This is the old United Nations University website. Visit the new site at http://unu.edu

Analysis of the reconnaissance project. Phase 1: The community

Abstract
Introduction
This was not KAP
Reconnaissance is more than RAP
Questionnaires used during the study of the community
Informants to be consulted during the study of the community
Summary of the first phase of the Reconnaissance project
Conclusions
Comments
Summary
References

Husni Thamrin, Asri Rasad, Noel W. Solomons, Mark L. Wahlqvist, and Rainer Gross

Husni Thamrin and Rainer Gross are affiliated with the Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ) in Jakarta, Indonesia. Asri Rasad is with the SEAMEO-TROPMED Center in Jakarta. Noel Solomons is affiliated with the Center for Studies of Sensory Impairment, Aging and Metabolism (CeSSIAM) in Guatemala City, Guatemala. Mark Wahlqvist is with the Department of Medicine at Monash University in Melbourne, Victoria, Australia.

Abstract

The Reconnaissance project, Nutritional and Food Security Situation of Rural and Urban Elderly from Selected East Asian and Latin American Developing Countries, was carried out in eight developing countries. The purpose of the project was to validate the protocol and identify potential problems before the full project was begun. In March 1993 a workshop was held in Wageningen, Netherlands, for participants from the countries and other interested professionals to analyse the results of the Reconnaissance in preparation for the initiation of the full study. The Reconnaissance consisted of two phases: the first studied the community and the second the individual. This paper summarizes the study of the communities through group discussions and semi-structured interviews of community representatives. The information obtained was used to revise and improve the protocol to be used when the full study, Cross-Cultural Research on the Nutrition of Older Subjects (CRONOS), is undertaken.

Introduction

Over the past five years, interest in urban populations and in the aged has grown. Concurrently, in Asia and Latin America, two regions of the developing world in which rural problems and maternal and child health have been the dominant paradigms in nutrition, interest in urbanization and the elderly has increased. The Reconnaissance project on the Nutritional and Food Security Situation of Rural and Urban Elderly from Selected East Asian and Latin American Developing Countries is an outcome of this interest. The project was designed to be conducted as a multicentre study in three Latin American and five Asian countries, using a common protocol. The logistics of involving eight distinct, diverse, and geographically dispersed countries in the development of each element of the protocol- interview, questionnaire, or measurement-were too unwieldy. Consequently, the protocol was written in one site (Indonesia) and circulated to the other centres.

The designers of the Reconnaissance project sought guidance from two previous models of multicentre studies on nutrition and ageing: the EURONU-SENECA study, Nutrition and the Elderly in Europe [1, 2], and the study by the Committee on Nutrition and Aging of the International Union of Nutritional Sciences (IUNS) entitled Food Habits in Later Life (FHLL) [3, 4]. The designers included as many of the physical measurement variables and survey questions from these studies as possible so that a common database could be established and expanded with additional countries.

In June 1992 the proposers of the Reconnaissance project travelled to Boston, Mass., USA, to attend the Book-Writing Meeting of the IUNS Committee on Nutrition and Aging to analyse the data and experience of the IUNS-FHLL project. The meeting was also attended by representatives of European institutions that had been part of the EURONUTSENECA project (Wageningen University) or were involved in the IUNS-FHLL project (the Philippines), as well as from other countries where study sites were located (Indonesia, China, and Guatemala). This allowed for an exchange of experiences among the investigators.

The project of the IUNS Committee was to develop a general protocol that could be used in different sites. The days in Boston were devoted to the serious analysis of the experience of IUNS-FHLL, with data from Australia, China, and Sweden. Ellen Messer, a cultural anthropologist, gave her perspective on some of the inconsistencies in the IUNS-FHLL process that were recognized at the Boston meeting. She suggested that the use of rapid assessment procedures (RAP) before beginning the IUNS-FHLL study might have avoided many confounding issues and problems in interpretation. Nevin Scrimshaw, host of the meeting, gave strong support to her comments. The collection of field data for IUNS-FHLL was complete by June 1992, and nothing could be done in retrospect to implement a baseline rapid assessment. But the eight-country Reconnaissance process was just beginning. So, it was suggested by the assembly, and accepted by the framers of the Reconnaissance, that a rapid qualitative analysis component [5] should be a prelude to the definitive protocol measurements.

