Distance to basic services and Retention of Health Workers in Tanzania: A multivariate logistic and GIS Model Approach.

Author(s) Benjamin Mayala1, Jonathan Mcharo2, Vitus Nyigo3.
Affiliation(s) 1Disease Surveillance and GIS, NIMR, Dar es Salaam, Tanzania, 2HRH, NIMR, Dar es Salaam, Tanzania, 3Traditional Medicine, NIMR, Dar es Salaam, Tanzania.
Country - ies of focus Tanzania
Relevant to the conference tracks Health Workforce
Summary Lack of basic services in some areas of the country is one of the major reasons for health workers to migrate to areas with better services but this causes other areas to remain with few health staffs, and resultant poor health care. A simple analysis to determine the availability of these services and at what distance a health worker can access them can be a great solution to policy makers.
Background Tanzania has been implementing the Primary Health Service Development Program policy to increase the accessibility of health care services to its people at a distance of 5km. Although this policy has been introduced in the various districts in the country, its implementation has been a challenge due to the fact that retaining health workers at these facilities remains problematic. Lack of basic services in some areas of the country is one of the major reasons for health workers to migrate to areas with better services which results in other areas retaining fewer health workers with poorer health care provision in the country. Lack of health care provision in terms of health workers can present barriers to patient’s access to health facilities, who might be forced to travel long distances to access health care. Understanding the problems of health workers, including the availability of basic services in their localities, is important. Distance to basic services has an impact on health worker retention. Travel times, lack of access to transportation, and seasonally inaccessible roadways can present barriers to health workers access to important services such as banks, security or schools for their children.
Objectives We hypothesized that the lack of basic services was a key factor for health worker migration to other districts that possessed better services. The main goal of this study was to determine the availability and distribution of basic key services that are important to health workers and how close they can be accessed. Specifically, we used spatial analysis to establish a network that considers geographic position and the existence of basic services that could influence the retention of health workers.
In order to achieve this goal we focused on the following questions:
• Is there a reliable water supply source and what is the nearest distance to this source?
• How accessible are the health facility (your place of work)?
• Are there roads which can easily access different routes to services?
• Are there schools that can be accessible by health workers?
• Is there reliable transport (public) that can be used by health workers to move from one point to another?
In particular Euclidian distances from health workers houses to the basic services were computed. Using data on health workers moving in/out of a particular district, we applied multivariate spatial logistic regression to determine the variables that were statistically significant.
Methodology The study was undertaken in 16 selected district in Tanzania in 2007. Districts were chosen to represent the spatial zonal distribution of the country. Four of the districts (Kigoma Urban, Mtwara Urban, Nymagana and Temeke) were considered urban and the other 12 (Biharamulo, Kilwa, Kondoa, Lushoto, Mafinga, Manyoni, Mbinga, Mpanda, Same, Ulanga, Urambo and Rungwe) were rural districts. Four villages were selected in each district, and three to four health workers houses were selected for inclusion in the mapping exercise. For each district we first assessed all the basic services available, then a hand held Global Positioning System (GPS) was used to map the geographic locations of those services, this includes banks, water source, post office, police post, shops/open market location, schools and bus stops. Other features that located were the health facilities and health worker households. We used ArcGIS 10.1 analysis tools to map the locations of basic services and distances to the nearest services from each health worker household was calculated using simple Euclidean distance. Multivariate logistic regression were used to model the distances to bank, post office, police, schools, referral hospital, water sources and bus stops. Then we compared the significance variables to the data on overall district move in/out of health workers.
Other covariate variables of interest that we included in the spatial logistic model that could be important in our analysis included elevation, rainfall and temperature data.
Results Main sources of Water: Our results indicated that most of the health worker households and the health facilities they work were not connected to piped water, because there was no such services in the districts. Therefore, health workers are forced to walk long distances (up to 15km) to access water services.
Distance to bank, police and post office services: Most of the health workers in public health facilities collect their salaries through a bank. Our analysis indicated some health workers travel a distance of 160 km to obtain bank, post or police services. This was seen to be a burden especially in those areas where public transport is a major problem (i.e. one bus a day), and that health workers had to take two or three days off, spend nights in a guest house near the bank, and also incur costs for fare and food, which are deducted from their salary.
Accessibility of Health facilities: in most the districts visited during the research, quite large number of health facilities are located near to the roads (0.2 to 3 km). This means that they (in principle) can be accessed by public transport. However, this is not the situation in most of the rural areas where we conducted this research. As mentioned before, public transport is not reliable and the situation is worse during rainy season.
Distance to School: The distance analysis to access schools indicated that schools can be accessed up to 12 km which is also a burden for most children in the rural areas without transport.
Our initial logistic regression models confirmed a statistical significance of distance to basic services and moved in/out of health workers.
Conclusion In this study we have used GIS and spatial logistic analysis to determine the spatial distribution of existing basic services for health workers. Apart from others, issues like better salary, promotion and various ways of motivating a health worker and understanding the distance to access basic service is important. This is due to the fact that lack of services in some area may cause health workers to migrate to those areas with better services. The application of GIS technology has shown how a health worker can access basic services in terms of travel distance. A multivariate analysis indicated the significance of some variables to the migration of health workers.

, , , , , , ,

No comments yet.

Leave a Reply