Skip to 0 minutes and 9 seconds Welcome again to the topic of health facility planning, accessibility, and GIS. In the previous videos, we discussed a number of issues. We started with distributing resources in geographic space. After that, we looked into the accessibility concept. And following that video, we moved into some of the commonly used GIS based measures of accessibility analysis. In this fourth video, we will use an example to illustrate how GIS can support the spatial planning of health facilities.
Skip to 0 minutes and 45 seconds Spatial planning of health facilities actually– and it was mentioned before– is about bringing health services as close as possible to the patients and at the same time ensure that every health facility is able to service sufficient patients. It is about securing a geographical match between allocation and need of health resources. When planning resources, we normally go through a number of steps. We analyze an existing situation. Once we have an understanding of the situation, we try to prioritize our planning interventions. And thirdly, we try to evaluate the impact that we might expect from our planning interventions. So issues that health facility planning deals with are things like, are all the people in a given study area properly served?
Skip to 1 minute and 41 seconds Are there no excesses in travel time for some people in order to reach a health facility? Another issue that we need to deal with is to identify good locations for new health facilities. And in some cases, we also need to check if every health facility is able to service sufficient people. Now, the example that I’m going to use is based upon a control measure which was explained in a previous video. Let us look at the map that we see here. We start by analyzing an existing situation of a large African city, Dar Es Salaam in Tanzania in this example.
Skip to 2 minutes and 21 seconds What we see in this map are the locations of health facilities and different colors indicating different travel times to the nearest health facility. Areas in green have limited travel times. Areas in red are areas where people live that have to walk for more than an hour in order to reach the health facility. So the good thing of using our GIS is that we’re able to visualize the areas that are poorly serviced by a given health care system. At the same time, we can use our GIS to provide numbers. We can quantify how many people actually have to travel long distance.
Skip to 3 minutes and 4 seconds In the table that you see in the slide, we have made different travel time units, units of 10 minutes, and we see how many people have which travel times. In the table, we have also separated between informal areas and planned residential areas because in many African cities, people living in informal areas have much worse access to health care. So using our GIS, we’re able to analyze an existing situation. We can see on the map which areas are problematic and we can quantify the numbers of peoples that are involved.
Skip to 3 minutes and 41 seconds Now, if we have a norm– you see the red line in the table– for example, that a maximum travel time would be 40 minutes, then we would know that all the people that travel for more than 40 minutes are the ones that actually need to be given health care to reduce their travel time. Analyzing the existing situation of health care can also be done in a separate way. And in this case again, we look at the map on the left and the graph on the right. In the map of the left, we see small bars that indicate how many patients every health facility in Dar Es Salaam is able to service. On the right hand, we see the graph.
Skip to 4 minutes and 24 seconds From left to right, we see on the left health facilities that service very few patients. And on the right hand side, we see health facilities that service very many patients. In many cases, we have to deal with minimum standards in planning. In the case of Dar Es Salaam, we have a minimum standard stating that a new health facility can only be constructed if it reaches at least 40,000 people that are not serviced at the moment. We will use such a norm in the examples that follow.
Skip to 5 minutes and 1 second Analyzing an existing situation with GIS is actually not very difficult. This is something that anybody with basic skills of GIS should be able to do. But actually, this is not the type of information that health planners are waiting for. Health planners need to know where should I plan my interventions. Health planners want to know how much benefits will such an investment generate. Here again, we see a map where we have prioritized the interventions. The map on the left shows areas where in the pink color, we see areas that are properly serviced by the existing health care system. We also see some areas with green and red colors.
Skip to 5 minutes and 47 seconds Now, the thing here is that the areas in red are those areas where many people could be reached if a new health facility would be located there. So we are prioritizing. We are not just putting health facilities in any location where we have a lot of high travel time, but we’re able to identify selected areas where the benefit of a new health facility would be highest. If we look at the table in the center of the slide on the left part, we see there for every health facility, how many patients a health facility could service. On the very left of the graph, we see a number of health facilities that can serve service considerable number of new patients.
Skip to 6 minutes and 37 seconds So these are locations which are very well capable of meeting a minimum standard norm. These are locations where a new health facility will easily serve the minimum of 40,000 people. So the good thing is that we have used our GIS to prioritize intervention to identify the best locations for a new health facility. We’re also able to say that in these locations, a new health facility would be able to serve a large proportion of patients.
Skip to 7 minutes and 13 seconds On the table on the right, on the bottom right of the slide, we see again a number of figures which are quantifications of how much the inequality in travel time will reduce its figures on how many additional people can be serviced with one, with two, or with three additional health facilities. So we’re able to identify the best locations for new clinics. We’re able to quantify what the expected benefits are and we can see in that way how much the health system performance will improve. This is an example of adding new facilities. Let me try to explain in a few words how this works.
Skip to 7 minutes and 58 seconds If we look at the map under A, this is the map where we saw the different travel times. All the people that travel for more than 40 minutes are considered to have no access to the health care system. Then again, we go back to our contour method and we calculate for all the unserviced areas the number of people within 40 minutes that can be reached and we summed that to the individual area just like the example given in a previous video. Using this community of opportunity measure, we are able to identify areas with a very high potential of servicing people that are currently not reached by the health care system.