In the final design, the Reconnaissance project proceeded in two sequential phases. In the first phase the community as a whole was the unit of study, and in the second phase the individuals in the community were the units. The community phase was added in an effort to avoid some of the pitfalls encountered by the organizers of the other large, multicentre surveys of nutrition of the elderly by the EURONUTSENECA study and the IONS Committee.

This was not KAP

The knowledge, attitudes, and practices procedure (KAP) is a semi-quantitative approach that can use either individuals or the community as the unit of analysis [6]. It is a rapid and fairly structured method. In fact, KAP would not be appropriate in the present context, since it is most useful in assessing how a population confronts a specific health problem or how it might respond to a new intervention. For example, if one went to an area in which a specific parasite was endemic, such as Onchocerca volvulus, the agent of river blindness, it might be appropriate, using KAP, to determine what the population knew about the vector, about the worm, and about the disease in scientifically sound terms (knowledge). Then one would determine their degree of concern about the prevention and treatment of the disease and what they thought about certain hypotheses (attitudes). Finally, it would examine what they were currently doing-or not doing- about the problem (practices). In fact, studies with these components have been conducted on river blindness [7, 8].

In the KAP, one surveys a cogent and coherent set of responses related to a concrete problem. In the Reconnaissance study, the variables were very diverse, and no specific intervention or set of interventions was ever contemplated as an immediate result of the research.

Reconnaissance is more than RAP

Rapid Assessment Procedures or RAP are a qualitative method that uses the household or community as the unit of analysis [5, 9-11]. The approach used in the Reconnaissance project was not RAP. The elements of conventional RAP, as they have evolved, have been well summarized by Herman and Bentley [11]:

Rapid assessment procedures (RAP) apply a set of common anthropological techniques to specific health issues. Because only health issues are addressed and the number of informants is limited, RAP is relatively fast and inexpensive. Moreover, lengthy statistical analysis is not needed, so that the data are available immediately for decision making.

Alone, however, they [formal surveys] are often insufficient in providing the type of in-depth information required to understand the complexity of human behaviour and to formulate prevention and control strategies and programmes.... qualitative information needs to be generated on the same topics in a way which permits a more detailed, a more rapid, and often, a more accurate understanding of the underlying social and cultural characteristics influencing or associated with specific patterns of behaviour.

In essence, it was decided that for research on health-seeking behaviour, such as the Reconnaissance project, basic anthropological field methods, such as formal interviews, conversations, observations, participant observations, and focus groups, should be used. The Reconnaissance phase involved the collection of secondary and statistical data for the initial assessment of the nutritional situation of the group living in the defined geographic area. Qualitative information, such as opinions and perception of health and nutrition, as well as perceptions of the elderly and their roles in the community, was to be collected before measuring the quantitative variables. This activity finally became a qualitative survey directed towards refining a structured set of variables. Its purpose was to decide what to measure and how to measure it efficiently and sensitively. For example, some questions might be superfluous and offensive, such as a question about the religion of a respondent in a Malaysian or Indonesian ethnic group already known to be Moslem. However, this question would be relevant in Guatemala, formerly a largely Roman Catholic country in which major portions of the population have undergone conversion by evangelical Protestants.

Alternatively, a question might be poorly understood and therefore answered with difficulty or answered inconsistently from one country to another or from one stratum to another within a country. To ensure that the same question is asked in each language, back-to-back translation should be done, i.e., translation into the local language and then back into the original language by two different bilingual individuals.

One approach to the strict validation of the language of a question or the acceptability of a measurement would be to present them before key informants or focus groups.

Questionnaires used during the study of the community

In general, the questionnaires used in the first phase of the Reconnaissance project, during which the community was the unit of analysis, related to norms and values and the general situation of the community. Topics covered included the demographic structure, the general living conditions in the community, the situation of the economically active population, the agricultural situation, access to schools and the school situation, the living conditions, the situation of women, the nutritional and health situation, the religious situation, and the health service situation. For each topic, detailed information was collected from each study site by interviews with informants or by focus group discussions.

