
==== Front
Prev Med Rep
Preventive Medicine Reports
2211-3355

S2211-3355(24)00247-X
10.1016/j.pmedr.2024.102832
102832
Infectious Disease
Association of chronic opioid therapy and opioid use disorder with COVID-19-related hospitalization and mortality: Evidence from three health systems in the United States
Nguyen Anh P. Anh.P.Nguyen@KP.org
a⁎
Binswanger Ingrid A. abcd
Narwaney Komal J. a
Ford Morgan A. a
McClure David L. e
Rinehart Deborah J. cf
Lyons Jason A. a
Glanz Jason M. ag
a Institute for Health Research, Kaiser Permanente Colorado, Aurora, CO, USA
b Colorado Permanente Medical Group, Denver, CO, USA
c Division of General Internal Medicine, University of Colorado School of Medicine, Aurora, CO, USA
d Bernard J. Tyson Kaiser Permanente School of Medicine, Pasadena, CA, USA
e Center for Clinical Epidemiology and Population Health, Marshfield Clinic Research Institute, Marshfield, WI, USA
f Center for Health Systems Research, Office of Research, Denver Health and Hospital Authority, Denver, CO, USA
g Department of Epidemiology, Colorado School of Public Health, Aurora, CO, USA
⁎ Corresponding author at: Institute for Health Research, Kaiser Permanente Colorado, P.O. Box 378066, Denver, CO 80237-8066, 303-636-2959, USA. Anh.P.Nguyen@KP.org
25 7 2024
10 2024
25 7 2024
46 10283210 5 2024
17 7 2024
18 7 2024
© 2024 The Author(s)
2024
https://creativecommons.org/licenses/by-nc-nd/4.0/ This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Objective

Chronic opioid use can lead to detrimental effects on the immune and various organ systems that put individuals prescribed chronic opioid therapy (COT) for pain and those with an opioid use disorder (OUD) at risk for severe COVID-19 disease. We assessed the association of COT and OUD with COVID-19-related hospitalization and death to inform targeted interventions to improve clinical outcomes in COVID-19 patients who use opioids.

Methods

We conducted a retrospective cohort study of adults ages ≥ 18 with laboratory-confirmed SARS-CoV-2 infection in 2020 and 2021 from three US health systems. We used Cox proportional hazards regression to estimate the 30-day risk of COVID-19-related hospitalization and death associated with two opioid exposures (COT and OUD) following an infection.

Results

The study cohort included 53,123 patients with SARS-CoV-2 infection and a mean (SD) age of 45.1 (16.5), of whom 1,059 (2.0 %) were exposed to COT and 269 (0.5 %) had an OUD diagnosis in the year prior to infection. There were 2,270 observed COVID-19-related hospitalizations or deaths (1.6 per 1,000 person-days, 95 % CI 1.5–1.7). In the fully adjusted model, COT was not associated with increased risk (HR 1.19; 95 % CI, 0.98–1.43), while past-year OUD was independently associated with severe COVID-19 disease (HR 1.82; 95 % CI, 1.18–2.80). Past-year OUD remained associated with increased risk in post-hoc analysis with COVID-19-related hospitalization alone as the outcome (HR 2.00; 95 % CI, 1.30–3.08).

Conclusions

Past-year OUD is a potential independent risk factor for severe COVID-19 disease that warrants monitoring to improve the prognosis of patients with COVID-19.

Keywords

Chronic opioid therapy
Opioid use disorder
COVID-19
Health services
Risk prevention
==== Body
pmc1 Introduction

The coronavirus disease 2019 (COVID-19) pandemic has in resulted in high morbidity and mortality in the United States, with nearly 7 million hospitalizations and more than 1 million deaths reported to date (Centers for Disease Control and Prevention, 2024). Numerous risk factors for severe COVID-19 disease, caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, have been identified, including demographic factors (such as older age), obesity, and cardiovascular diseases (Booth et al., 2021, Wolff et al., 2021, Zhang et al., 2023). There are concerns that individuals with chronic opioid use – those with exposure to chronic opioid therapy (COT) for pain and those with an opioid use disorder (OUD) – may face unique risks for severe complications from COVID-19 (Volkow, 2020, Schimmel and Manini, 2020). Opioids have been linked to immunosuppression in experimental studies (Roy et al., 2011, Vallejo et al., 2004, Eisenstein, 2019), and epidemiological research suggests that individuals on COT and those with OUD are at risk for compromised immunity as a result of chronic opioid use, leaving them susceptible to infections and severe illnesses (Roy et al., 2011, Edelman et al., 2019, Wiese et al., 2018, Gomes et al., 2022). Individuals with substance use disorders, including OUD, have been shown to have increased risk for contracting a COVID-19 infection (Wang et al., 2020, Wang et al., 2022). Further, patients prescribed COT and with an OUD have high prevalence of established risk factors, which may be linked to toxicities of chronic opioid use (Radke et al., 2014), that could lead to elevated morbidity and mortality from COVID-19 (Tuan et al., 2021, Schieber et al., 2023, Melamed et al., 2020). Current evidence, however, on the risk of hospitalization and death from COVID-19 among individuals prescribed COT (Tuan et al., 2021) and those with OUD (Qeadan et al., 2021, Krawczyk et al., 2023, Allen et al., 2023) is limited. Understanding COVID-19 risks associated with these opioid exposures is important to inform current practices, such as clinical monitoring and vaccination recommendations, to improve the prognosis of COVID-19 patients who use opioids.

