Factors Associated With Drug Overdoses During the COVID-19 Pandemic.

To the Editor:

We read with interest your Journal's commentary describing the impact of the COVID-19 pandemic on access to needle and syringe programs[1] and recommendations for policies to improve the treatment of substance use disorders (SUDs) with telehealth.[2] Here, we consider these topics in the context of overdoses, treatment, and service access during the pandemic.

To address potentially catastrophic interactions between the overdose epidemic and COVID-19 pandemic, clinicians have rapidly provided alternative and additional services. However, preliminary evidence suggests overdoses may be increasing.[3] Between April 27 and May 13, 2020, the Addiction Policy Forum fielded to their US network of patients, families, and survivors of SUDs an IRB-approved anonymized survey (n = 1148 consenting respondents) assessing COVID-19-related impacts.[4] To identify possible factors linked to overdoses, we describe responses from individuals experiencing overdoses during the pandemic (Table 1).

TABLE 1

Demographic and COVID-19 Variables

	None (n = 1079)	Overdose (n = 47)	P, (OR)∗
Sex (%)			0.942
Female	677 (65.6)	28 (70.0)
Male	333 (32.3)	12 (30.0)
Other	22 (2.1)	0 (0.0)
Age (%)			0.334
18–25	43 (4.2)	3 (7.3)
26–40	344 (33.3)	16 (39.0)
41–60	467 (45.2)	18 (43.9)
61–64	90 (8.7)	1 (2.4)
65–74	80 (7.7)	2 (4.9)
75 yrs or older	9 (0.9)	1 (2.4)
Hispanic or Latino Ethnicity (%)	81 (7.9)	1 (2.3)	0.246 (0.28)
Race/Ethnicity (%)			0.176
American Indian/Alaskan Native	17 (1.7)	0 (0.0)
Asian	12 (1.2)	1 (2.3)
Black or African American	46 (4.5)	1 (2.3)
Native Hawaiian/Pacific Islander	1 (0.1)	1 (2.3)
Other	47 (4.6)	2 (4.7)
White	900 (88.0)	38 (88.4)
Education (%)			0.018
Less than high school	11 (1.1)	1 (2.3)
High school/equivalent	120 (11.6)	6 (14.0)
Some college, no degree	242 (23.5)	15 (34.9)
Associate degree	108 (10.5)	9 (20.9)
Bachelor's degree	283 (27.4)	6 (14.0)
Graduate or professional degree	267 (25.9)	6 (14.0)
Involved Substances
Polysubstance (%)	704 (66.0)	36 (76.6)	0.156 (1.68)
Alcohol (%)	700 (65.6)	30 (63.8)	0.876 (0.93)
Nicotine (%)	426 (39.9)	26 (55.3)	0.047 (1.86)
Stimulants (%)	450 (42.2)	25 (53.2)	0.174 (1.56)
Opioids (%)	492 (46.1)	34 (72.3)	<0.001 (3.05)
Sedatives (%)	228 (21.4)	11 (23.4)	0.718 (0.92)
Marijuana (%)	406 (38.1)	17 (36.2)	0.879 (1.12)
Other substances (%)	95 (8.9)	7 (14.9)	0.189 (1.79)
Personal Involvement
Family member (%)	423 (39.4)	25 (53.2)	0.068 (1.75)
In recovery (%)	582 (54.2)	18 (38.3)	0.037 (0.53)
Currently using (%)	111 (10.3)	10 (21.3)	0.028 (2.34)
In treatment (%)	80 (7.4)	4 (8.5)	0.775 (1.15)
COVID status (%)			0.030
Never tested, no symptoms	915 (85.0)	34 (72.3)
Never tested, yes symptoms	96 (8.9)	5 (10.6)
Tested and diagnosed	2 (0.2)	0 (0.0)
Tested negative	64 (5.9)	8 (17.0)
COVID impacts
Substance use has increased (%)	203 (19.0)	26 (60.5)	<0.001 (6.52)
Yes, COVID impacted services (%)	340 (33.1)	28 (66.7)	<0.001 (4.04)
Accessed telehealth (%)	188 (18.4)	11 (28.2)	0.141 (1.75)
Accessed more doses (%)	30 (2.9)	5 (12.8)	0.007 (4.85)
Accessed curbside medication (%)	43 (4.2)	5 (12.8)	0.027 (3.34)
Unable to access naloxone (%)	23 (2.2)	6 (15.4)	<0.001 (7.88)
Unable to access syringes (%)	18 (1.8)	6 (15.4)	<0.001 (10.1)
Unable to access needed services (%)	139 (13.6)	12 (30.8)	0.008 (2.82)

P value calculated by Fisher exact test due to small cell sizes. Odds ratios given for 2 × 2 tables.

Forty-seven individuals (4.17%) reported that they or their family members had experienced an overdose during the COVID-19 period. The most commonly reported educational attainment among individuals reporting an overdose was some college but no degree (34%). Fifty-five percent of individuals who overdosed reported using nicotine, 72% reported opioid use. Among those reporting an overdose, 53% identified as a family member of someone impacted by substance use, 38% identified as in recovery, and 21% reported current use. Regarding COVID-19 status, 72% reported they were never tested and had no symptoms or diagnosis.

Individuals who reported an overdose reported changes or disruptions in treatment (67%) and that substance use had increased due to the pandemic (61%). Some of these individuals reported access to greater take-home doses (13%) and curbside medication pickup (13%), but also reported inability to access naloxone (15%), needle exchange services (15%), or general needed services (31%).

Survey limitations include the small convenience sample and online self-report format which did not distinguish between individuals who had personally overdosed and those whose family members had overdosed. These findings may not indicate which individuals may have been at greater risk for overdose during this period, as we did not recruit matched cohorts of equivalent baseline overdose risks. Rather, they suggest some initial factors to explore in future research of the relationship between COVID-19 and overdoses. They suggest that educational attainment, a proxy for socioeconomic status, is linked to overdose during the COVID-19 period, as has previously been shown in analyses of overdoses before the COVID-19 pandemic.[5] The information suggests that disruptions in care and increased substance use are important to target to reduce likelihoods of overdoses. Importantly, naloxone and syringe exchange disruptions were more common among those who reported an overdose, as was usage of spatially distanced services such as curbside pick-up and extended take-home medications. This underscores the need to expand access to naloxone and other overdose reduction services and evaluate the efficacies of specific interventions as in-person interactions are reduced.