Trends in overdose experiences and prevention behaviors among people who use opioids in Baltimore, MD, 2017-2019

Abstract

Background: Little is known about trends in overdose behaviors. This study explored non-fatal overdose and engagement in overdose prevention behaviors and compared these trends to city-wide overdose fatality rates from 2017 to 2019 in Baltimore, MD.

Methods: The analysis included people who used opioids (PWUO; N = 502) recruited through a community-based study. Enrollment date was used to categorize participants into annual quarters. Logistic regression models examined change in overdose experiences and prevention behaviors with time. Baltimore's fatal overdoses were also mapped over the study period to assess overlaps in trends.

Results: The majority of the sample were male(68 %), Black(61 %), reported past 6 months homelessness(56 %), and were on average 45 years old. Most had witnessed(61 %), and 28 % had personally experienced an overdose in the past 6 months. Witnessing overdose marginally increased(aβ = 0.182;p = 0.058) while experiencing overdose did not significantly change by enrollment quarter. Most participants had or had been prescribed naloxone(72 %), and one fifth(22 %) regularly carried naloxone, with both access to(aβ = 0,408;p = 0.002) and carrying naloxone(aβ = 0.302;p = 0.006) increasing over time. Overdose communication remained stable, with 63 % of participants reporting discussing overdose sometimes/often. Among participants who injected (n = 376), regularly injecting alone decreased(aβ=-0.207;p = 0.055), and reporting others often/always having naloxone with them when injecting increased over time(aβ = 0.573;p < 0.001).

Conclusions: Witnessed overdose marginally increased from 2017 to 2019, aligning with city trends of fatal overdose. Overdose prevention behaviors significantly increased over time. Despite reporting having naloxone or a naloxone prescription, most PWUO did not regularly carry naloxone, and many used alone. Social network diffusion interventions may be a strategy to promote normative overdose prevention behaviors.

Keywords: Harm reduction; Naloxone; Overdose prevention; People who use opioids.

Figure 1:. Overdose experience in study PWUO and Baltimore City overdose fatality rate per 100,000 residents, by quarter

^a Witnessed an overdose adjusted beta: 0.182; 95% Confidence Interval: −0.006, 0.371; p-value: 0.058 ^b Experienced an overdose adjusted beta: −0.003, 95% Confidence Interval: −0.205, 0.199; p-value: 0.977

Figure 2:. Overdose prevention behaviors among people who use opioids, by quarter

Note.“ Have naloxone when injecting” and “Inject alone” measures only include individuals who reported injection drug use (n = 376) a: Have or prescribed naloxone adjusted beta: 0.408; 95% Confidence Interval: 0.152, 0.664; p-value: 0.002 b: Talk about overdose adjusted beta: −0.038; 95% Confidence Interval: −0.225, 0.148; p-value: 0.688 c: Inject alone adjusted beta: −0.207; 95% Confidence Interval: −0.419, 0.005; p-value: 0.055 d: Regularly carry nalone adjusted beta: 0.304; 95% Confidence Interval: 0.068, 0.540; p-value: 0.012 e: Have nalone when injecting adjusted beta: 0.573; 95% Confidence Interval: 0.304, 0.841; p-value: <0.001