Corrigendum to “Extent and implications of omitted ties on network measures in a longitudinal social network survey of people who use drugs” [Drug Alcohol Depend. 238 (2022) 109554](S0376871622002915)(10.1016/j.drugalcdep.2022.109554)

The authors wish to correct an error in the point estimates and summary statistics for our manuscript “Extent and implications of omitted ties on network measures in a longitudinal social network survey of people who use drugs”. While conducting entity resolution (ER) for new waves of data collection, we discovered a coding error in the ER algorithm that impacted the network data used in the current manuscript. As soon as we discovered this error, we corrected the coding error, re-ran the ER algorithm, had interviewers review the new potential matches for confirmation, and then re-ran the analyses from the paper to determine the effect of the recoding on the results. The coding change affected only 19 of the 1033 ties reported by participants in the dataset. Specifically, 19 network ties previously thought to be to non-SNAP participants were confirmed to be ties to other SNAP participants and one tie changed in status from prompted to free-listed. While these ties were originally accounted for in the analysis of the complete set of reported relationships, the change-in-status of one tie changed the summary statistics and prevalence ratios reported inTables 2 and 3 (and the corresponding mentions in the text). Changes to the prevalence ratios were at the hundredths decimal place; thus, the direction, magnitude, and significance of the revised prevalence ratios are consistent with those originally reported. Similarly, the addition of 19 pre-prompt edges in the sociometric network resulted in minor changes to the local and global sociometric network measures reported inTable 4 and in the text but did not change any of the conclusions. For example, the statistics presented in Table 4 are altered slightly, and the corresponding changes to the 95% bootstrapped confidence intervals are minor. None of these changes resulted in a change in statistical significance or altered the overall interpretation of findings. In summary, for all analyses, the direction, magnitude, and significance of effects remained the same after amending the network data based on the corrected ER algorithm. Updated results for Tables 1–4 and Figure S1 are included below. The only change toTable 1 was in footnote i. Figures 1 and 2, and Table S1 were not impacted by the corrections. We also included updated paragraphs for section 3.2 and 3.3 of the manuscript that describe summaries of the local and global sociometric network measures. The authors would like to apologize for any inconvenience caused.