In this study, we evaluated the association of opioid exposures and risk of COVID-19 related hospitalization and death. We used data on a large cohort of adult patients with a COVID-19 infection from three health systems that included patient demographic characteristics, medical comorbidities, and COVID-19 vaccination status. We examined COT and OUD as separate opioid exposures. Our strategy to assess both COT and OUD provided analytic leverage to consider potential similar and different drivers of severe COVID-19 disease in these two opioid-exposed populations.

2 Methods

2.1 Study design and setting

We conducted a retrospective cohort study with data from three health systems: Kaiser Permanente Colorado (KPCO), an integrated health plan and care delivery system that serves members in urban and suburban regions of Colorado; Denver Health (DH), an urban safety-net health system; and Marshfield Clinic Health System (MCHS), an integrated health system that serves largely rural populations across Wisconsin. Data for the study were extracted from a secure virtual data warehouse (VDW) at each health system that collects comprehensive patient and clinical data using a common data model (Ross et al., 2014). Data in the VDWs are derived from electronic health records and claims for external care services. These data include patient demographic information, social history, diagnoses (International Classification of Diseases, Tenth Revision [ICD-10]), care utilization, laboratory results, and pharmacy dispensation (National Drug Codes). Vital status and cause-of-death information were ascertained through a data linkage with the National Death Index. The data-only study was approved by the Kaiser Permanente Interregional Institutional Review Board (IRB) with a waiver of informed consent and with IRB ceding from the other health systems. The study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

2.2 Study population

The primary study cohort consisted of adult patients ages ≥ 18 with SARS-CoV-2 infection confirmed by a laboratory-based test between March 2020 and December 2021. The first observed date of a positive test result represented the index infection date. We required patients to have at least 12 months of health plan enrollment (at KPCO and MCHS) or empanelment (at least 1 primary care visit over a period of 18 months at DH) prior to the index date to allow accurate capture of patient characteristics. Follow-up began after the index date and continued until one of the following censoring events, whichever occurred first: a) the outcome of interest (COVID-19-related hospitalization or death), b) death from other causes, c) health plan disenrollment (at KPCO and MCHS) or health system dis-empanelment (at DH), d) 30 days after the index date, or e) end of the study period on December 31, 2021. The minimal person-time individuals can contribute to the analysis was 1 day, and the maximum was 30 days.

2.3 Measures

2.3.1 Study outcome

The outcome of interest was severe COVID-19 disease, identified by a composite measure of hospitalization with an accompanying discharge diagnosis of COVID-19 (ICD-10 U07.1) or death with an underlying or contributing cause of COVID-19. We assessed the outcome in the 30 days after the index date to capture events proximal to the infection.

2.3.2 Exposures

We assessed two separate opioid exposures: COT and OUD. COT was defined as three or more opioid analgesic prescriptions dispensed in a 90-day period with 80 days of medication coverage in the 12 months before the index date. We also identified an OUD diagnosis (ICD-10 F11, with or without evidence of OUD treatment) on or in the 12 months before the index date. Individuals could have both COT and OUD.

2.3.3 Covariates

We assessed patient and clinical characteristics previously shown to be associated with COVID-19 hospitalization and death (Booth et al., 2021, Wolff et al., 2021, Zhang et al., 2023, Yek, 2021). Covariates included age, race or ethnicity (Non-Hispanic Asian/Pacific Islander, Non-Hispanic Black, Hispanic of any race, Non-Hispanic White, or other [including more than one race]), sex, and key comorbidities identified via diagnoses: cancer, chronic cardiac disease (coronary artery disease, health failure and cardiomyopathies, and valvular heart disease), chronic kidney disease, chronic liver disease (cirrhosis, non-alcohol fatty liver disease, alcoholic liver disease, and autoimmune hepatitis), chronic neurologic disease (dementia, cerebrovascular disease, myopathies, and paralytic syndromes), chronic obstructive pulmonary disease, immunosuppression (HIV/AIDS, immunodeficiency, solid organ transplant, receipt of chemotherapy or other immunosuppressive therapy), diabetes, and obesity (Yek, 2021). We also assessed COVID-19 vaccination (≥1 dose), alcohol use disorder, tobacco use or use disorder (diagnosis or self-reported social history), and mental health disorders (anxiety and stress-related, mood, personality, and psychotic disorders) (Zhang et al., 2023, Wang et al., 2020, Schieber et al., 2023). Demographic characteristics were assessed on the index date, and clinical characteristics were assessed in the 12 months prior to the index date.

