Results of the multilevel linear regression of provider communication during sick child visits in seven sub-Saharan African countries*
| Model 1 | Model 2 | Model 3 | ||||
| β | 95% CI | β | 95% CI | β | 95% CI | |
| Visit level | ||||||
| Caregiver’s age | 0.11 | 0.06 to 0.16 | 0.10 | 0.06 to 0.15 | ||
| Caregiver’s education | ||||||
| No school | Reference | Reference | ||||
| Any primary | 1.21 | 0.14 to 2.28 | 1.08 | 0.01 to 2.15 | ||
| Any secondary | 4.25 | 2.97 to 5.52 | 3.98 | 2.70 to 5.26 | ||
| Any tertiary or higher | 5.63 | 3.41 to 7.85 | 4.92 | 2.67 to 7.16 | ||
| Caregiver is child’s parent | 2.35 | 1.02 to 3.68 | 2.35 | 1.02 to 3.67 | ||
| Caregiver paid for the visit | 1.78 | 0.67 to 2.90 | 0.66 | −0.53 to 1.85 | ||
| Caregiver has insurance | 0.97 | −0.33 to 2.27 | 0.57 | −0.73 to 1.87 | ||
| Duration of the visit (minutes) | 0.07 | 0.05 to 0.10 | 0.07 | 0.05 to 0.09 | ||
| Reason for visit was child vomiting | 1.78 | 0.92 to 2.64 | 1.80 | 0.94 to 2.66 | ||
| Reason for visit was fever | 0.79 | −0.07 to 1.65 | 0.93 | 0.07 to 1.79 | ||
| Provider level | ||||||
| Provider’s years of education | 0.38 | 0.15 to 0.60 | ||||
| Provider’s years since graduation | 0.05 | −0.02 to 0.12 | ||||
| Number of years provider worked in this facility | 0.05 | −0.05 to 0.15 | ||||
| Provider’s cadre | ||||||
| Doctor or clinical officer | Reference | |||||
| Nurse | 2.90 | 1.18 to 4.61 | ||||
| Other | −1.92 | −4.49 to 0.65 | ||||
| Provider is a manager | 2.55 | 1.29 to 3.81 | ||||
| Provider is female | 1.29 | 0.04 to 2.54 | ||||
| Received training on IMCI | 4.38 | 2.99 to 5.76 | ||||
| Number of in-service training topics in past year | 0.00 | −0.21 to 0.22 | ||||
| Facility level | ||||||
| Private facility | 3.40 | 1.87 to 4.92 | ||||
| Supportive supervision visit last 6 months | 1.69 | 0.34 to 3.05 | ||||
| Infrastructure index | 2.69 | −1.20 to 6.57 | ||||
| Country | ||||||
| Kenya (2010) (reference) | Reference | Reference | Reference | |||
| Malawi (2013) | −5.14 | −7.31 to 2.98 | −3.48 | −5.64 to 1.32 | −3.69 | −5.91 to 1.47 |
| Namibia (2009) | 5.67 | 3.06 to 8.28 | 4.85 | 2.22 to 7.48 | 5.13 | 2.35 to 7.91 |
| Rwanda (2007) | −11.02 | −13.61 to 8.42 | −11.00 | −13.81 to 8.20 | −10.60 | −13.54 to 7.67 |
| Senegal (2012–2014) | −12.06 | −14.34 to 9.77 | −10.96 | −13.24 to 8.67 | −10.82 | −13.58 to 8.06 |
| Tanzania (2015) | 0.88 | −1.11 to 2.88 | 1.69 | −0.31 to 3.70 | 3.02 | 0.98 to 5.06 |
| Uganda (2007) | 12.10 | 9.12 to 15.08 | 13.61 | 10.64 to 16.57 | 13.66 | 10.72 to 16.61 |
| Random effects | ||||||
| Variance between providers | 252.4 | 239.5 | 225.8 | |||
| Intraclass correlation between providers (p) | 0.371 | 0.359 | 0.346 | |||
| Total observations | 14 985 | 14 985 | 14 985 | |||
*The outcome, provider communication, is on a scale from 0% to 100% and is measured during exit interviews with the caregiver.
IMCI, integrated management of childhood illness.