Demography

The demographic structure of the community was determined by gathering information on population size, family size, ethnic group, spoken and written languages, religion, and other organized groups, such as clubs for housewives or Boy Scouts.

General living conditions

Information on general living conditions in the community was gathered by questions about changes in the community during the last 10 to 20 years, major problems in the community, major problems in living together in the community and family, income, migration to and from the community, and infrastructure in the community. Sample questions included: What are the major problems of the community? What are the major problems of the household? Are there opportunities to earn money? What? Do family dependents live in another place? If yes, where? How is a person defined as old? What is the status of an old man or woman?

Agriculture

Information on the agricultural situation was gathered by questions about cultivation patterns (food and cash crops), land ownership and access to land, major agricultural problems, rating of agricultural advisory services, rating of the credit system (if any), availability of food at the community and household levels (home-grown or purchased food, seasonality, food storage), and food price movements over recent years. Sample questions included: What is the ownership structure of farmed arable land? Must any rent be paid? If so, how much? What are the main forms of agriculture? Yield? Diseases? Fertilizers? Main demand for labour (men/women, seasonal changes)? Marketability? What are the most important types of livestock? Earnings? Diseases? Feeding? What is the main demand for labour (men, women, seasonal changes)? Marketability? Is there access to the credit system, with equal opportunity for men and women? How has it worked in practice? Can some improvement be made to the credit system? Is there access to an agricultural advisory system? How has it worked in practice? How often does an advisor turn up? Can some improvement be made to the advisory system?

Status of women

Information on the status and general situation of women was gathered by questions about marriage age, ownership rights for women, the proportion of women owning land, the proportion of women engaged in agricultural work and paid employment, rating of the workload of the women, opinions as to whether women are adequately represented in society, and the status of women (girls, young women, mothers, grandmothers, women living alone, etc.). Sample questions asked of women included: What is your daily schedule? How much of your workload involves getting water and fuel? How much agricultural work do you do? What are your other income generating activities? What is the relative income distribution between men and women? Who has control over the income? Who decides upon purchases? Who looks after the children most of the time? How much time is needed by the children to walk to school? At what age do girls or young women usually marry? How many children do women usually have? If you had another child, would you like a boy or a girl? How many children per family die on average? What is the ideal number of children? How many boys and how many girls? Are there possibilities for family planning?

Nutrition

Information on the nutritional situation was gathered by questions about the availability of food at the community and household level (home-produced or purchased food, seasonality, food storage), changes in food prices over the past year, views on the distribution of food within the family, rating of the food programme (if any), rating of breastfeeding and infant nutrition, rating of nutrition for pregnant women and nursing mothers, the proportion of undernourished children, and nutrition of the elderly. Sample questions included: What are the staple foods? Are the staple foods always available? If not, during which months are they unavailable? Why are they unavailable? Is food stored? Which food? Where does the food come from (produced at home, purchased, where purchased)? What is the nutritional condition of children? Are the elderly properly fed? If not, why not? Can some improvements be made in the nutrition of the elderly? Are there foods unsuitable for elderly people? If yes, why? How many meals does an elderly person have daily? Do men and women need different food and different care? What is the nutritional practice for elders suffering from a disease? What food is eaten at parties and celebrations?

Health

Information on the health situation was gathered by questions about major illnesses, major causes of death, rating of the health system (governmental and traditional), and the position and rating of the family planning programme. More detailed questions on this topic were asked of the health representatives, such as access to health centres, the longest time and distance to the health centre, the number of persons cared for, the daily visiting rate (infants, schoolchildren, women, men, the elderly), the nearest hospital, the supply of medicine, cooperation between health centres or posts and traditional midwives and healers, rating of the immunization programmes, and community participation. Questions were also asked about the environment, such as water supply, sewage, and refuse disposal. Sample questions discussed with the group included: What is the ideal number of children for a family? Of this number, how many are boys and how many girls? At what age should a woman have children, and until what age? How does one know whether an old person is healthy or sick? How is illness diagnosed in an elder? What are the major illnesses suffered by older men and women in the community? What are their symptoms, incidences, causes, and treatments? How far is it to the health service? What is expected from a good health service? Does this apply to the local health service? Can some improvements be made to the health service? How often are personnel (health assistants, nurses, doctors, etc.) on duty at the health stations (posts, centres)? Do the members of the community visit the health service frequently enough? If not, why do people not visit the health service more often? How much does treatment at the health service cost? Is the price too high? At what age should an older woman or man have a checkup at the health post, and how often? Is there health insurance?

