Feasibility Study of Using Mobile Phone-Based Experience Sampling to Assess Drug Checking by Opioid Street Drug Users

Abstract

Background: To date, evaluations of take-home fentanyl (and/or benzodiazepine) test strip use - the most common form of drug checking services - and potential effects on overdose risk have relied on retrospective accounts for some preceding time period, usually a week to several months. Such accounts, however, are subject to recall and memory biases. This pilot study assessed the feasibility of using experiential sampling to collect daily information in situ on drug checking and associated overdose risk reduction - the primary outcomes - among a sample of street opioid users and compared the results to retrospective reports. Methods: We recruited 12 participants from a Chicago-based syringe services program. Participants were 18 years of age or older, reported using opioids purchased on the street 3+ times per week in the past month, and had an available Android mobile phone. A phone-based app was programmed to collect daily drug checking information and provided to each participant along with a supply of fentanyl and benzodiazepine test strips and instructions for use over 21 days. Comparable retrospective data were collected via follow-up in-person surveys at the conclusion of daily report collection. Results: We found a reasonably high rate of daily reporting (63.5%) with participants submitting reports on 160 "person-days" out of 252 possible days. Participants submitted daily reports an average of 13 of 21 days. Reports of test strip use frequency varied between the retrospective and daily reports with a relatively higher percentage of days/time using test strips obtained from the daily reports. We also found higher proportions reporting overdose risk reduction behaviors on the daily reports compared with the retrospective reviews. Conclusions: We believe the results support using daily experience sampling to collect information on drug checking behaviors among street drug users. Although resource intensive in comparison to retrospective reports, daily reporting potentially provides more detailed information on test strip use and its association with overdose risk reduction and, ultimately, fewer overdoses. Needed are larger trials and validation studies of daily experience sampling to identify the optimum protocol for collecting accurate information on drug checking and overdose risk reduction behavior.

Figure 1

Total Number of Daily Drug Checking Reports Submitted by Participant and Study Day Note. All figures are based on the daily drug testing reports submitted over the 21-day study period by the 12 pilot test participants. The left panel shows the total number of unduplicated reports submitted by each participant ordered from the fewest to the most reports submitted. The right panel shows the number of unduplicated reports received each day during data collected across all participants. Multiple reports submitted on the same day were counted as one report for that day. The dashed line in the left panel shows the mean days reported and in the right panel shows the mean number of participants reporting at least once per day.

Figure 2

Flow Chart of Daily Test Strip Results from Submission to Risk Reduction Post-Testing Note. All figures are based on the number of valid daily reports submitted and/or the results obtained for drug checking at each reporting/testing step. All percentages shown are conditional, contingent on the number of reports submitted, the results of the submitted reports, or the behaviors taken given the obtained testing results. Because only a small number of tests (3) were conducted exclusively using the benzodiazepine test strips, the positive results and overdose risk behavior percentages are based on the reports of fentanyl test strip use.