2.4 Statistical analysis

We first calculated crude rates and Kaplan-Meier survival estimates of the composite outcome by each opioid exposure (COT and OUD). We preformed log-rank tests to evaluate the equality of survival curves. To assess the association of opioid-related exposures and time to COVID-19-related hospitalization and death, we fit Cox proportional hazards models to estimate hazards ratios (HRs) and 95 % confidence intervals. To evaluate how covariates influenced associations, we tested two models: a minimally adjusted model that conditioned on demographic characteristics and separately a fully adjusted model that included demographic and clinical characteristics. Minimally and fully adjusted models also included an indicator for study health system. We calculated correlation of coefficients to assess for multicollinearity. Multiple imputation by chained equations was used to fill in missing values of race or ethnicity (4.3 % of the cohort) and sex (0.1 % of the cohort) (White et al., 2011). We verified the proportional hazards assumption using a global nonzero slope test and graphical inspections of Schoenfeld residuals. In post-hoc analysis, we estimated a model with COVID-19-related hospitalization only as the outcome and censoring at death of any cause. To assess the relative magnitude of associated risk for opioid exposures compared to other risk factors, we estimated individual models for each risk factor while adjusting for age, race and ethnicity, sex, and study site. We also evaluated a model with opioid dose for COT as an exposure. To assess the robustness of results, we conducted a series of sensitivity analyses. In separate models, we excluded individuals who were vaccinated for COVID-19 during the 30-day follow-up, excluded individuals who had both COT and OUD, defined opioid-related exposures 6 months prior to the index date, assessed a 14-day risk window, and treated death from causes other than COVID-19 related (N=8) as a competing risk. Further, we calculated E-values for the fully adjusted model to assess sensitivity to potential unmeasured confounding. The E-value is the minimum strength of association between an unmeasured confounder and both exposure and outcome needed to explain away the observed exposure-outcome association (VanderWeele and Ding, 2017). Statistical analyses were performed using Stata version 18.0 (StataCorp). Statistical significance was based on 2-sided tests with a threshold of p < 0.05.

3 Results

3.1 Characteristics of the study cohort with COVID-19 infection

The study consisted of 53,123 adult patients with laboratory-confirmed SARS-CoV-2 infection who had at least 1 day of follow-up after the index date across the three health systems. The mean (SD) age for the cohort was 45.1 (16.5) years (Table 1). The most common physical comorbidities in the cohort were obesity (16.9 %), chronic obstructive pulmonary disease (12.3 %), and diabetes (11.2 %). At the time of the index date, approximately a quarter (25.7 %) of individuals in the cohort were vaccinated with at least one dose for COVID-19. Patient with a COVID-19-related hospitalization or death within 30 days of the index date were generally older, had a heavier burden of comorbidities, and were less likely to be vaccinated for COVID-19.Table 1 Demographic and Clinical Characteristics of Adults with COVID-19 Infections from Three US Health Systems 2020–2021.

	Full Cohort (N=53,123)		Patients who did not experience a COVID-19- related hospitalization or death (N=50,853)		Patients who experienced a COVID-19- related hospitalization or death (N=2,270)		p-value	
	No. (%)		No. (%)		No. (%)			
Opioid Exposures								
Chronic Opioid Therapy	1,059 (2.0)		933 (1.8)		126 (5.6)		<0.001	
Opioid Use Disorder	269 (0.5)		247 (0.5)		22 (1.0)		0.001	
								
Demographic Characteristics								
Age, mean (SD)	45.1 (16.5)		44.4 (16.2)		60.20 (16.4)		<0.001	
18–29	10,844 (20.4)		10,741 (21.1)		103 (4.5)			
30–39	10,855 (20.4)		10,672 (21.0)		183 (8.1)			
40–49	10,550 (19.9)		10,289 (20.2)		261 (11.5)			
50–59	9,673 (18.2)		9.194 (18.1)		479 (21.1)			
≥60	11,201 (21.1)		9,957 (19.6)		1,244 (54.8)			
Race and Ethnicity							<0.001	
White	33,379 (62.8)		31,936 (62.8)		1,443 (63.6)			
Hispanic of any race	12,251 (23.1)		11,689 (23.0)		562 (24.8)			
Black	2,403 (4.5)		2,308 (4.5)		95 (4.2)			
Asian/Pacific Islander	1,470 (2.8)		1,408 (2.8)		62 (2.7)			
Other	1,331 (2.5)		1,276 (2.5)		55 (2.4)			
Missing	2,289 (4.3)		2,236 (4.4)		53 (2.3)			
Male	23,301 (43.9)		22,113 (43.5)		1,188 (52.3)		<0.001	
								
