Table 4

Poisson regression of mortality status (person-time of all deaths) by enrolment status and by covariates (distance, place of residence, household size, educational level), Nouna DSS, Burkina Faso
Robust 95% Conf. Interval
Died IRR Std. Err Z P>|z| Lower Higher
Enrolment_1 0.97 0.11 - 0.22 0.826 0.78 1.22
Enrolment_2 0.87 0.14 - 0.86 0.390 0.65 1.18
Firstyear 0.74 0.07 - 3.19 0.001 0.62 0.89
Secondyear 0.70 0.08 - 3.28 0.001 0.57 0.87
Thirdyear 0.72 0.09 - 2.54 0.011 0.55 0.93
Distance_1 1.54 0.09 6.78 0.000 1.36 1.75
Distance_2 2.01 0.14 10.39 0.000 1.76 2.30
Size_1 0.75 0.04 - 6.11 0.000 0.68 0.82
Education_1 0.73 0.04 - 5.15 0.000 0.65 0.82
Education_2 0.53 0.09 - 3.59 0.000 0.38 0.75
Person-time (exposure)

(Std. Err. adjusted for 16397 clusters in id individuals).

Number of obs = 52382 Wald chi2(10) = 247.23.

Log pseudolikelihood = − 9375.1604 Prob > chi2 = 0.0000.

Enrolment_0 is the reference group and represents the group of individuals whose households could not have been members in the scheme because their area was yet to be covered by the scheme.

Enrolment_1 = represents the group of individuals whose households could have been members (areas covered by Nouna scheme), but were not.

Enrolment_2 = represents the group of individuals whose households were members, and their areas were covered by Nouna CBHI scheme, and.

Person-time is defined as time spent up to death within the Nouna DSS from inception throughout 2007.

Enrolment_0 is the reference group (deaths before the insurance scheme) and additional dummy variables were created (firstyear, secondyear, and thirdyear) for time spent in the scheme to assess any “dose-response” relationship.

Firstyear = year 1 of the Nouna CBHI (2004–2005).

Secondyear = year 2 of the Nouna CBHI (2005–2006).

Thirdyear = year 3 of the Nouna CBHI (2006–2007).

For distance, dummy variable were created in relation to Nouna town where is the district hospital (and not distance to health centers because life saving skills and equipment are mainly at the district hospital).

Distance_0 = reference group for distance to district hospital (Nouna town).

Distance_1 = group of households not in Nouna town and less than 35 km from the district hospital.

Distance_2 = group of households at more than 35 km from district hospital.

(Size) Household size_0 = reference group (household size 1–10).

(Size) Household size_1 = group of households with more than 10 members.

Education_0 = reference group for educational level (none).

Education_1 = primary school level.

Education_2 = secondary level or more.

Hounton et al.

Hounton et al. BMC Health Services Research 2012 12:363   doi:10.1186/1472-6963-12-363

Open Data