|Alcohol consumption and pregnancy in the Mexican national addiction survey |
Consumo de bebidas alcohólicas y embarazo en la encuesta nacional de adicciones, México
|1 Instituto Mexicano de Psiquiatría. Calzada México Xochimilco 101, Col. San Lorenzo Huipulco, C. P. 14370, México D.F., México. Fax: (525) 513-3446. firstname.lastname@example.org |
2 Dirección General de Epidemiología, Francisco de P. Miranda 177, 4o piso, Col. Merced Gómez, Lomas de Plateros, C. P. 01480 México D.F., México.
3 Instituto Nacional de Salud Pública, Av. Universidad 655 Col. Santa Maria Ahuacatitlan, C. P. 62508, Cuernavaca, Morelos, México
4 Hospital General de México, Calle Dr. Balmis 148 Col. Doctores, Unidad 204, Epidemiología Clínica, Edificio de Estudios Especiales, C. P. 06720, México D.F., México.
|Abstract In 1988, the General Directorate of Epidemiology and the Mexican Institute of Psychiatry conducted the first National Addiction Survey (ENA), providing regional and national data on alcohol, tobacco, and drug use. The ENA providing a subsample of women who have been pregnant at some time in their lives. There were 5,234 affirmative responses. Women were asked if they had suffered any of three adverse outcomes during their last pregnancy: spontaneous abortion, stillbirth, and congenital abnormalities. Prevalence of spontaneous abortion was 3.8%, stillbirth 1.2%, and congenital abnormalities 1.1 %. Multiple logistic-regression models were used to analyze the effect of alcohol consumption on these problems. Consumption during pregnancy was related only with the prevalence of congenital abnormalities, with prevalence odds of 3.4. Among habitual users during the last 12 months, oniy women in the highest use category showed an important relationship with the three problems mentioned. Follow-up studies on the Mexican population are recommended in order to obtain more conclusive findings. |
Key words Pregnancy; Alcohol Consumption; Spontaneous Abortion, Stillbirth; Congenital Abnormalities
Resumo En l988 la Dirección General de Epidemiología y el Instituto Mexicano de Psiquiatría llevaron a cabo la primera Encuesta Nacional de Adicciones (ENA), que proporciona información a nível nacional y regional sobre el uso de alcohol, tabaco y drogas ilícitas. La ENA incluye una submuestra de mujeres que se embarazaron alguma vez en la vida. Se obtuvieron 5.234 respuestas afirmativas a la pregunta de se alguma vez se habían embarazado. A este grupo se les preguntó si en su último embarazo tuvieron un aborto espontáneo, un bebe nacido muerto o un bebe con anomalías congênitas. La prevalencia de aborto espontáneo fue de 3,8%, 1,2% de nascidos muertos y 1,1% de anomalías congénitas. Se utilizaron modelos de regressión logística múltiple para analizar el papel del consumo de bebidas alcohólicas sobre estos problemas. El consumo de alcohol durante el embarazo sólo se asoció con la prevalencia de anomalias congênitas, con una razón de prevalencia de 3,4. Entre las bebedoras habituales de alcohol en los últimos 12 meses, sólo las mujeres que se ubicaron en la más alta categoria de consumo mostraron relaciones importantes con los tres problemas mencionados. Se sugiere la necesidad de realizar estudios de seguimento para llegar a conclusiones más definitivas.
In recent years, interest has arisen concerning the possible effects of alcohol consumption on pregnancy outcomes. Such concern ranges from Fetal Alcohol Syndrome (FAS), involving major congenital abnormalities and developmental measurements in babies of women clinically diagnosed with Alcohol Dependeney Syndrome (ADS). (Jones & Smith 1973), to fetal alcohol effects (FAE), involving minor problems. Some of these are low birth weight or preterm delivery in social drinkers. In fact, the effects of alcohol consumption can be considered a continuum ranging from FAS to FAE. The greater the consumption during the initial stages of pregnancy, the greater and more serious the consequences for the infant (Abel, 1984).
Although there are no longer any doubts that high alcohol consumption during the first trimester of pregnancy produces teratogenesis (Kline et al., 1989), controversy still surrounds the effects of different amounts of alcohol in social drinking, the importance of consumption periods, and the number of occurrences of heavy drunkenness on problems such as spontaneous abortion, preterm delivery, and specific aspects of growth and development (Borges, 1990).
