Statistical analysis

We used the chi-squared test to compare the distributions of categorical variables and the Kruskal-Wallis test for continuous variables. We calculated the prevalence of INSTI resistance mutations over the study period and by year for those whose first genotype was sent ≤3 months after diagnosis and for those whose first genotype was sent >3 months after diagnosis. We calculated binomial exact 95% confidence intervals for prevalence estimates. We performed a sensitivity analysis and calculated the prevalence of INSTI resistance mutations using alternative cut-offs of 1 month and 6 months to define the treatment-naïve population. We used logistic regression to determine trends in the prevalence of INSTI resistance mutations over the study period. Statistical significance was defined at the P<0.05 level. We used STATA 14.2 for all analyses (College Station, TX).