Physiologic oxygen responses to smoking opioids: an observational study using continuous pulse oximetry at overdose prevention services in British Columbia, Canada

In British Columbia, Canada, smoking is the most common modality of drug use among people who die of opioid toxicity. We aimed to assess oxygen saturation (SpO ) while people smoked opioids during a pilot study that introduced continuous pulse oximetry at overdose prevention services (OPS) sites. This was an observational cohort study, using a participatory design. We implemented our monitoring protocol from March to August 2021 at four OPS. We included adults (≥ 18 years) presenting to smoke opioids. A sensor taped to participants' fingers transmitted real-time SpO readings to a remote monitor viewed by OPS staff. Peer researchers collected baseline data and observed the timing of participants' inhalations. We analyzed SpO on a per-event basis. In mixed-effects logistic regression models, drop in minimum SpO ≤ 90% in the current minute was our main outcome variable. Inhalation in that same minute was our main predictor. We also examined inhalation in the previous minute, cumulative inhalations, inhalation rate, demographics, co-morbidities, and substance use variables. We recorded 599 smoking events; 72.8% (436/599) had analyzable SpO data. Participants' mean age was 38.6 years (SD 11.3 years) and 73.1% were male. SpO was highly variable within and between individuals. Drop in SpO ≤ 90% was not significantly associated with inhalation in that same minute (OR: 1.2 [0.8-1.78], p = 0.261) or inhalation rate (OR 0.47 [0.20-1.10], p = 0.082). There was an association of SpO drop with six cumulative inhalations (OR 3.38 [1.04-11.03], p = 0.043); this was not maintained ≥ 7 inhalations. Demographics, co-morbidities, and drug use variables were non-contributory. Continuous pulse oximetry SpO monitoring is a safe adjunct to monitoring people who smoke opioids at OPS. Our data reflect challenges of real-world monitoring, indicating that greater supports are needed for frontline responders at OPS. Inconsistent association between inhalations and SpO suggests that complex factors (e.g., inhalation depth/duration, opioid tolerance, drug use setting) contribute to hypoxemia and overdose risk while people smoke opioids.