Substance Use and Mental Health								
Alcohol Use Disorder	968 (1.8)		925 (1.8)		43 (1.9)		0.793	
Tobacco Use or Use Disorder	13,379 (25.2)		12,554 (24.7)		825 (36.3)		<0.001	
Mental Health Disorder	12,908 (24.3)		12,315 (24.2)		593 (26.1)		0.038	
								
Medical History								
Obesity	8,998 (16.9)		8,341 (16.4)		657 (28.9)		<0.001	
Diabetes	5,932 (11.2)		5,201 (10.2)		731 (32.2)		<0.001	
Cancer	1,701 (3.2)		1,500 (3.0)		201 (8.9)		<0.001	
Chronic Cardiac Disease	4,876 (9.2)		4,239 (8.3)		637 (28.1)		<0.001	
Chronic Kidney Disease	2,669 (5.0)		2,208 (4.3)		461 (20.3)		<0.001	
Chronic Liver Disease	1,127 (2.1)		1,028 (2.0)		99 (4.4)		<0.001	
Chronic Neurologic Disease	1,028 (1.9)		841 (1.7)		187 (8.2)		<0.001	
Chronic Obstructive Pulmonary
Disease	6,534 (12.3)		5,987 (11.8)		547 (24.1)		<0.001	
Immunosuppression	684 (1.3)		597 (1.2)		87 (3.8)		<0.001	
COVID-19 Vaccination	13,675 (25.7)		13,403 (26.4)		272 (12.0)		<0.001	
Note: P-values were obtained from chi-square tests that compared patients who did and did not experience the outcome; age categories were compared between these two groups.

3.2 Chronic opioid therapy and opioid use disorder

The cohort included 1,059 (2.0 %) patients exposed to COT and 269 (0.5 %) patients with past-year OUD (Table 1); of these patients, 62 had both COT and OUD. Patients with COT were older (mean [SD] age of 58.5 [13.3]), more likely to have comorbidities, and had a higher level of COVID-19 vaccination compared to COVID-19 patients without COT (Supplemental Table A.1). Patients with past-year OUD were slightly younger (mean [SD] age of 43.7 [15.3]) and more likely to have some comorbidities (e.g., chronic obstructive pulmonary disease, chronic liver disease, and mental health disorder) than COVID-19 patients without past-year OUD (Supplemental Table A.2). Patients with and without past-year OUD did not differ in the level of vaccination for COVID-19.

3.3 Incidence of COVID-19-related hospitalization or death

There were 2,270 COVID-19-related hospitalizations and deaths within 30 days of the index date (1.6 per 1,000 person-days, 95 % CI 1.5–1.7), of which 223 were deaths (Table 2). The incidence of severe COVID-19 disease for patients prescribed COT (4.7 per 1,000, 95 % CI 3.9–5.6) was higher than for individuals not prescribed COT (1.5 per 1,000, 95 % CI 1.5–1.6). Similarly, the incidence was higher in individuals with past-year OUD (3.1 per 1,000, 95 % CI 2.1–4.8) compared to individuals without an OUD (1.6 per 1,000, 95 % CI 1.5–1.7). Further, Kaplan-Meier estimates showed significantly lower survival probabilities for individuals with COT compared to no COT, and separately, OUD compared to no OUD (Fig. 1).Table 2 Association of Opioid Exposures and COVID-19 Related Hospitalization and Death among Adults with COVID-19 Infections from Three US Health Systems 2020–2021.

Opioid Exposure	Person Days	Events	Crude Rate per 1,000 Person-Days (95 % CI)	Minimally Adjusted HR (95 % CI)	Fully Adjusted HR (95 % CI)	
Chronic Opioid Therapy				
No	1,397,059	2,144	1.5 (1.5–1.6)	Reference	Reference	
Yes	26,865	126	4.7 (3.9–5.6)	1.74 (1.45–2.09)	1.19 (0.98–1.43)	
Opioid Use Disorder					
No	1,416,932	2,248	1.6 (1.5–1.7)	Reference	Reference	
Yes	6,992	22	3.1 (2.1–4.8)	1.81 (1.18–2.78)	1.82 (1.18–2.80)	
Overall	1,423,924	2,270	1.6 (1.5–1.7)			
Abbreviation: HR=hazards ratio; CI=confidence interval.

