Skip to main content

Table 4 Discrete-time survival models for the effect of Connect experiment intervention versus comparison areas, August 2007–June 2015, Rufiji, Ulanga and Kilombero districts, Tanzania

From: The impact of paid community health worker deployment on child survival: the connect randomized cluster trial in rural Tanzania

Mortality Indicatora Baseline versus endline:
Baseline (August 2007–July 2011) Endline (August 2011–June 2015)
Hazard Ratiob (95% CI) P-valuec Hazard Ratio (95% CI) P-value
Under-5 mortality (0-59 months) 0·99 (0·88–1·11) 0·867 0·95 (0·86–1·07) 0·443
  Exposure period and childhood age specific rates:
Early stage (August 2011–July 2013) Late stage (August 2013–June 2015) Overall (August 2011–June 2015)
Hazard Ratiob (95% CI) P-valuec Hazard Ratio (95% CI) P-value Hazard Ratio (95% CI) P-value
Neonatal mortality (0-28 days) 1·10 (0·89–1·36) 0·392 0·98 (0·74–1·30) 0·902 1·05 (0·85–1·29) 0·659
Post-Neonatal mortality (1-59 months) 0·85 (0·76–0·96) 0·008 1·03 (0·85–1·24) 0·772 0·91 (0·82–1·02) 0·103
Under-5 mortality (0-59 months) 0·92 (0·83–1·03) 0·137 1·01 (0·85–1·20) 0·891 0·95 (0·86–1·07) 0·443
  1. a)Source: Rufiji and Ifakara HDSS data, 2016
  2. b)Models were adjusted for calendar year
  3. c)Standard errors are adjusted for within village cluster effects. Hazard Ratios (HR) for regression models employ dependent variables taking value 1 if the child died during the study period and 0 otherwise
\