Malaria Control During Mass Population Movements and Natural Disasters (2003)

In some situations, special studies might be required to more accurately estimate the prevalence or incidence of malaria-associated illness, to evaluate the efficacy of malaria therapy, to determine the principle malaria vectors in the area, or to test new interventions or monitor the effectiveness of existing ones. These operational research issues are necessary in order to make sound programmatic decisions. However, there is a sensitivity about the use of the term “research,” even though these research issues are operationally oriented activities.

Historically, there has been resistance to conducting research during complex emergencies, particularly during the acute phases of an emergency. The concern was that it is unethical to impose the burden of research on vulnerable populations that are traumatized, frequently in ill health, and attempting to survive in new surroundings. Slowly, this sentiment has been shifting toward the recognition that not only is research needed but that it is unethical to continue to practice public health measures that may or may not reflect “best practices” (Burkle, 1999; Waldman and Martone, 1999; Banatvala and Zwi, 2000; Brundtland, 2000; McClelland et al.,

2000; Ezard, 2001a). However, what constitutes best practices in the context of an emergency is often not known, as the base of applied research in this area is limited (Burkle, 1999). Best practices have been based mostly on summaries of personal and/or organizational experiences, lacking theoretical perspectives and rigorous application of standard clinical, epidemiological, or other scientific methodologies needed to answer research or programmatic questions (Waldman and Williams, 2001). Little research has been documented, as evidenced by a small inventory of health research in emergencies (World Health Organization, 1999b). But initiating research and monitoring and evaluating ongoing malaria control activities are necessary to improve humanitarian relief practices (Dick and Simmonds, 1983; Walkup, 1997; Banatvala and Zwi, 2000). In November 1997 an international meeting on applied health research in complex emergencies was convened at the World Health Organization, and the Advisory Group on Research in Emergencies was established. Malaria was identified as one of the priority areas of communicable disease, and recommendations were specifically made to conduct research into the use of rapid diagnostic tests and prevention of transmission by use of insecticide-impregnated bed nets (World Health Organization, 1997b; Advisory Group on Research in Emergencies, 2000).¹

While the need for malaria-related research is evident, the feasibility of such projects should be considered prior to attempting to initiate research or special studies during a complex emergency. Outside researchers can be viewed as unwelcome burdens during emergencies, particularly during the initial phases of relief operations. However, if collaborative partnerships are forged from the beginning among those conducting the research, the nongovernmental organizations (NGOs) or other agencies providing the malaria control activities, and representatives of the intended beneficiary population, such studies will be feasible. Attention should be placed on the issues identified as problematic by field staff or the host government, and these should be relevant from a practical standpoint. Permission needs to be obtained from many levels—host governments, participating agencies,

the United Nations High Commissioner for Refugees or other lead relief organization, the participants themselves, and so forth. Consideration must be given to the practicalities of travel logistics, hiring of additional staff (both from the host government and from within the displaced population), the need for ongoing communication among all involved parties, cost effectiveness, and, particularly, security.

The research questions should be examined to determine whether they address only the displaced population or whether data gleaned from such studies have wider application to the host population (Williams and Bloland, 2001). For example, the authors of this monograph conducted antimalarial drug efficacy trials in refugee camps in western Tanzania in 1998. Those results complemented drug efficacy studies that had been conducted on the host population by the Ministry of Health. Together, those data helped to inform the decision to change the treatment of uncomplicated malaria in both the host country and the refugee population. However, a caveat was raised recently about competing tensions between the needs of the scientific community and those of the community involved. Ezard (2001a) provides an example of a refugee situation in which antimalarial drug efficacy data were urgently needed at the same time that local capacity needed to be strengthened to support self-determination. While attempting to recognize the needs for collaboration with the host country’s health care professionals, prioritization of antimalarial drug efficacy studies did not match the local priority of establishing a new national health care system. Although ultimately successful in meeting both sets of priorities to some degree, Ezard stresses the need to recognize the underlying tensions between the achievement of short-term research goals and longer-term transfer of skills and development of local self-determination.