Note: Minimally adjusted HRs from Cox proportional hazards regression model that included age, race and ethnicity, sex, and health system site; fully adjusted HRs from Cox proportional hazards regression model that included age, race and ethnicity, sex, health system site, alcohol use disorder, tobacco use and use disorder, mental health disorder, obesity, diabetes, cancer, chronic cardiac disease, chronic kidney disease, chronic liver disease, chronic neurologic disease, chronic obstructive pulmonary disease, immunosuppression, and COVID-19 vaccination status.

Fig. 1 Kaplan-Meier Estimates of 30-Day Survival of Adults with COVID-19 Infections by Opioid Exposures from Three US Health Systems 2020–2021.

3.4 Association of opioid exposures and COVID-19 related hospitalization or death

In the minimally adjusted model that included only demographic characteristics, exposure to COT was positively associated with COVID-19-related hospitalization and death (HR, 1.74; 95 % CI, 1.45–2.09) as was past-year OUD (HR, 1.81; 95 % CI, 1.18–2.78) (Table 2). In the fully adjusted model with the addition of clinical characteristics, the associated risk for COT was attenuated and no longer significant (HR 1.19; 95 % CI, 0.98–1.43), while past-year OUD remained independently associated with severe COVID-19 disease (HR 1.82; 95 % CI, 1.18–2.80). From individual models for each risk factor that adjusted for demographic characteristics and health system site, the magnitude of the associated increase in risk for past-year OUD (HR 1.92; 95 % CI, 1.26–2.95) rivaled recognized risk factors of severe COVID-19 disease, including chronic cardiac disease (HR 1.91; 95 % CI, 1.73–2.11), chronic obstructive pulmonary disease (HR 1.86; 95 % CI, 1.69–2.05), and obesity (HR 1.72; 95 % CI, 1.57–1.89) (Supplemental Table A.3).

We observed no COVID-19-related deaths among individuals with past-year OUD. In post-hoc analysis with COVID-19-related hospitalization alone as the outcome, past-year OUD remained associated with increased risk (HR 2.00; 95 % CI, 1.30–3.08) (Supplemental Table A.4). Additionally, we did not observe a dose–response association for COT dose (HR 1.01; 95 %, 0.96–1.06) (Supplemental Table A.4).

Sensitivity analyses showed that results were robust to specifications of the study population, exposures, risk window, and modeling strategy, with estimates similar or larger in magnitude (though confidence intervals overlapped) (Supplemental Table A.5). We estimated an E-value (lower confidence limit) of 3.04 (1.64) for the association between OUD and COVID-19-related hospitalization and death from the fully adjusted model in Table 2, indicating that substantial unmeasured confounding would be needed to explain away the observed association.

4 Discussion

In this multisite cohort of adults with laboratory-confirmed SARS-CoV-2 infection, patients exposed to COT and patients with past-year OUD exhibited higher rates of severe COVID-19 disease compared to patients without these opioid exposures. After adjustment for patient demographic and clinical characteristics, past-year OUD remained independently associated with a nearly 2-fold increase in risk of severe COVID-19 disease. Results of this study suggest OUD is an important marker for elevated severe COVID-19 disease that should be evaluated alongside established risk factors to improve clinical outcomes in patients with COVID-19.

Our study contributes to existing research on opioid exposures and severe COVID-19 disease by assessing both COT and OUD and leveraging data on a large cohort from three health systems. Findings related to OUD are consistent with prior studies that used data from 2020 and early 2021, which showed that COVID-19 patients with substance use disorder generally and OUD specifically had higher risk of hospitalization (Wang et al., 2020, Qeadan et al., 2021, Krawczyk et al., 2023, Allen et al., 2023, Baillargeon et al., 2021). The evidence on mortality risk for patients with OUD is mixed (Qeadan et al., 2021, Krawczyk et al., 2023, Allen et al., 2023). Among patients with past-year OUD in our study, we did not observe any COVID-19-related deaths within 30 days following the index infection date. Research on risks associated with COT is more limited. A study analyzing data from 2020 found that COVID-19 patients on COT had increased risk of hospitalization, intensive care once admitted, and mortality (Tuan et al., 2021). While exposure to COT was positively associated with severe COVID-19 disease in the minimally adjusted model in our study, the associated risk was attenuated in the fully adjusted model. Differences in time periods (our study used data from 2020 and 2021) and covariates included in the data may explain why our findings depart from prior work.

