small subset of patients, although the median change in health care spend was greater for COVID-19 cases than their matched controls in both populations. Nonetheless, our study was strengthened by its large, diverse population and study design.

CONCLUSIONS

Among BCBSRI members, COVID-19 was associated with excess health care spend over the subsequent year, particularly for outpatient and professional services, and excess spend was greatest in the months immediately following COVID-19 diagnosis. Our findings highlight the importance of societal preparations to support individuals’ long-term health care needs within pandemic preparedness efforts, including anticipation of long-term health care utilization and costs.

Author Affiliations: Department of Epidemiology, Brown University (LCC, LL, TZ, FLB), Providence, RI; Blue Cross Blue Shield of Rhode Island (AP, MarC, MatC), Providence, RI.

Source of Funding: This work was funded by Blue Cross Blue Shield of Rhode Island. Dr Chambers, Ms Lovgren, Ms Zandstra, and Dr Beaudoin were supported, in part, by funding from the Hassenfeld Family Foundation to the Brown University School of Public Health. Dr Chambers was supported, in part, by the National Institutes of Health (grant R25MH083620).

Author Disclosures: Dr Chambers, Ms Lovgren, Ms Zandstra, and Dr Beaudoin report having work funded by Hassenfeld Family Initiatives LLC. Mr Collins reports being employed by Blue Cross Blue Shield, which has an interest in identifying cost associated with post–COVID-19 syndrome.

Authorship Information: Concept and design (AP, MarC, FLB, MatC); acquisition of data (AP); analysis and interpretation of data (LCC, AP, MarC, LL, TZ, FLB, MatC); drafting of the manuscript (LCC); critical revision of the manuscript for important intellectual content (LCC, AP, MarC, LL, TZ, FLB, MatC); statistical analysis (AP); and administrative, technical, or logistic support (LL, TZ).

Address Correspondence to: Laura C. Chambers, PhD, MPH, Brown University Department of Epidemiology, Box G-S121-2, 121 S Main St, Providence, RI 02903. Email: laura_chambers@brown.edu.

REFERENCES

1. Post-COVID conditions: information for healthcare providers. CDC. Updated September 11, 2023. Accessed August 1, 2022. https://www.cdc.gov/coronavirus/2019-ncov/hcp/clinical-care/post-covid-conditions.html

2. Long COVID: household pulse survey. CDC. Updated August 16, 2023. Accessed August 1, 2022.
https://www.cdc.gov/nchs/covid19/pulse/long-covid.htm

3. A clinical case definition of post COVID-19 condition by a Delphi consensus, 6 October 2021. World Health Organization. October 6, 2021. Accessed August 1, 2022. https://www.who.int/publications/i/item/WHO-2019-nCoV-Post_COVID-19_condition-Clinical_case_definition-2021.1

4. Smith S, Rahman O. Postintensive Care Syndrome. In: StatPearls. StatPearls Publishing LLC; 2023.

5. Janiri D, Carfì A, Kotzalidis GD, et al; Gemelli Against COVID-19 Post-Acute Care Study Group. Posttraumatic stress disorder in patients after severe COVID-19 infection. JAMA Psychiatry. 2021;78(5):567-569. doi:10.1001/jamapsychiatry.2021.0109

6. Kyzar EJ, Purpura LJ, Shah J, Cantos A, Nordvig AS, Yin MT. Anxiety, depression, insomnia, and trauma-related symptoms following COVID-19 infection at long-term follow-up. Brain Behav Immun Health. 2021;16:100315. doi:10.1016/j.bbih.2021.100315

7. Mazza MG, De Lorenzo R, Conte C, et al; COVID-19 BioB Outpatient Clinic Study Group. Anxiety and depression in COVID-19 survivors: role of inflammatory and clinical predictors. Brain Behav Immun. 2020;89:594-600. doi:10.1016/j.bbi.2020.07.037

