Table 2.
Teamwork and coordination characteristics reported by accredited social health activists (ASHA) and Anganwadi workers (AWW) after implementation in the Team-Based Goals and Incentives intervention trial in Begusarai, Bihar, 2012–2014*
| Modelled indicators | AWW n=285 |
ASHA n=274 |
||||
| Control (n=148) | Intervention (n=137) | P value | Control (n=153) | Intervention (n=121) | P value | |
| Attended three or more subcentre meetings in the past 3 months (%) | 70.1 | 72.4 | 0.72 | 74.9 | 86.8 | 0.02 |
| Working in a team | ||||||
| Consider … part of their team | ||||||
| Other FLW of the village (other cadre) (%) | 87.6 | 90.7 | 0.49 | 80.7 | 70.7 | 0.15 |
| Subcentre auxiliary nurse midwife (ANM) (%) | 80.8 | 93.3 | 0.01 | 57.4 | 76.4 | <0.01 |
| Same-cadre front-line workers (FLW) at subcentre (%) | 13.1 | 24.3 | 0.05 | 27.4 | 27.3 | 0.99 |
| Other-cadre FLWs at the subcentre (%) | 12.2 | 19.1 | 0.19 | 10.1 | 16.7 | 0.13 |
| Can always get help from team when needed (%) | 41.5 | 51.3 | 0.15 | 47.4 | 60.0 | 0.07 |
| Always expected to plan with team (%) | 51.1 | 67.9 | 0.01 | 58.4 | 71.4 | 0.08 |
| Always expected to meet regularly with team (%) | 51.8 | 63.7 | 0.08 | 64.4 | 60.5 | 0.52 |
| Working with ANM | ||||||
| Times met with ANM outside subcentre meetings in the past 3 months (mean)† | 3.7 | 4.0 | 0.34 | 3.6 | 4.1 | 0.13 |
| Any joint visits with ANM in the past month (%) | 50.4 | 48.6 | 0.79 | 68.9 | 62.2 | 0.41 |
| Working with opposite - cadre FLW | ||||||
| Average joint home visits in the past week (mean) | 1.5 | 1.7 | 0.38 | 1.5 | 1.3 | 0.48 |
| Average times met with opposite-cadre FLW in the past week to discuss work (mean) | 1.9 | 2.2 | 0.08 | 2.0 | 2.0 | 0.89 |
| Average times asked opposite-cadre FLW to conduct visit (because you could not) in the past 30 days (mean) | 1.3 | 1.6 | 0.19 | 1.4 | 1.7 | 0.39 |
| Average times opposite-cadre FLW asked to conduct visit (because she could not) in the past 30 days (mean) | 1.0 | 1.3 | 0.18 | 1.0 | 1.6 | 0.01 |
*Survey-weighted percentages and counts are reported to account for the survey design. Survey regression models were performed that accounted for village as the primary sampling unit and subcentre as the primary stratum within the sampling unit, and with proportional sampling weights at the FLW level. Survey logistic regression models were conducted for binary outcomes and survey Poisson regressions for count variables. Separate regression models were conducted for each outcome, for each cadre (AWW, ASHA); p values reflect comparisons of the intervention and control groups for each cadre.
†Three AWWs had missing responses for this outcome.