Several pathways may explain increased disease severity in COVID-19 patients who use opioids. Experimental studies have shown that opioids suppress immune response by inhibiting the function of immune cells (such as natural killer cells, neutrophils, and macrophages), antibody production, and cytokine expression (Schimmel and Manini, 2020, Roy et al., 2011, Vallejo et al., 2004, Eisenstein, 2019). Compromised immunity increases vulnerability to infections and more severe illnesses (Roy et al., 2011, Edelman et al., 2019, Baillargeon et al., 2021, Wiese et al., 2016). Use of opioids (prescribed or non-prescribed) can also lead to pulmonary complications, such as respiratory depression and pulmonary edema, and toxicities in other organ systems that could increase risk for complications of COVID-19, including acute respiratory distress syndrome (Radke et al., 2014, Melamed et al., 2020, Qeadan et al., 2021, Baillargeon et al., 2021). Apart from opioid use itself, patients prescribed COT and with past-year OUD have higher prevalence of recognized risk factors for COVID-19 morbidity and mortality, such as diabetes, obesity, and cardiovascular diseases, that could contribute to observed increased risk (Wang et al., 2020, Tuan et al., 2021). The relationship between chronic opioid use and severe COVID-19 disease may differ between patients prescribed COT and patients with past-year OUD. As shown in our data, demographic characteristics and disease burden are markedly different between these two groups, which may influence variation in vulnerability to severe COVID-19. Additionally, differences in the duration, dose, and type of opioids used by each group could impact risk. Our finding that OUD is a potential independent risk factor for severe COVID-19 suggests non-biological mechanisms warrant consideration. For patients with OUD, stigma associated with opioid use and barriers to accessing health services can delay timely care, resulting in greater severity of COVID-19 complications and worse outcomes (Volkow, 2020, Melamed et al., 2020, McCradden et al., 2019). Further research is needed to evaluate these mechanisms in patients prescribed COT and with past-year OUD.

Our results show that older age and medical comorbidities may explain observed risk in patients exposed to COT. Comorbidities, such as immunosuppression and chronic obstructive pulmonary disease, may be independent of or related to a history of COT. If the latter, our estimates for COT may be biased toward the null if these conditions mediate the relationship between COT and severe COVID-19 disease. It is also possible that adjusting for risk factors, such as cancer and chronic neurologic disease, that may be indicated for COT treatment may have attenuated the association. We were not able to explicitly identify indications for COT and examine their impact on risk, which could be a direction for future work.

Strategies to address increased risk of severe COVID-19 disease in patients with opioid exposures are needed. Patients with COT are more likely to have risk factors that make them potential candidates for antiviral treatments (Centers for Disease Control and Prevention, 2020). Ensuring access to and uptake of these treatments could serve to reduce COVID-19 risks in patients with COT. The independent association of past-year OUD and severe disease emphasizes the importance of screening for OUD, particularly among patients with SARS-CoV-2 infection. Limited evidence also suggests that treatment for OUD may protect against COVID-19-related hospitalization and death (Qeadan et al., 2021). Addressing barriers to broader reach of effective treatments for OUD, such as stigma and insurance challenges (Olsen et al., 2021, Volkow, 2018, Nguyen et al., 2022), continues to be important and also could have positive spillover impacts on timely care engagement for comorbid health conditions, including COVID-19 complications. Interventions to prevent the spread of infections (such as enhanced infection control protocols) could also reduce the burden of COVID-19 morbidity and mortality among individuals with OUD, who are often exposed to congregate settings that increase the likelihood of disease transmission, such as residential treatment, shelters, jails, and prisons (Volkow, 2020). While our findings did not show that patients with OUD were less likely to be vaccinated for COVID-19 during the study period, they confirmed that vaccination conferred significant protection against severe disease. Thus, efforts to maintain current vaccination among individuals with OUD are critical. Finally, with expiration of the federal COVID-19 public health emergency, addressing COVID-19 risks among patients at high risk should remain a public health priority, particularly among patients with OUD who face challenges to accessing timely and quality care that contribute to poor outcomes (Krawczyk et al., 2023).

This study has limitations. We used diagnosis data to identify OUD, which may be under-diagnosed, leading to exposure misclassification. We included patients with laboratory-confirmed SARS-CoV-2 infection to ensure accurate identification of the population at risk but could have under-detected patients with infections who did not undergo laboratory testing. Rapid antigen tests became widely available after our study period (Peeling et al., 2022). We considered the index date as the date of a positive test result, which may differ from the date of infection. While we accounted for a robust set of covariates, residual confounding, such as due to unmeasured socioeconomic status, may bias results. However, the E-value calculated for OUD suggests that significant unmeasured confounding would be needed to negate the observed association with severe COVID-19 disease. Although our study included a large cohort from three health systems, we lacked sufficient power to evaluate heterogeneity within opioid exposure groups, such as clinical outcomes by treatment status among patients with OUD. Finally, our study was conducted using data from 2020 and 2021. Changes in COVID-19 disease severity over time could impact findings. Despite these limitations, our study leveraged data on a large cohort of patients from three health systems that included detailed clinical and mortality information.

5 Conclusion

Patients with COVID-19 who have opioid exposures, either COT or OUD, have elevated rates of severe COVID-19 disease compared to patients without these exposures. Findings suggest OUD is an independent risk factor for severe COVID-19 disease, after adjusting for patient demographic and clinical characteristics. Interventions to prevent and mitigate COVID-19 risks in this population, such as maintaining current COVID-19 vaccination, are needed.