8. Cutler DM. The costs of long COVID. JAMA Health Forum. 2022;3(5):e221809. doi:10.1001/jamahealthforum.2022.1809

9. Koumpias AM, Schwartzman D, Fleming O. Long-haul COVID: healthcare utilization and medical expenditures 6 months post-diagnosis. BMC Health Serv Res. 2022;22(1):1010. doi:10.1186/s12913-022-08387-3

10. Kim Y, Gordon A, Rowerdink K, Herrera Scott L, Chi W. The impact of the COVID-19 pandemic on health care utilization among insured individuals with common chronic conditions. Med Care. 2022;60(9):673-679. doi:10.1097/MLR.0000000000001747

11. Czeisler MÉ, Marynak K, Clarke KEN, et al. Delay or avoidance of medical care because of COVID-19-related concerns - United States, June 2020. MMWR Morb Mortal Wkly Rep. 2020;69(36):1250-1257. doi:10.15585/mmwr.mm6936a4

12. Whaley CM, Pera MF, Cantor J, et al. Changes in health services use among commercially insured US populations during the COVID-19 pandemic. JAMA Netw Open. 2020;3(11):e2024984. doi:10.1001/jamanetworkopen.2020.24984

13. Johns Hopkins ACG System (Version 11.1). Johns Hopkins University. Accessed August 1, 2022. https://www.hopkinsacg.org

14. Caring for people with post-COVID conditions. CDC. Updated September 1, 2022. Accessed July 1, 2021. https://www.cdc.gov/coronavirus/2019-ncov/long-term-effects/care-post-covid.html

15. Total U.S. resident population by age, sex, and series: April 1, 2020 (in thousands). US Census Bureau. Accessed August 1, 2022. https://www.census.gov/data/tables/2020/demo/popest/2020-demographic-analysis-tables.html

16. Sah P, Fitzpatrick MC, Zimmer CF, et al. Asymptomatic SARS-CoV-2 infection: a systematic review and meta-analysis. Proc Natl Acad Sci U S A. 2021;118(34):e2109229118. doi:10.1073/pnas.2109229118

17. Vaccines and vaccination. CDC. Updated September 15, 2021. Accessed August 1, 2022. https://www.cdc.gov/coronavirus/2019-ncov/science/science-briefs/fully-vaccinated-people.html

18. The effectiveness of vaccination against long COVID: a rapid evidence briefing. UK Health Security Agency. 2022. Accessed August 1, 2022. https://www.icpcovid.com/sites/default/files/2022-02/Ep%20241-9%20UK%20Health%20Security%20Agency%20The%20effectiveness%20of%20vaccination%20against%20long%20COVID%20Feb%202022.pdf

19. Kuodi P, Gorelik Y, Zayyad H, et al. Association between BNT162b2 vaccination and reported incidence of post-COVID-19 symptoms: cross-sectional study 2020-21, Israel. NPJ Vaccines. 2022;7(1):101. doi:10.1038/s41541-022-00526-5

20. Azzolini E, Levi R, Sarti R, et al. Association between BNT162b2 vaccination and long COVID after infections not requiring hospitalization in health care workers. JAMA. 2022;328(7):676-678. doi:10.1001/jama.2022.11691

21. Antonelli M, Pujol JC, Spector TD, Ourselin S, Steves CJ. Risk of long COVID associated with delta versus omicron variants of SARS-CoV-2. Lancet. 2022;399(10343):2263-2264. doi:10.1016/S0140-6736(22)00941-2

22. Taquet M, Sillett R, Zhu L, et al. Neurological and psychiatric risk trajectories after SARS-CoV-2 infection: an analysis of 2-year retrospective cohort studies including 1 284 437 patients. Lancet Psychiatry. 2022; 9(10):815-827. doi:10.1016/S2215-0366(22)00260-7

23. Type, proportion, and duration of persistent COVID-19 symptoms. UpToDate.Accessed August 1, 2022. https://medilib.ir/uptodate/show/130356