This is the old United Nations University website. Visit the new site at http://unu.edu
Abstract
Introduction
The examination process
Issues for consideration when revising individual questions
Conclusions
References
Geok Lin Khor, Panata Migasena, Keyou Ge, Rainer Gross, Adriane Lacle, and Annhild Kvalbein
Geok Lin Khor is a professor in the Department of Nutrition and Community Health in Selangor Darul Ichsan, Malaysia. Panata Migasena is a professor in the Department of Tropical Nutrition and Food Sciences in the Faculty of Tropical Medicine at Mahidol University in Bangkok. Keyou Ge is the director of the Institute of Nutrition and Food Hygiene in the Department of Nutrition and Community Health of the Chinese Academy of Preventive Medicine in Beijing. Rainer Gross is senior advisor to the Regional Community Nutrition Program of the South-East Asian Ministries of Education Organization (SEAMED) at the University of Indonesia, Jakarta, for the Deutsche Gesellschaft fur Technische Zusammenarbeit (GTZ). Adriane Lacle is the director of the Project on Research in the Elderly in the National Institute for Health Research (INISA) at the University of Costa Rica in San Jose. Ann-hild Kvalbein was formerly an exchange student at the University of Jakarta from the Nutrition Department of the University of Oslo.
After the experiences at the individual sites during the Reconnaissance project had been presented in a plenary format, five groups were formed. Four of these groups examined the individual questions on the original questionnaire with specific attention to (1) nutritional status and biological variables, (2) food intake and food security variables, (3) health status assessment variables, and (4) lifestyle, socio-demographic, and social behaviour and practices variables. In this process, the groups identified items on the questionnaire that had been found, at one or another site, to be of doubtful appropriateness or feasibility. The area of concern of the fifth group was the sampling frame and selection procedures for sites and individuals within sites.
The participants in the process of developing a protocol for the CRONOS project (Cross-Cultural Research on the Nutrition of Older Subjects) follow a recent tradition set in motion by the EURONUT-SENECA study [1] and the study Food Habits in Later Life (FHLL) of the International Union of Nutritional Sciences (IUNS) [2]. This tradition is to develop a coalition of investigators at diverse geographic sites to undertake a multicentre study on health and nutrition of the elderly. The age limits here are lower than the 75-year minimum age for the EURONUT-SENECA study and the 70-year minimum for the IUNS study. Moreover, the eight countries of the CRONOS study are all transitional or pre-industrial countries of the so-called third world.
As originally set up, the protocol of the study on the Nutritional and Food Security Situation of Rural and Urban Elderly from Selected East Asian and Latin American Developing Countries was conceived as a common protocol for a definitive study at nine centres in eight countries, five in Asia and three in Latin America. The availability of only partial funding allowed only for a feasibility study (the Reconnaissance project) of 24 persons at each site, except for 36 persons in the Philippines [3]. The present series of papers recounts the analysis of that experience. This analysis led to a substantially revised- and presumably improved-version of the protocol. The present paper details the results of the individual working groups that dealt with the appropriateness and feasibility of the interview questions and the physical and laboratory measurements.
At the meeting in Wageningen, Netherlands, the experiences of each country at both the community and the individual levels were presented in a plenary session [4]. Problems that arose in interviewing and taking measurements were discussed. This evaluation was an opportunity to improve the protocol and procedures to be used in the definitive phase of the study. Five working groups were formed to consider nutritional status and biological variables; food intake and food security variables; health status assessment variables; lifestyle, socio-demographic, and social behaviour and practices variables; and the sampling frame and selection procedures for sites and individuals within sites.
The first four working groups had a mandate that was specifically relevant to the present paper, namely, the analysis of the appropriateness and feasibility of the interview questions and measures.
Of the 104 original questions in the Reconnaissance manual, a series of specific questions was singled out as questionable by one or another of the working groups in their deliberations on the experience in the pilot phase.
Demographic and socio-economic variables
China
It is not possible to circumscribe a geographic area or neighbourhood in Beijing that would be considered “low-income” in contrast to another considered “middle-income.” It is difficult within the assumptions of socialism to admit to the existence of fixed classes, and it is the contention of the investigators that no “slum” settings existed and that persons of different incomes lived side-by-side in residential areas.
