Diet and alcohol consumption in the general population; preliminary findings - includes related information

Alcohol Health & Research World, Summer, 1989 by Sharon Brooks, Gerald Williams, Frederick Stinson, Mary C. Dufour

Diet and Alcohol Consumption in the General Population

Preliminary Findings

All cells in the body need a source of energy to accomplish their specific functions. Adequate intake of basic nutrients provides the energy necessary to maintain the body's physical activity and general health. Food supplies the essential materials (proteins, carbohydrates, and fats) from which energy is produced and from which new molecules are assembled to support body activity (Spencer and Mason 1983). Nutrition is the process by which ingested foods produce such energy.

The most important nutritional disturbances involve either a shortage or an excess of the dietary building blocks that provide both energy and materials for the synthesis of proteins, carbohydrates, and fats (Walker 1982). Malnutrition--a shortage of these building blocks--can lead to various health problems or increase one's susceptibility to illness. High levels of alcohol consumption can result in primary malnutrition (inadequate dietary intake) and secondary malnutrition (inappropriate digestion, absortion, transport, storage, metabolism, or elimination of nutrients) (World et al. 1985). If alcohol consumption is related to malnutrition and if malnutrition results in various health problems, then excessive alcohol consumption can contribute to decreased states of wellness, in addition to its direct linkages with some diseases, such as cirrhosis and pancreatitis (see the articles by Rothschild, et al., on pp. 229-231, and by Korsten, on pp.232-237, in this issue).

Much attention has been focused on the relationship between nutrition and alcohol. Many studies have concentrated on the nutritional status of the alcoholic, ignoring other types of drinkers. The remaining research, which has focused on different types of drinkers, has been concerned with the average calories consumed from carbohydrates, proteins, and fats versus the average calories consumed from alcohol. Some studies have even demonstrated that skipping meals and eating less food--markers of nutrient intake--vary across drinking levels (Hillers and Massey 1985; Jones et al. 1982; Bebb et al. 1971). In a previous report (Williams et al. 1986) from NIAAA's Alcohol Epidemiologic Data System (AEDS), analyses of data from the Health Promotion and Disease Prevention Questionnaire of the 1985 National Health Interview Survey (NHIS) indicated that heavier drinkers are more likely than light or moderate drinkers to report that they "rarely or never" eat breakfast and that they "rarely or never" snack. In general, the literature indicates that excessive alcohol consumption often is associated with poor diet and may lead to malnutrition.

This Epidemiologic Bulletin presents some preliminary findings on whether certain dietary beliefs and behaviors are associated with particular levels of drinking in the general adult population. The data reported here come from the National Cancer Institute (NCI) Epidemiology Study, which was a part of the 1987 NHIS (see sidebar for a brief description of this study).

1987 Drinking Patterns

Table 1 presents age- and sex-specific distributions for abstainers as well as for light, moderate, and heavier drinkers (see sidebar for an explanation of these drinking level categories). Excluded from analyses presented here were respondents with missing or unknown responses to questions concerning number of drinks consumed for all three alcoholic beverage types (beer, wine, and liquor). These exclusions resulted in a sample size of 21,826.

The data in Table 1 are percentages based on weighted survey data. The NHIS sample is a multistage probability sample of the general household population, which oversamples racial/ethnic minorities and older persons to ensure reliability of estimates for these subgroups. It is necessary to weight differentially the data from different respondent types in order to adjust for this oversampling and to adjust for unequal probabilities of sample selection when developing estimates intended to be representative of the general household population. Appropriate weights for such adjustments are provided in the 1987 NCI Epidemiology Study and were used by AEDS in developing the estimates presented here.

Consistent with AEDS findings from 1983 NHIS data (Malin et al. 1985/86), the proportion of abstainers increases with age; the proportions of light, moderate, and heavier drinkers decrease with age. Females are more likely to be abstainers or light drinkers, and males are more likely to be heavier drinkers.

Diet and Alcohol Consumption

In preliminary analyses of the 1987 NCI Epidemiology Study addressing the relationship between certain dietary beliefs and behaviors, AEDS examined the following six variables, which appear to have face validity as positive or negative indicators of respondents' awareness of the importance of diet for health: * believing that diet has little effect on

disease * eating fewer than two meals per day

on weekdays * reporting a diet low in fiber * reporting a diet high in fat * making lasting and major changes in