Economics

Information on the situation of the economically active population (by age and sex) was gathered by questions about economic activities (main types, major employers, unemployment and underemployment, daily wages, seasonality, sex specificity, rural or urban) and migration pattern (seasonality, sex specificity).

Education

Information on the educational situation was gathered by questions about access to schools, literacy rate (sex specific), formal or informal education and further schooling, rating of schoolwork and the school situation, and number of schoolchildren (schooling rate, dropout rate, absence, reasons for dropout and absence).

Living conditions

Information on living conditions was gathered by questions about the size of the house, the number of bedrooms, the location of the kitchen, building materials, drinking water supply, energy supply, and sewage and refuse disposal.

Informants to be consulted during the study of the community

Key informants

Qualitative information was gathered by semi-structured interviews with key informants, such as the mayor and community leaders, teachers and religious leaders, agricultural representatives or field workers, and health workers. At least two representatives in each community were asked the same questions.

Focus groups

To get more qualitative information on the community, group discussions with community members were also included in the first phase of the Reconnaissance, in addition to the semi-structured interviews with informants. The community members involved in the group discussions included informal leaders, adults, youths, and some elderly people. The topics to be discussed by the group were a subset of those for key informants: general living conditions, food, health, agriculture, and the situation of women.

Summary of the first phase of the Reconnaissance project

Semi-structured interviews and group discussions based on the Reconnaissance manual were held in all eight countries. Table 1 summarizes elements of the semi-quantitative survey in the eight countries and of the group discussions in four Asian sites (China, Indonesia, Malaysia, and the Philippines). Reports from the Latin American countries and from Thailand failed to mention in detail the elements of the group discussions at the community level.

The rural participants were from a single village in some of the countries and from as many as four villages in others. In all of the countries, both middle income and low-income urban participants were from the same metropolitan area. There was little consistency from country to country in the types of individuals interviewed. In each country for which information was available, community leaders were interviewed. In addition, some or all of the following types of people were interviewed: health service personnel including midwives, agricultural workers, religious leaders, teachers and other educational authorities, priests and other religious leaders, and nutrition professionals. Very little information was provided about the composition of the focus groups. In the Philippines separate groups of rural women and rural elderly people were interviewed.

Findings and interpretation

At a workshop in Wageningen, the results of the first phase of the Reconnaissance exercise, in which the community was the unit of analysis, were presented by representatives from each of the participating countries. No systematic attempt has been made here to analyse the data from these exercises. However, some generalizations can be made.

Inter-site consistency

There were inconsistencies from site to site with respect to the application of the instrument used to study the community as the unit of analysis. In all countries except Guatemala, the rural setting was chosen a priori, and the qualitative, rapid instrument was applied. In Guatemala the qualitative pre-study was performed to assess the suitability of different rural sites for inclusion. Three sites were assessed and one was chosen on the basis of the results.

Each country differed with respect to the numbers and types of key informants and focus group participants. The complete panel was not used in any region.

In the Philippines the Reconnaissance studies were performed by different teams in the urban and rural areas. However, the planning and implementation of both studies were undertaken with strict inter-institutional coordination.

Attitudes of individual investigators

The attitudes of the various investigators who participated in the Reconnaissance project were revealed either implicitly or explicitly in the presentations made at the workshop. The participants at Wageningen included the senior principal investigators, the managing field investigators (if they were not the same as the principal investigators), and the consultants. The attitudes of the participants toward the procedures of the Reconnaissance project were dependent on whether they had been in the original deliberations in Boston when the decision was made to include the community-level qualitative survey (associated with generally positive attitudes), whether they had been a senior partner or a responsible principal investigator at a given site (also associated with generally positive attitudes), and whether they had been personally responsible for administering the questionnaire (associated with the least positive attitudes).