6 Disclosures

Dr. Binswanger receives royalties for educational content on the health of incarcerated persons from UpToDate. Authors have disclosed no other conflicts of interest.

Funding

This study was supported by a grant from the National Institute on Drug Abuse of the National Institutes of Health under grant award number R01DA047537 . The content expressed in this manuscript is solely the responsibility of the authors and does not necessarily represent the views, position, or official policy of the National Institutes of Health.

CRediT authorship contribution statement

Anh P. Nguyen: Writing – review & editing, Writing – original draft, Investigation, Formal analysis, Conceptualization. Ingrid A. Binswanger: Writing – review & editing, Supervision, Funding acquisition, Conceptualization. Komal J. Narwaney: Writing – review & editing, Resources. Morgan A. Ford: Writing – review & editing, Project administration. David L. McClure: Writing – review & editing. Deborah J. Rinehart: Writing – review & editing. Jason A. Lyons: Writing – review & editing, Data curation. Jason M. Glanz: Writing – review & editing, Supervision, Funding acquisition, Conceptualization.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Dr. Binswanger receives royalties for educational content on the health of incarcerated persons from UpToDate. Authors have disclosed no other conflicts of interest.

Appendix A Supplementary data

The following are the Supplementary data to this article:Supplementary Data 1

Data availability

Data will be made available upon request with appropriate data use agreements and approvals.

Acknowledgment

We thank M. Joshua Durfee, MSPH and Sai Sudha Medabalimi, MS, MPharm for data collection, management, and quality assurance; and Melanie Stowell, MSc and Judith Hase, BS for project coordination and assistance with data collection.

Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.pmedr.2024.102832.
==== Refs
References

