No abstract

BackgroundThis study describes job satisfaction and intention to stay on the job among primary health-care providers in countries with distinctly different human resources crises, Afghanistan and Malawi.MethodsUsing a cross-sectional design, we enrolled 87 health-care providers in 32 primary health-care facilities in Afghanistan and 360 providers in 10 regional hospitals in Malawi. The study questionnaire was used to assess job satisfaction, intention to stay on the job and five features of the workplace environment: resources, performance recognition, financial compensation, training opportunities and safety. Descriptive analyses, exploratory factor analyses for scale development, bivariate correlation analyses and bivariate and multiple linear regression analyses were conducted.ResultsThe multivariate model for Afghanistan, with demographic, background and work environment variables, explained 23.9% of variance in job satisfaction (F(9,73) = 5.08; P < 0.01). However, none of the work environment variables were significantly related to job satisfaction. The multivariate model for intention to stay for Afghanistan explained 23.6% of variance (F(8,74) = 4.10; P < 0.01). Those with high scores for recognition were more likely to have higher intention to stay (β = 0.328, P < 0.05). However, being paid an appropriate salary was negatively related to intent to stay (β = -0.326, P < 0.01). For Malawi, the overall model explained only 9.8% of variance in job satisfaction (F(8,332) = 4.19; P < 0.01) and 9.1% of variance in intention to stay (F(10,330) = 3.57; P < 0.01).ConclusionsThe construction of concepts of health-care worker satisfaction and intention to stay on the job are highly dependent on the local context. Although health-care workers in both Afghanistan and Malawi reported satisfaction with their jobs, the predictors of satisfaction, and the extent to which those predictors explained variations in job satisfaction and intention to stay on the job, differed substantially. These findings demonstrate the need for more detailed comparative human resources for health-care research, particularly regarding the relative importance of different determinants of job satisfaction and intention to stay in different contexts and the effectiveness of interventions designed to improve health-care worker performance and retention.

No abstract

Executive SummaryMaternal, perinatal and neonatal mortality remain high in Malawi and throughout subSaharan Africa. The Health Foundation funded community and facility interventions aimed at reducing this burden. MaiKhanda was set up as an independent NGO in Malawi to deliver both interventions with technical support from IHI and WCF. The community intervention involved mobilisation of rural communities through women's groups, and later, maternal and neonatal health task forces. 729 women's groups completed a participatory learning and action cycle to identify and prioritise maternal and neonatal health problems, decide upon local solutions/advocate and lobby for alternatives, and, implement and evaluate such strategies. The facility intervention involved coaching of health facility staff in quality improvement methodology, including Plan-Do-Study-Act cycles, change ideas, bundles and packages and death reviews to improve obstetric and newborn care at 29 health centres (randomised) and 9 hospitals (not randomised).The community and facility interventions were evaluated via a two-by-two factorial cluster randomised controlled trial. All pregnant women in surveillance areas were eligible to take part and consenting women were followed-up to two months after birth via a low-cost community surveillance system using village-based key informants. Primary outcomes were maternal, perinatal and neonatal mortality. A separate non-controlled time-series evaluation of the quality improvement work at the 9 hospitals was under-taken with maternal and neonatal case-fatality rates as primary outcomes. Parallel process evaluations seeking to understand the processes, mechanisms, and intermediate outcomes of the interventions and the context within which they succeeded or failed were also undertaken and are reported in full. A preliminary cost-effectiveness analysis was also undertaken.Both the interventions underwent changes throughout the evaluation period (2007)(2008)(2009)(2010) and could have had sub-optimal dosages. Implementation could also have been improved. There were also political, management and resource challenges in the three districts (Lilongwe, Salima and Kasungu), and at the health facilities, which may have accounted for the lower impact of the interventions than hoped for. Our 21 main findings are as follows. Each of these is extensively discussed with respect to its implications, precision, potential bias, and relevant literature, in chapter 5.Impact on death rates in the overall population and within health facilities 1. The randomised controlled trial suggests neonatal mortality decreased by 22% in areas with both the facility intervention at the health centre and the community intervention in surrounding villages compared with control areas; and, that perinatal mortality decreased by 16% in areas with the community intervention relative to those without the community intervention (see section 4.1). (section 4.2.4). This seems likely given the lack of observed effects on deaths at the health centres and hos...

No abstract