TABLE 1. Summary of the first phase of the Reconnaissance project by country

Country	Location		Individuals interviewed	Groups interviewed
	Rural	Urban
China	Bei-Hi village, Shun-yi province	Beijing: Xi-Cheng (middle-income), Chong-Wen (low-income)	Health service personnel, community leaders	Adults, elderly
Indonesia	Sad Mantra, West Java	Jakarta: Harapan Mulya (middle-income), Kemayoran (low- income)	Community leaders, health representatives, agricultural workers, religious leaders, education authorities, teachers	Informal community leaders, adults, young men, older persons
Malaysia	Rural area near Kuala Lumpur	Kuala Lumpur	Community leaders	Population groups
Philippines	San Isidro, Baao	Barangay Santol in district IV of Quezon City	Rural: leader, agriculturist, priest, teacher, nutrition scholar, midwife Urban: Community leaders, health service workers, teachers, religious leaders	Rural: women and elderly (separately) Urban: no details
Thailand	4 villages in Khon Kaen	Bang Kane district of Bangkok	None	None
Brazil	Caetano and Capoeira, Bahai State	São Paulo county: Vila Madalena suburb (middle-income), Sao Remo shanty town (low-income)	2 community leaders at each site	No details
Guatemala	3 villages near Guatemala City (San Pedro Ayampuc, San Juan, and Villa Canales)	None	No details	No details
Mexico	Solis Valley in Central Mexico	Mexico City: Colonia Guerrero (middle-income), Colonia Isidro Fabela (low-income)	No details	No details

The investigators had generally been trained as nutritionists with a background in dietetics or biochemistry or as physicians with postgraduate specialization in nutrition or gerontology. It was the general impression that the non-physicians had a more positive attitude-both in the execution and in the discussion of the results-and more readily embraced the qualitative instrument designed to study the community. Physicians had been skeptical of the process during the survey and were unconvinced of its value in the collective evaluation.

Conclusions

Some consensus was achieved as to the perceived value of a rapid assessment application in a multicentre study such as this one. It was felt that if it were to be done, it should be done in the initial planning stage, before a quantitative protocol had been finalized, to allow feedback in the design of questions and procedures. Everyone was in accord as to the need to have 100% of the sites record all core data so as not to defeat the purpose of a multicultural approach.

Comments

The opportunity to perform a semi-quantitative, community-level Reconnaissance provided some unanticipated insights, both about the communities to be enrolled and about the attitudes and aptitudes of the investigators. It is clear that a multicentre study must have a core of variables that are standard for all participating sites; it is also evident that in cross cultural studies, the rationale for the investigation is set by the differences across cultures. On the way to the pilot collection of specified standard measures, each of the eight centres performed a qualitative, structured rapid assessment in the communities.

Rapid assessment methods emerged as a consequence of anthropological theory [12, 13] and have recently attracted their adherents and proponents [5, 9-11]. Given the extent of their discussion, it behooves nutritionists to become familiar with them, and a project such as the Reconnaissance offered an opportunity for researchers at many levels to gain direct experience with a rapid, qualitative, community assessment tool, albeit one that was constructed for the present endeavour and one that does not have the pedigree of KAP, RAP, RRA (rapid rural appraisal), or PRA (participatory rural appraisal).

Rapid methods are a prelude to structured, detailed research but not a substitute for it. In general, both KAP and RAP are data-gathering efforts (research) that set the stage for actions (interventions). Clearly, one wants both culturally sensitive and valid interpretable measurements in a survey design, and a qualitative method may allow the investigators to hone their questions and structure their measurements to afford the greatest precision, clarity, and acceptance.

Even within a given country, attitudes and responses to different questionnaires and physical measurements may vary from rural to urban populations and across age or socioeconomic classes within populations. This is a challenge to the design of a multicentre study. The purist of multicentre research will insist that the protocols be uniform across sites and that the measures be made in a standardized and uniform manner. However, this may be quite unrealistic.