Allen B. Basaraba C. Corbeil T. Racial differences in COVID-19 severity associated with history of substance use disorders and overdose: findings from multi-site electronic health records in New York City Prev. Med. 172 2023 107533
Baillargeon J. Polychronopoulou E. Kuo Y.F. Raji M.A. The impact of substance use disorder on COVID-19 outcomes Psychiatr. Serv. 72 5 2021 578 581 10.1176/appi.ps.202000534 33138712
Booth A. Reed A.B. Ponzo S. Population risk factors for severe disease and mortality in COVID-19: a global systematic review and meta-analysis PLoS One 16 3 2021 e0247461 33661992
Centers for Disease Control and Prevention. COVID-19 and Your Health. COVID-19 Treatments and Medications. Published February 11, 2020. Accessed April 29, 2024. https://www.cdc.gov/coronavirus/2019-ncov/your-health/treatments-for-severe-illness.html.
Centers for Disease Control and Prevention. COVID Data Tracker. Published April 29, 2024. Accessed April 30, 2024. https://covid.cdc.gov/covid-data-tracker.
Edelman E.J. Gordon K.S. Crothers K. Association of prescribed opioids with increased risk of community-acquired pneumonia among patients with and without HIV JAMA Intern. Med. 179 3 2019 297 304 10.1001/jamainternmed.2018.6101 30615036
Eisenstein T.K. The role of opioid receptors in immune system function Front. Immunol. 2019 10 10.3389/fimmu.2019.02904 30723470
Gomes T. Kitchen S.A. Tailor L. Trends in hospitalizations for serious infections among people with opioid use disorder in Ontario, Canada J. Addict. Med. 16 4 2022 433 10.1097/ADM.0000000000000928 34711742
Krawczyk N. Rivera B.D. Basaraba C. COVID-19 complications among patients with opioid use disorder: a retrospective cohort study across five major NYC hospital systems Addiction 118 5 2023 857 869 10.1111/add.16105 36459420
McCradden M.D. Vasileva D. Orchanian-Cheff A. Buchman D.Z. Ambiguous identities of drugs and people: a scoping review of opioid-related stigma Int. J. Drug Policy 74 2019 205 215 10.1016/J.DRUGPO.2019.10.005 31671303
Melamed O.C. Hauck T.S. Buckley L. Selby P. Mulsant B.H. Article commentary: COVID-19 and persons with substance use disorders: inequities and mitigation strategies Subst Abuse. 41 3 2020 286 291 10.1080/08897077.2020.1784363
Nguyen A.P. Glanz J.M. Shoup J.A. Mortality among patients prescribed buprenorphine for opioid use disorder after disenrollment from an insurance plan and healthcare system J. Gen. Intern. Med. 37 11 2022 2882 2884 10.1007/S11606-021-07344-W 35083649
Olsen Y. Fitzgerald R.M. Wakeman S.E. Overcoming barriers to treatment of opioid use disorder J. Am. Med. Assoc. 325 12 2021 1149 1150 10.1001/jama.2021.1741
Peeling R.W. Heymann D.L. Teo Y.Y. Garcia P.J. Diagnostics for COVID-19: moving from pandemic response to control Lancet 399 10326 2022 757 768 10.1016/S0140-6736(21)02346-1 34942102
Qeadan F. Tingey B. Bern R. Opioid use disorder and health service utilization among COVID-19 patients in the US: a nationwide cohort from the Cerner Real-World Data EClinicalMedicine. 2021 100938 10.1016/j.eclinm.2021.100938
Radke J.B. Owen K.P. Sutter M.E. Ford J.B. Albertson T.E. The effects of opioids on the lung Clin. Rev. Allergy Immunol. 46 1 2014 54 64 10.1007/s12016-013-8373-z 23636734
Ross T.R. Ng D. Brown J.S. The HMO research network virtual data warehouse: a public data model to support collaboration EGEMS. 2 1 2014 1049 10.13063/2327-9214.1049 25848584
Roy S. Ninkovic J. Banerjee S. Opioid drug abuse and modulation of immune function: consequences in the susceptibility to opportunistic infections J. Neuroimmune Pharmacol. 6 4 2011 442 465 10.1007/s11481-011-9292-5 21789507
Schieber L.Z. Dunphy C. Schieber R.A. Lopes-Cardozo B. Moonesinghe R. Guy G.P. Jr. Hospitalization associated with comorbid psychiatric and substance use disorders among adults with COVID-19 treated in US Emergency Departments From April 2020 to August 2021 JAMA Psychiat. 80 4 2023 331 341 10.1001/jamapsychiatry.2022.5047
Schimmel J. Manini A.F. Opioid use disorder and COVID-19: biological plausibility for worsened outcomes Subst. Use Misuse 55 11 2020 1 2 10.1080/10826084.2020.1791184 31426693
Tuan W.J. Spotts H. Zgierska A.E. Lennon R.P. COVID-19 outcomes among adult patients treated with long-term opioid therapy for chronic non-cancer pain in the USA: a retrospective cohort study BMJ Open 11 11 2021 e056436
Vallejo R. de Leon-Casasola O. Benyamin R. Opioid therapy and immunosuppression Am. J.Ther. 11 5 2004 354 365 10.1097/01.mjt.0000132250.95650.85 15356431
VanderWeele T.J. Ding P. Sensitivity analysis in observational research: introducing the E-value Ann. Intern. Med. 167 4 2017 268 274 10.7326/M16-2607 28693043
Volkow N.D. Medications for opioid use disorder: bridging the gap in care 2018 Published online The Lancet
Volkow N.D. Collision of the COVID-19 and addiction epidemics Ann. Intern. Med. 2020 10.7326/m20-1212
Wang Q.Q. Kaelber D.C. Xu R. Volkow N.D. COVID-19 risk and outcomes in patients with substance use disorders: analyses from electronic health records in the United States Mol. Psychiatry. 1 2020 10.1038/s41380-020-00880-7
Wang L. Wang Q. Davis P.B. Volkow N.D. Xu R. Increased risk for COVID-19 breakthrough infection in fully vaccinated patients with substance use disorders in the United States between December 2020 and August 2021 World Psychiatry 21 1 2022 124 132 10.1002/wps.20921 34612005
White I.R. Royston P. Wood A.M. Multiple imputation using chained equations: issues and guidance for practice Stat. Med. 30 4 2011 377 399 10.1002/sim.4067 21225900
Wiese A.D. Griffin M.R. Stein C.M. Mitchel E.F. Grijalva C.G. Opioid analgesics and the risk of serious infections among patients with rheumatoid arthritis: a self-controlled case series study Arthritis Rheumatol. 68 2 2016 323 331 10.1002/art.39462 26473742
Wiese A.D. Griffin M.R. Schaffner W. Opioid analgesic use and risk for invasive pneumococcal diseases: a nested case-control study Ann. Intern. Med. 168 6 2018 396 404 10.7326/M17-1907 29435555
Wolff D. Nee S. Hickey N.S. Marschollek M. Risk factors for Covid-19 severity and fatality: a structured literature review Infection 49 1 2021 15 28 10.1007/s15010-020-01509-1 32860214
Yek C. Risk factors for severe COVID-19 outcomes among persons aged ≥18 years who completed a primary COVID-19 vaccination series — 465 health care facilities, United States, December 2020–October 2021 MMWR Morb. Mortal Wkly. Rep. 71 2022 10.15585/mmwr.mm7101a4
Zhang J.J. Dong X. Liu G.H. Gao Y.D. Risk and protective factors for COVID-19 morbidity, severity, and mortality Clin. Rev. Allergy Immunol. 64 1 2023 90 107 10.1007/s12016-022-08921-5 35044620
