Printer Friendly

An environmental health evaluation tool for locating and assessing disaster relief and refugee camps.

The location of disaster relief and refugee camps has been shown to have significant effect on the rates of diseases associated with the environment, such as tuberculosis and malaria (1, 2). The location of most camps is often influenced by the political, social, economic or military realities of the host countries, and relief agencies must often choose from among a few sites that may not be optimum. Evidence is shown that no matter how much money is spent on environmental health services, camps in poorer environmental locations may have higher rates of disease associated with the environment than do camps located at better sites (3). The overall, or end, objective of providing good refugee health by optimizing environmental health at a given site can be met in two ways: 1) select the best site to begin with, and 2) provide sufficient environmental services in established camps to offset constraints of the site.

To accomplish the task for this research, that is to establish a weighted value tree for camp evaluation, the following objective were established: 1) interview former and current providers of environmental health services in disaster relief and refugee camps to determine, rate and weight environmental health objectives and their respective attributes; 2) develop a model, a weighted value tree, based upon the normalized weights of those interview-determined objectives and attributes; and 3) test the model in existing relief operations, comparing locations to rates of selected environmentally-incurred diseases.


In order to determine optimum camp location from an environmental health standpoint, when two or more locations are available, and to assess water and sanitation services in existing camps, a weighted value tree was identified as a potential candidate for providing that assessment. This form of decision analysis has had previous applications which have demonstrated its capability in prioritizing environmental services and resource allocation (4, 5, 6). The objectives of this research were met by modifying the procedure proposed by Edwards and von Winterfeldt and Edwards (7, 8).

Step 1. Rank the environmental health objectives in order of importance. Then rank each objectives' attributes. An example of an environmental health objective is "adequate water." Two measurable attributes of this objective would be an optimum quantity of water and sufficiently good quality of water.

Step 2. Weight the objectives and attributes in importance, preserving ratios. Start by assigning a "10" to the least important objective or attribute. Then rate the objectives or attributes in a pairwise manner.

Step 3. Sum the weights and divide each by the sum. This results in a weighted average ||W.sub.i~~ based on relative importance, or importance weight for each attribute. The individual weights of an objective and for each of its respective attributes, from which the importance weights are determine, are here referred to as normalized weights.

Step 4. Calculate site scores based on the sum scores of the importance weights |SS=|Sigma~ |W.sub.i~(|x.sub.i~)~. The symbol |x.sub.i~ represents each attribute and its respective objective.

Step 5. Decide. Utility theory mandates that the decision alternative or attribute with the higher importance weight is preferred. When selecting a camp site location from among two or more choices, the location with the highest importance weight would be selected. If the exercise is used to assess an existing camp, resources should be allocated to those objectives with the highest importance weights.


The interview candidates for this study were identified within international relief agencies. United Nations High Commissioner for Refugees (UNHCR) Program Officers, other international agency personnel and experts who have worked in water, sanitation and health programs throughout the world have first-hand knowledge of the problems encountered in delivery of environmental health services. They have had practical experience with the realisms of service constraints in refugee camps and have often been personally involved in setting up such camps. Interviewing these people in a structured manner, and in a way that stimulated thinking, was expected to provide the necessary data for the weighted value tree.

There were thought to be an estimated TABULAR DATA OMITTED 100 international experts who might qualify as candidates. A sampling of 10 percent of these was considered sufficient. An assumption of decision analysis is that experts, given the same information, will tend to make the same decisions (9). Those who participated in the interviews were first asked to verify that the list of objectives and attributes was complete and to add or delete as necessary. Then the candidate ranked and weighted each objective and set of attributes according to Steps 1 and 2 of the methodology.

Weighted value tree

The weighted average or importance weight for each objective was determined by the following process: raw weight scores for each group (ultimate objectives or objectives' attributes) were summed, then each objective or attribute score was divided by that sum. This resulted in a normalized weight. The normalized weight for the individual ultimate objective was then multiplied by the normalized weight for each of its corresponding attributes to produce the importance weight. For example, the normalized weight for the water supply objective might have been 0.261. If normalized weight score for the water quantity attribute was 0.667, the importance weight ||W.sub.i~~ was then calculated as 0.174 (0.261 X 0.667). The normalized weights for all ultimate objectives and for each objective's set of attributes were added to assure the sum was 1. The importance weights, along with their respective attributes and objectives, were then graphically presented as a polycotomous weighted value tree model.

Due to economic and political constraints, three camps in Thailand, selected by UNHCR and the United Nations Border Relief Organization (UNBRO), were visited. Testing of the environmental health priorities determined from the interviews was carried out in May and June 1990. The three camps visited were Ban Vinai, containing predominantly Hmong hillstribe people; Ban Napho, with Lao and Thai Esson; and Site 2, a camp of Cambodian refugees. The camp survey form was filled out at each site.

Morbidity data was also collected from the senior camp medical officer. This information was to be used to compare camp scores against environmentally-incurred disease. The morbidity data, however, were incomplete and inconsistent between camps. Therefore, a modified decision analysis rating and ranking methodology was used by camp medical staff to assess levels of selected environmentally-incurred illnesses. The summary camp scores and selected environmentally-incurred morbidity rates are shown in Table 1.

The categorical rates of morbidity (low, moderate, high and very high) correspond roughly to the following incidence rates per 1,000 people: |is less than~ 0.1, 0.1-1, 1-10, and |is greater than~ 10. In the analysis they were assigned the values 4, 3, 2 and 1, respectively.


