Health information should be reported on a simple, standardized surveillance form. (A sample form, adapted from WHO Emergency Relief Operations, is located at the end of this section.) Each health facility should be held accountable for completing the reporting form at the appropriate interval and for returning it to the person or agency charged with compiling the reports, analyzing the information, and providing feedback. Each refugee settlement or camp should have a person responsible for coordinating the HIS. Forms should be translated into the appropriate local language(s) if community health workers are involved in information collection.
Health facilities should keep a daily record of patients; age, sex, clinical and laboratory diagnosis, and treatment should be specified. If personnel time is limited, a simple tally sheet should be used. In addition, the patient should be issued a health record card on which the date, diagnosis, and treatment are recorded. Each time a patient contacts the health-care system, whether for curative or preventive services, this should be noted on the health record card. Laboratory data should accompany diagnostic information whenever possible. Collecting Processing, Storing, and Shipping Diagnostic Specimens in Refugee Health-Care Environments * provides an overview of procedures for collecting and processing diagnostic specimens in the field.
* Available from IHPO, CDC, 1600 Clifton Road, MS F-03, Atlanta, GA 30333, 404-639-0308.
Data collection should be limited to that information that can and will be acted upon. Information that is not immediately useful should not be collected during the emergency phase of a refugee relief operation. Overly detailed or complex reporting requirements will result in noncompliance.
The most valuable data are generally simple to collect and to analyze. Standard case definitions for the most common causes of morbidity and mortality should be developed and put in writing. The data collected will fall into one of the following categories: a) demographic, b) mortality, c) morbidity, d) nutritional status, and e) health program activities.
Population. Camp registration records should provide most of the demographic information needed. If registration records are inadequate, a population census may be necessary. Conducting a census is often politically sensitive and may be delayed by the administrative authorities for a long period of time. Consequently, innovative methods may need to be devised. For example, organize a nutritional screening of all children less than 5 years of age. Count the children and estimate the percentage of the total population less than 5 years of age by doing a sample survey. From this information, estimate the total population size. For other methods to determine population size and structure see "Rapid Health Assessment".
It is important that population figures be updated on a regular basis, taking into account new arrivals, departures, births, and deaths. The total population is used as the denominator in the calculation of disease incidence, birth, and death rates. This total is also necessary to determine requirements for food and medical supplies and to estimate program coverage rates. Information about the population structure is needed to calculate age- and sex-specific morbidity and mortality rates, to estimate ration requirements, and to determine the target population for specific interventions, i.e., antenatal care and immunizations.
The rate of new arrivals and departures gives an indication of the stability of the population and will influence policy decisions about long-term interventions, such as TB therapy. This information is also used to predict future resource and program needs.
A birth registration system is usually simple, since the community expects an increase in the family food ration as a result of a new birth. Births might be reported in the community to volunteer health workers or traditional birth attendants. Alternatively, if good antenatal care services are established, follow-up of pregnant mothers will allow for a relatively complete registration of births. Examples of mortality surveillance systems are described in "Rapid Health Assessment". Deaths may be underreported if there is a fear of possible ration reduction; thus, an agreement might be negotiated with camp authorities not to decrease rations after a death occurs at least during the emergency phase. Arrivals and departures should be monitored through the camp registration system.
Mortality. Each health facility should keep a log of all patient deaths (with cause of death and relevant demographic information) and report the deaths on a standardized form. Because many deaths occur outside of the health-care system, a community-based mortality surveillance system should be established. Such a system may include the employment of grave watchers, the routine issuance of burial shrouds, and the use of community informants (see "Rapid Health Assessment").
Death rates are the most specific indicators of a population's health status and are the category of data to which donors and relief agencies most readily respond. During the emergency phase of a relief operation, death rates should be expressed as deaths/10,000/day to allow for detection of sudden changes. In refugee camps, relief programs should aim at achieving a CMR of less than 1/10,000/day as soon as possible. This rate still represents approximately twice the "normal" CMR for non-displaced populations in most developing nations and should not signal a relaxation of efforts. After the emergency phase, death rates should be expressed as deaths/1,000/month to reflect the usual reporting frequency and to facilitate comparison with baseline, non-refugee death rates.
Age- and sex-specific mortality rates will indicate the need for interventions targeted at specific vulnerable groups. During the early stage of a relief operation, specific death rates for persons less than 5 years of age and greater than 5 years of age may suffice. Later, further disaggregation by age may be feasible -- for example, less than 1 year, 1-4 years, 5-14 years, and greater than 15 years. Different male- and female-specific death rates may reflect inequitable access to resources or health services. Cause-specific mortality rates will reflect those health problems having the greatest impact on the refugee community and requiring the highest priority in public health program planning.
Morbidity. Health facilities and feeding centers should report morbidity information on the same form on which mortality is reported. Each disease reported in the system must have a written case definition that will guide health workers in their diagnosis and ensure the validity of data. Where practical, case definitions that rely on clinical signs and symptoms should be tested periodically for sensitivity and specificity as compared with a laboratory standard (e.g., malaria).
Knowledge of the major causes of illness and the groups in the affected population that are at greatest risk allows for the efficient planning of intervention strategies and the most effective use of resources. Morbidity rates are more useful than a simple tallying of cases, as trends can be followed over time, or rates compared with those from different populations. The monitoring of proportional morbidity (e.g., percentage of all morbidity caused by specific diseases) may be useful when specific control measures are being evaluated, although caution is needed in the interpretation of trends. A relative decrease in disease-specific proportional morbidity may merely reflect an absolute increase in the incidence of another disease.
Nutritional status. Data regarding nutritional status can be obtained through a nutritional assessment survey or a mass screening exercise. Surveys should be repeated at regular intervals to determine changes in nutritional status; however, not so frequently as to obscure true differences between surveys. All children less than 5 years of age should undergo a nutritional screening upon arrival at the camp and should continue to be weighed and measured monthly at MCH clinics in the camp. Information collected during these screenings should be included in HIS reports. If the initial screening identifies high prevalence rates of undernutrition, cross-sectional surveys should be repeated at intervals of 6-8 weeks until the undernutrition prevalence rate is below 10%. Thereafter, surveys every 6-12 months will suffice, unless routine surveillance data indicate that nutritional status has deteriorated. Measurement of nutritional status is described in the Rapid Nutrition Assessment Manual. (*)
(*) Available from IHPO, CDC, 1600 Clifton Road, MS F-03, Atlanta, GA 30333, 404-639-0308.
The prevalence of acute malnutrition acts as an indicator of the adequacy of the relief ration. A high prevalence of malnutrition in the presence of an adequate average daily ration may indicate inequities in the food distribution system, or high incidence rates of communicable diseases (e.g., measles and diarrhea). The presence of nutritional deficiency disorders (i.e., pellagra, anemia, or xerophthalmia) indicates the need for ration supplementation.
Programs. Each health facility should keep a log of all activities. Immunizations should be recorded in a central record, as well as on the person's health record card. Records of health sector activities will be useful in determining whether certain groups in the population are underserved, and in planning measures to reach a broader population base. Although approximate immunization coverage may be estimated from the number of vaccine doses administered, the preferred method is by annual population surveys.