Malaysia
There are clearly residential class differences, but the government provides low-income housing. Rather than seek persons in the truly marginal and precarious conditions as their “poor” group, the investigators chose to study the “settled poor.’
Options
Change the criterion for enrolment from one based on an area of residence to one based on the family or make the selection method flexible, adapted to the specific conditions of the country.
Other observations
There was an assumption by the designers of the Reconnaissance manual that the lower-class urban community would be the urban extension of a rural region from which migrants flowed into the city, which was the case in Brazil. However, in some situations, the rural persons were from communities close to and convenient to the metropolitan area under study.
Another consideration is the tendency for younger urban adults to have migrated much more in their lifetime than the older inhabitants.
Because different individuals administer the questionnaires, there is a possibility that the connotations of the questions could differ, leading to inconsistency in the responses. As a control for the stability of interview responses, a random 10% repetition by another interviewer was recommended. However, in China the elderly subjects in the pilot study were reluctant to be interviewed more than once.
Income variables
Philippines and China
The informants were found to be sensitive about questions relating to income. The informants provided answers when pressed, but some investigators were skeptical about the reliability of the data provided.
Living conditions variables
Philippines
Some of the subjects were sensitive to questions about their housing conditions.
Social behaviour
The EURONUT-SENECA study
The organizers realized that the questions about activity patterns missed information about elderly people who were still working. There were no questions about specific types of activity. As many persons over 60 years of age still are employed or active in agriculture, information about activity may be biased or distorted unless more detailed information about activity patterns is obtained.
Other observations
The questions about social contacts and cultural events seemed to have some Western bias. For instance, many elderly people who work have social contact on the job. It is not necessary to conceive of each elder as a potentially isolated individual living at home. Also, cultural events may be common in some locations and rare in others. In rural Mexico, for instance, respondents reported cultural events, as they interpreted the expression, to be infrequent.
If other family members are present during the interview, people may be reluctant to give honest answers to some questions, especially questions about personal problems or “emotional closeness,” for fear of offending or embarrassing them. This is well known to anthropologists, who try to avoid such situations.
Self-perceived health variables
The question about “climbing stairs” is irrelevant in rural areas where there are no stairs.
Food intake and food security variables
Older people may have difficulty remembering their food intake during the last 24 hours. In the pilot study we found some interesting variants, such as Mexicans who preferred to impress the interviewers by reporting their “best” meal of the past week and Filipinos who seemed reluctant to discuss their dietary intake.
Administration of the interview and examination
The willingness of the subjects to expose bare skin for skinfold or circumference measurements is a concern. In Islamic cultures such as Malaysia, exposing the body is a taboo. In Mexico and Guatemala, on the other hand, subjects showed no reluctance to expose the body for measurement, even by examiners of the opposite sex. In Indonesia the subjects were reluctant to accept the clamp of the skinfold calipers.
A diverse array of perceived problems was encountered when the original Reconnaissance protocol (Nutritional and Food Security Situation of Rural and Urban Elderly from Selected East Asian and Latin American Developing Countries) was applied. These observations led to the revisions that were incorporated into the CRONOS protocol [5]. To some extent, although unintended, this process has been blessed by the luxury of an opportunity for reflection and reassessment before definitive generation of data from a multicentre protocol. It points out that many erroneous procedures can be incorporated into a multicentre study if pretesting is not allowed. It is expected that this experience will make the protocol a better guide for research to provide insights into the process of ageing and urbanization in the developing world.
1. 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.
2. Hsu-Hage BH-H, Wahlqvist ML. The IUNS study of Food Habits in Later Life: a critique. Food Nutr Bull 1997;18:23947.
3. Barba CVC, Rabuco LB. Overview of ageing, urbanization, and nutrition in developing countries and the development of the Reconnaissance project Food Nutr Bull 1997;18:220-5.
4. Thamrin H. Rasad A, Solomons NW, Wahlqvist ML, Gross R. Analysis of the Reconnaissance project. Phase 1: The community. Food Nutr Bull 1997;18:248-55.
5. Gross R. ed. CRONOS (Cross-Cultural Research on the Nutrition of Older Subjects). 3rd ed. Food Nutr Bull 1997;18:267-303.