Skip to 8 minutes and 46 seconds So using this methodology, we come back to this map where we are able to identify and quantify optimal locations for new health facilities. Interventions are not only about adding new facilities. This happens a lot, of course, especially in the global south. With rapidly expanding urban areas, with rapidly expanding national populations, normally there is need for additional facilities. But not in all cases do we need to add facilities. The strange thing is that sometimes, we see that health facilities have been located in places where they do not have real added value. This is because the health system is very much influenced by politics.
Skip to 9 minutes and 34 seconds I do not really want to go into too much detail of this slide, but basically, what we have been able to prove using our calculations is that some of the health facilities in Dar Es Salaam city actually could be closed without reducing the accessibility to medical services for the patients. In other words, we could save resources. We could be more efficient without reducing the quality of service to the patients.
Skip to 10 minutes and 7 seconds In conclusion, the approach that I’ve illustrated here actually consists of a number of building blocks. The first is, it is very well rooted in theory. The second is that it works very well with the accessibility concept. The final thing that is important is that I have only shown how to extend the health system by adding new facilities or how to reduce the health system by removing them. But a very similar approach can also be used to reorganize the hierarchy of a health system. Reorganizing the hierarchy, I mean it is possible that some dispensaries actually can service so many people that the variety of their services should be expanded.
Skip to 10 minutes and 57 seconds They should be upgraded from a dispensary maybe to a health center or even to a hospital level. The same applies to the reverse. In some cases, we see a hospital which is actually in a location which is not very suited. It cannot reach many people, and therefore, it might be better to reduce the variety of services, and in that way limit the expenditures of health care a bit. These were not discussed in this short presentation, but they are all based on a similar calculation and it works for all these issues.
Skip to 11 minutes and 39 seconds Of course, we need to discuss a bit also about this health service planning. The things that you have seen have a number of strong points. I mentioned them already. Theory is OK, methodology is OK. Also, the set of analytical tools that are delivered by the GIS are enough to support strategic health planning. Nowadays also to do this is not a very complex issue. With some reasonable level of skills, you should be able to do this yourselves. Of course, an important drawback of the example that I gave you was one of a rapidly growing city. And we did not include anything on the future growth of the city whereas planning is future oriented.
Skip to 12 minutes and 25 seconds So this is a limitation of the work that we did. If we conclude on what we discussed, we can see that health facility planning consists of a number of stages. We want to exist. We want to analyze an existing situation. We need to prioritize planning interventions and we need to evaluate expected impacts. This is very important because we want to provide accurate information to our policymakers. We want them to improve their decision making. And in the end, if these policy makers improve their decision makers, it’s the people that benefit. It leads to improvement of human well-being of people. I would like to thank you very much for your attention. I hope you will enjoy the rest of the course.
Skip to 13 minutes and 14 seconds And who knows? In future, we might even meet again.
Accessibility and Health Facility Planning
In the preceding articles we discussed issues of resource distribution in geographic space, the accessibility concept, and the three most commonly used measures of accessibility analysis in the Geo-Health domain.
Here we focus on how accessibility analysis can be used for strategic spatial planning of health facilities. Health service planners must consider a range of issues when evaluating alternative geographical distributions of their services. From the perspective of the provider, a desired spatial configuration should ensure efficient use of scarce resources. From a patient’s point of view, travel time to a health facility should be within acceptable bounds. The challenge thus is to achieve maximum health benefits to the greatest number of people while making efficient use of scarce health resources. The accessibility concept, is used as an indicator to assess in how far the joint objectives of spatial equality/equity and efficiency are achieved. Broadly speaking, GIS-based health facility planning will go through three stages.
Analysis of the existing situation.
Health planners require ‘diagnostic’ instruments with which they can identify places and populations that are not being well served. Apart from geographically localising such areas, quantification is needed of the number of people that are under-serviced. In addition, there is need to assess if health facilities exist that do not significantly contribute to overall spatial performance. Such facilities - if existent - represent inefficient use of scarce health resources. To perform such an analysis using GIS is straightforward.
Identification and prioritisation of potential planning interventions.
Following the ‘diagnostics’ phase health planners will need to take a future perspective and determine which actions best remedy identified shortcomings. We give one example here using a contour measure. In case of large(r) un-serviced areas (areas where people live with excessive travel time to reach a health facility) one would consider adding additional health facilities. This can be done as follows. First the study area is partitioned into serviced and un-serviced areas/populations using a given travel time threshold. Second, based upon the un-serviced population, we perform a summation - for each possible origin - of the number of destinations (with population as weight variable) using the same travel time threshold. The outcome gives an indication of the potential customer base for every un-serviced origin location. If the market potential of a prospective site is large enough the establishment of a new facility is justified. To perform this type of using GIS is less straightforward but possible using standard functionality.
Evaluation of expected impacts of planning interventions
For informed decision making, however, planners also need to know what the expected improvement in spatial performance of a prospective planning intervention will be. This information is generated in an iterative manner as follows: (i) add the prospective site to the set of existing supply points, (ii) redo the ‘diagnostic’ analysis and (iii) compare the new accessibility scores with the one of the original situation. The GIS analysis needed equals that done during the ’diagnostics’ phase.
Reference: Amer, S., Ottens, H.F.L. (promoter) and de Jong, T. (co-promoter) (2007) Towards spatial justice in urban health services planning : a spatial - analytic GIS - based approach using Dar es Salaam, Tanzania as a case study. Utrecht, Utrecht University, 2007. ITC Dissertation 140, ISBN: 90-6164-253-1.
© University of Twente