Table 3 Interrupted time series model output
IRR | 95% confidence interval | p value | |
|---|---|---|---|
Intercept: baseline average number of surgeries among Tanzanians, 2010, Q1 | 24.8 | 19.1–32.4 | < 0.001 |
T, slope prior to intervention among Tanzanians (ratio of average number of surgeries per quarter) | 1.06 | 1.04–1.08 | < 0.001 |
Z, ratio of average surgery counts between Refugees and Tanzanians at baseline (2010, Q1) | 3.21 | 2.35–4.38 | < 0.001 |
Z x T, ratio of slopes prior to intervention between Refugees and Tanzanians | 0.96 | 0.94–0.99 | 0.005 |
X, ratio of average surgery counts in the period immediately following intervention (2015, Q1) to the last pre-intervention quarter (2014, Q4) among Tanzanians | 0.62 | 0.46–0.84 | 0.002 |
X x T, ratio of postintervention and pre-intervention slopes for Tanzanians | 0.96 | 0.93–0.98 | 0.001 |
Z x X interaction, ratio of the post-intervention change in average number of surgeries between Refugees and Tanzanians, in the period immediately following intervention (i.e. 2015, Q1) compared to the last pre-intervention quarter (2014, Q4) | 1.92 | 1.35–2.74 | < 0.001 |
Z x X x T interaction, ratio of ratios of post-vs. pre-intervention slopes between Refugees and Tanzanians (“difference in difference” on multiplicative scale) | 1.04 | 1.00–1.07 | 0.028 |