In the evaluation of individual ultimate objectives, there is little argument that water supply and, specifically, water quantity stand out as major priorities in the location of disaster relief and refugee camps. However, any model for site selection must take into account the predominant environmental source of immediate mortality or acute morbidity. The example is given of high rates or respiratory illnesses being the major morbidities in Italy during the winter where displaced persons were housed in tents in an area of abundant water. Increased rates of pneumonia and acute respiratory infections have been noted elsewhere by health workers following inundations and floods, due to cold stress from wind and cold weather (10). Water at these locations was presumed abundant. Similarly, heat stress is exacerbated by humidity in hot climates. Protection from the sun is, therefore, a high priority.

It does little good to put refugees or displaced persons in an area with abundant water, based on the presumed number one priority for water, when freezing conditions or excessive sun are preventing occupants from surviving long enough to need water, unless immediate and adequate protection is provided. Although there was an expectation during this research that the interview would elicit this, such was not realized. Several candidates commented that objectives and criteria would change based on different site characteristics, but failed to offer suggestions, unless specifically requested to do so. Each candidate was requested to provide priorities and criteria based on site variability. However, these researchers apparently failed to provide sufficient amplification or clarification.

Therefore, the final model or models should take this ability to prioritize campsites based on the prevailing major environmental trauma. If, however, a site survey is not possible, or complete information on the site is otherwise difficult to determine, then environmental health services may be prioritized and planned according to the normalized weights of the ultimate objectives. Such situations may occur where refugees or displaced persons are already at a site of undetermined characteristics and the site is inaccessible to planners and provisioners. Those priorities are more simply stated in Table 2.

Both weather protection and access had the largest differences in prioritization among interviewed candidates. This is also reflected in candidate comments that site selection should be specific to the site. One rationale for retaining access as a higher priority item is that some form of weather protection is usually available at any site, be it even an excavation in a soil bank or simple structure made from local plants and materials. Access is more difficult to correct and, often, little can be done to access remote camps.


A weighted value tree, which establishes the priorities of environmental health services for disaster relief and refugee camps, may be used as a tool for assessing new sites for compliance. It can also be used to assess existing camps to assure that resources are directed toward the most important attributes. Such a weighted value tree was developed, based on interviews with international experts on disaster relief and refugee camp environmental health. The tree was then tested in existing refugee camps in Thailand, after being translated into a camp survey form. That survey form, as modified for a hot, moist climate, was found to be valid within that climate.
Table 2
Environmental health priorities for disaster relief and refugee
Priority Weight(1) Environmental heath component
1 0.245 Water supply
2 0.169 Camp access
3 0.144 Sewage disposal
4 0.117 Drainage
5 0.096 Solid waste disposal
6 0.093 Vectors and pests
7 0.091 Weather protection
8 0.045 Air pollution and noise
1 Environmental health objective normalized weight

Where at all possible, campsite surveys should be conducted at the prospective site. When that is possible, and sufficient information can be gathered, priorities for environmental health services can be established using the decision analysis technique of this study.

Where environmental health priorities cannot be determined at a site, or where information from a selected site is too difficult to determine, the value tree may be used in establishing environmental health service priorities. Those again have been very simply set forth in Table 2. The preferred method, however, would be to use the formal decision analysis method, as proposed in this research.


1. de Ville De Goyet, C. and M.F. Lechat (1976), "Health aspects in natural disasters," Tropical Doctor 6(4):152-157.

2. Simmonds, S.P.V. and S.W. Gunn (1983), Refugee Community Health Care, Oxford University Press, New York, NY.

3. Shook, G.A. (1983), "A sanitary survey form for locating refugee camps," J. Env. Health 45(6):295-298.

4. Keeney, R.L. (1980), Siting Energy Facilities, Academic Press, New York, NY.

5. Hankins, R.W.and P.J. Fos (1989), "Objectives for a system of health care delivery for HIV infected people," SocioEconomic Planning Science 23(4):181-193.

6. Cantor, L.W. and R.C. Knox (1985), Groundwater Contamination Control, Lewis Publishers, Chelsea, MI.

7. Edwards, W. (1977), "How to use multiattribute utility measurement for social decisionmaking," IEEE Transactions on Systems, Man and Cybernetics 7(5):326-340.

8. von Winterfeldt and Edwards (1987), Decision Analysis and Behavioral Research, Cambridge University Press, Cambridge, UK.

9. Griffin, R.W. (1987), Management, Houghton-Mifflin, Dallas, TX.

10. Beinin, L. (1979), "Sanitary consequences of inundations," Disasters 3(2):213-216.

Gary Shook, MSPH, Sc.D., EHS, Asian Institute for Technology, ADPC, GPO Box 2754, Bangkok 10501, Thailand.
COPYRIGHT 1993 National Environmental Health Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Author:Fos, Peter
Publication:Journal of Environmental Health
Date:May 1, 1993
Previous Article:Albuterol nebulizer use in Bahrain and the burning oil wells of Kuwait.
Next Article:Programme planning: a management system for environmental health services.

Related Articles
Public Health International.
IRAN - Sept. 24 - Tehran Prepares For Refugees.
ARABS-ISRAEL - Nov. 22 - UN Aid Worker Killed.
New emergency Preparedness products. (Products & Services).
In the shadow of the volcanoes: the impact of intervention on the nutrition and health status of Rwandan refugee children in Zaire two years on from...

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters