Background: Nigeria faces a substantial burden of illicit drug use alongside intensified enforcement activity. However, the geographic correspondence between enforcement indicators and population-level drug use burden remains poorly characterised. This study provides an ecological assessment of enforcement–burden alignment in Nigeria and introduces a Total Seizure-to–Any-Drug-User Ratio (TSUR) as a policy-relevant surveillance metric. Methodology: We conducted a retrospective ecological analysis of Nigeria’s six geopolitical zones. Drug use burden was obtained from the 2018 Drug Use in Nigeria survey; seizure, arrest, and conviction data were extracted from National Drug Law Enforcement Agency (NDLEA) reports (2021–2022). We calculated a Total Seizure-to–Any-Drug-User Ratio (TSUR; kg seized per 1,000 estimated past-year users) using 2018 survey denominators and 2021–2022 NDLEA seizure totals. Associations were assessed using Spearman’s rank correlation (n = 6 zones). To evaluate robustness to temporal mismatch, we conducted sensitivity analyses assuming ±10–20% changes in zonal prevalence. Results: Past-year drug use prevalence ranged from 10.0% to 22.4% across zones. National seizures increased substantially between 2021 and 2022 and were dominated by cannabis by weight. TSUR varied markedly across geopolitical zones (300 kg per 1,000 users), indicating substantial geographic differences in seizure intensity relative to the 2018 baseline distribution of drug users. Zone-level correlations between seizure weight and estimated users were positive but statistically unstable, given the small number of aggregate units. Conclusions: Marked regional variation exists in both drug use burden and enforcement activity across Nigeria. Using a historical demand-side baseline and subsequent enforcement indicators, this analysis demonstrates only partial geographic concordance between seizures and estimated user burden. The TSUR provides a transparent, scalable metric for contextualising enforcement activity alongside epidemiological estimates in data-constrained settings, when interpreted cautiously.
Albert et al. (Tue,) studied this question.