The entire dilemma about refining the measures in a common, multicentre protocol, across diverse linguistic groups and cultures from Asia to Latin America, collides with the fundamental reason why a cross-cultural study is attractive. In the case of ageing, it separates what is essential to senescence from that which might be situational to a given population or environment [14]. Clearly, if a valid or acceptable way to obtain information in one or more of the sites of a cross-cultural study cannot be found, one loses strength and dimension for that measurement. Its absence weakens the robustness of the collaborative network findings. Given time and cost considerations in mounting a study with a common protocol and interactive purpose among various cultures and nations, however, a certain assurance that each of the designated countries will be able to provide the data-whether it be physical, biochemical, or verbal -is necessary.

Summary

Qualitative appraisal has been conducted in the multicentre study, Nutritional and Food Security Situation of Rural and Urban Elderly from Selected East Asian and Latin American Developing Countries. The countries involved were China, Indonesia, Malaysia, the Philippines, Thailand, Brazil, Guatemala, and Mexico. Semi-structured interviews were con ducted with community representatives or informants in all countries except Thailand. However, only the reports from Indonesia and the Philippines specifically discussed the semi-structured interviews with community representatives. Group discussions with community members were also conducted in China, Indonesia, Malaysia, and the Philippines. The reports from Thailand, Brazil, Guatemala, and Mexico did not mention this activity.

References

1. de Groot LCPGM, van Staveren WA, Hautvast JGAJ, eds. EURONUT-SENECA. Nutrition and the elderly in Europe. Eur J Clin Nutr 1991;45(suppl 3):1-196.

2. de Groot LCPGM, van Staveren WA, Hautvast JGAJ. Experience of the EURONUT-SENECA study in design, implementation, and data analysis. Food Nutr Bull 1997;18:235-8.

3. Hsu-Hage BH-H, Wahlqvist ML. The IUNS study of Food Habits in Later Life: a critique. Food Nutr Bull 1997;18:239-47.

4. Wahlqvist ML, Davies L, Hsu-Hage BH-H, KourisBlazos A, Scrimshaw NS, Steen B. van Staveren WA, eds. Food habits in later life: descriptions of elderly communities and lessons learned. Jointly published on CD-ROM by the United Nations University Press, Tokyo, and the Asia Pacific Journal of Clinical Nutrition, 1996.

5. Scrimshaw SCM, Hurtado E. Rapid assessment procedures for nutrition and primary health care. Los Angeles, Calif, USA: Center for Latin American Studies, University of California Los Angeles, 1987.

6. Tyler SA, ed. Cognitive anthropology. New York: Holt, Rinehart, and Winston, 1969.

7. Ongore D, Kamunvi F. Knight R. Minawa A. A study of knowledge, attitudes and practices (KAP) of a rural community on malaria and the mosquito vector. East Afr Med J 1989;66:79-90.

8. Richards F. Klein RE, Gonzalez-Peralta C, Zea-Flores R. Zea-Flores G. Castro-Ramirez C. Knowledge, attitudes and perceptions (KAP) of onchocerciasis: a survey among residents in an endemic area in Guatemala targeted for mass chemotherapy with ivermectin. Soc Sci Med 1991;32:1275-81.

9. Scrimshaw SCM. Adaptation of anthropological methodologies to rapid assessment of nutrition and primary health care. In: Scrimshaw NS, Gleason GR, eds. Rapid assessment procedures: qualitative methodologies for planning and evaluation of health related programmes. Boston, Mass, USA: International Nutrition Foundation for Developing Countries, 1992:2539.

10. Scrimshaw NS, Gleason GR, eds. Rapid assessment procedures: qualitative methodologies for planning and evaluation of health related programmes. Boston, Mass, USA: International Nutrition Foundation for Developing Countries, 1992.

11. Herman E, Bentley M. Rapid assessment procedures (RAP) to improve the household management of diarrhea. Boston, Mass, USA: International Nutrition Foundation for Developing Countries, 1993.

12. Pelto PJ, Pelto GH, eds. Anthropological research: the structure of inquiry. New York: Cambridge University Press, 1978.

13. DeWalt BR, Pelto PJ, eds. Micro and macro levels of analysis in anthropology: issues in theory and research. Boulder, Col, USA: Westview Press, 1985.

14. Dieck M. Cross-national research and international comparative research in gerontology: a preliminary study. Berlin: Deutsches Zentrum für Altersfragen, 1984.