Table 2.
Results of linear mixed models of observed competent and respectful labour and delivery care in Kenya, 2010 (standard errors clustered by facilities)
| Competent care (N = 622) | Respectful care (N = 598) | |||||
|---|---|---|---|---|---|---|
| Coeff † | LCL‡ | UCL | Coeff † | LCL‡ | UCL | |
| Patient | ||||||
| Time of day | ||||||
| Evening and night (6 pm–6 am) | ref | ref | ||||
| Afternoon (12 pm–5 pm) | 2.51 | −2.92 | 7.93 | 3.40 | −2.32 | 9.12 |
| Morning (7 am–12 pm) | 5.00* | 0.13 | 9.87 | 6.96** | 2.10 | 11.82 |
| Provider | ||||||
| Gender | ||||||
| Male | ref | ref | ||||
| Female | −1.65 | −6.02 | 2.71 | −3.34 | −8.96 | 2.27 |
| Cadre | ||||||
| Lower § | ref | ref | ||||
| Higher¶ | 1.33 | −2.10 | 4.76 | 2.36 | −2.03 | 6.74 |
| Facility | ||||||
| Type | ||||||
| Public health centre or dispensary | ref | ref | ||||
| Private health centre or maternity | 0.92 | −9.50 | 11.34 | 8.41 | −6.79 | 23.60 |
| Public hospital | 5.14 | −3.37 | 13.65 | −2.10 | −16.00 | 11.80 |
| Private hospital | 13.70** | 4.95 | 22.46 | 12.88 | −1.74 | 27.51 |
| Clinical staff per maternity bed | ||||||
| Less than 5 | ref | ref | ||||
| 5 or more | 7.94** | 2.52 | 13.37 | 4.59 | −1.80 | 10.98 |
| Regions | ||||||
| Coast | ref | ref | ||||
| Central | 23.87*** | 17.16 | 30.57 | 18.67*** | 9.39 | 27.95 |
| Eastern | 12.90** | 3.74 | 22.05 | 4.72 | −5.80 | 15.25 |
| Nairobi | 24.57*** | 18.04 | 31.09 | 17.86** | 5.38 | 30.34 |
| Northeastern | 2.18 | −5.11 | 9.47 | 0.73 | −9.64 | 11.10 |
| Nyanza | 7.58 | −0.95 | 16.11 | 24.55*** | 15.92 | 33.18 |
| Rift Valley | 14.25*** | 8.42 | 20.09 | 7.25 | −3.44 | 17.95 |
| Western | 8.75** | 2.75 | 14.74 | 3.03 | −4.57 | 10.63 |
| Intercept | 38.10 | 27.67 | 48.54 | 43.25 | 27.79 | 58.70 |
| Total variance | 236.79 | 447.58 | ||||
| Provider variance | 110.82 | 77.70 | 158.06 | 151.47 | 101.74 | 225.50 |
| Residual variance | 125.97 | 95.00 | 167.05 | 296.11 | 235.00 | 373.12 |
| Proportion of variance explained | 32.9% | 25.4% | ||||
A total of 626 deliveries were observed in Kenya but 4 women were missing data on all 20 indicators to measure the competent care index and 28 were missing all nine indicators required to measure the respectful care index, leading to analytical samples of 622 and 598 women respectively for each outcome.
The coefficient is the expected difference in visit quality (scale 0 to 100) given a 1 unit difference in the exposure, holding all other covariates constant.
Lower and upper confidence limits.
Lower cadres include enrolled nurses and midwives, community health nurses and nurse aides.
Higher cadres include MDs, clinical technicians, medical assistants, BScN nurses and midwives and registered nurses and midwives.
*P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001.