Roll Back Malaria has become very active in supporting and encouraging operational research related to malaria control and development and assessment of new tools and technologies appropriate for use in complex emergencies (see Allan and Guillet, 2002). Some of these technologies have been mentioned previously, such as wash-durable insecticide-treated nets and other insecticide-treatable emergency shelter materials. Many NGOs have recognized the need for additional malaria-specific operational research activities and are becoming very involved in conducting valuable research in emergency settings.

If definitive diagnosis of malaria infection is not routine, rapid blood smear surveys of febrile patients can assist in evaluating the reliability of presumptive diagnoses. Surveys such as this are indicated in any area where malaria is believed to be transmitted and should be repeated during different times of the year, especially in areas where malaria transmission is believed to be seasonal. In Niger during the rainy season, for instance, over 99 percent of febrile patients were presumptively diagnosed as having malaria. But blood smears obtained from those patients confirmed malaria infection in only 62 percent. During the dry season, 79 percent of febrile patients were presumptively diagnosed as having malaria, but only 5.4 percent actually had parasites on blood smear (Olivar et al., 1991).

Rapid prevalence surveys can be done by simply obtaining and reading blood smears from 50 to 100 patients with fever or a history of fever who attend a health clinic. In areas where children are at greater risk than adults, as typically occurs in highly endemic areas, separate surveys based on age group should be done.

The best method to assess the efficacy of a chosen malaria therapy is by the in vivo test. The World Health Organization originally standardized the methodology, and there have since been numerous modifications. For the purposes of evaluating the efficacy of malaria therapy on a national level, emphasis has been on long follow-ups of patients (14 to 28 days or even longer). For the purposes of rapid assessment of therapy efficacy in an emergency setting, much useful information can be obtained by monitoring patients for even a minimal amount of time (such as 7 days), although longer follow-up of patients is preferable if at all possible. Suggested methods are presented in Appendix B.

Any malaria control program that includes vector control or avoidance strategies requires a detailed entomological assessment to identify the primary vectors responsible for transmitting malaria. To ensure a valid

assessment, these surveys must be conducted by a trained entomologist knowledgeable in malaria vectors and vector behavior.

As discussed in Chapter 8, little research has been done to answer behavioral questions related to malaria control activities in complex emergencies. In addition to the rapid assessments described earlier, other methods could be used, such as heath care facility surveys, with exit interviews to determine patients’ understanding of their prescribed antimalarial medications or random surveys of households in displaced areas to inventory the types of drugs used for home or self-treatment of malaria.

Other research areas for which there are almost no data from displaced populations include patterns of treatment-seeking behaviors in a refugee camp or settlement area, the effect of economic dependency on the retention of distributed personal protective items (such as bed nets), patterns of compliance with antimalarial treatment during pregnancy, and understanding and correct usage of antimalarial drug treatment policies by relief health care workers. Samples of questions to guide sociobehavioral research can be found in the previous chapter. As noted previously, few NGOs have the internal capacity to conduct this type of research; such research should be guided by a trained social scientist.

Reproductive health and gender-related research have been identified as areas of pressing concern in emergency situations (World Health Organization, 1997b; Enarson, 1998; Palmer and Zwi, 1998; Palmer et al., 1999; Khaw et al., 2000). Studies of health care delivery could focus on better ways to integrate malaria prevention and treatment during pregnancy with other reproductive health-related interests, such as general management of complications related to pregnancy, the care of pregnant women infected with the human immunodeficiency virus, and the need to strengthen the linkages between traditional midwife care and antenatal services.

Questions pertaining to the feasibility and cost effectiveness of implementing preventive intermittent treatment for pregnant women also need to be addressed. A protocol to determine the rates of malaria parasitemia and anemia in women attending antenatal clinics and placental parasitemia and low-birthweight babies among women delivering in health care facilities

is currently under development by the World Health Organization and the Centers for Disease Control and Prevention (M. Parise, Centers for Disease Control and Prevention, unpublished document, 2002). This protocol includes a rapid assessment guide to gather behavioral data related to sociobehavioral components of malaria and pregnancy. Although the protocol has been used only in nonemergency situations, it could be adapted for use in the more stable postemergency phase of displacement.

• Malaria prevalence surveys, therapy efficacy assessments, entomological surveys, and behavioral research should be considered essential routine activities to inform programmatic decisions.

• Findings of operational research should be documented in order to improve evidence-based humanitarian relief practice.