In Mexico, as in other countries, interest has been rekindled and in 1988 the General Directorate of Epidemiology and the Mexican Institute of Psychiatry conducted the first National Addiction Survey (Encuesta Nacional de Adicciones [ENA]), which provided representative data on the use of alcoholic beverages, tobacco, and drugs, both nationally as well as in seven different regions (Tapia-Conyer et al., 1990). Given tlie growing concern over women's consumption of these substances during pregnancy and the harmful effects on the infant, the ENA questionnaire included questions on the subgroup of women that had been pregnant at some time in their lives.
The objective of the present study is to provide estimates of the relationship between alcohol consumption and some adverse effects on pregnancy, such as spontaneous abortion, stillbirth, and congenital abnormalities in Mexico.
The ENA is a probability survey of households allowing one to estimate the prevalence of tobacco, alcohol, and licit and illicit drug use in a population 12-65 years of age residing in urban areas of Mexico (>2,500 inhabitants). It is part of a national system of health surveys, so that its design derives from the overall sampling framework of that system (Tapia-Conyer et al., 1990).
A total of 13,005 interviews were conducted in 1988,96.7% of which were complete (12,581) and 2.6% rejected. Of the 7,672 female respondents, 5,234 women had been pregnant at some time in their life. The last group of women were also asked if spontaneous abortions, stillbirths, or congenital abnormalitles had occurred during their last pregnancy. Valid responses ranged from 4,844 for abortion to 4,574 for stillbirths and 4,567 for congenital abnormalitles.
During in-home, face-to-face interviews carried out in the ENA, respondents gave information on various aspects of alcohol consumption. Here we will analyze the variables related to alcohol use, such as consumption of alcoholic beverages during the last pregnancy, presence of ADS symptoms during the last 12 months, frequency of consumption during the last 12 months, frequency of intoxication during the last 12 months, and consumption patterns. The latter were based on the frequency, amount, and variability of consumption reported by interviewees (Cahalan & Cisin, 1968), resulting in the following levels:
1) Abstainers: drink no alcoholic beverages or do so less than once a year.
2) Infrequent drinkers: drink once a year or more, but not more than once a month.
3) Low-level drinkers: drink once a month or more, no more than five drinks per occasion.
4) High-level drinkers: drink once a week or more often and have five or more drinks at a sitting occasionally (at least once a year).
5) Heavy drinkers: drink five or more drinks at a sitting once a week or more.
ADS is defined according to the 9th Revision of the International Ciassification of Diseases. It includes indicators for compulsive use, incapability of control, tolerance, reduction of the behavioral repertory, abandoning of alternative activities and other pleasures for alcohol use, persistence of abuse despite harmful consequences, quick recurrence after a period of abstinence, and withdrawal symptoms. Dependency is considered to exist when at least three of these elements are present during a 12-month period (Medina-Mora et al., 1989). Details of the scales used for this research have been presented elsewhere (Tapia et al., 1990).
Each of the above-mentioned adverse pregnancy outcomes is related to alcoholic beverages and other risk factors observed at the time of interview, such as: region of the country, year of migration to the current place of residence, marital status, age, educational status, work status and habitual tobacco use. The importance of these variables was shown through an unconditional multiple logistic-regression model (Kleinbaum et al., 1982). The analysis took weighted data from the interview into account.
Prevalences obtained for spontaneous abortion (3.8%), stillbirth (1.2%), and congenital abnormalities (1.1%) are low. Even research of the magnitude of the ENA produces small sample sizes for this type of problem. They range from 198 cases for spontaneous abortion to 59 for congenital abnormalities.
Table 1 shows the prevalence obtained from the weighted data for the three pregnancy outcomes, according to some ENA socio-demographic variables. Generally speaking, problems are more frequent in the central region of Mexico, and for abortion and stillbirth, older women report the highest prevalence. There is no apparent relationship between problems and years of residence or marital status. Moreover, the relationship between problems at work during the last month or education do not seem to be important for abortion and congenital abnormalities. The prevalence of habitual tobacco consumption does seem important. The small number of cases found in this study precludes definitive statements about these differences.
Table 2 presents diverse prevalence, according to variables associated with alcohol consuption. The highest prevalence rates are for natural abortion: 21% for women classified with ADS. However, the finding is unstable given the reduced prevalence of the latter. In relation to congenital abnormalities some cells were left empty, especially in the highest consumption categories, due to the low prevalence of that condition.
In order to choose variables explaining the three problems under study, several logistic regression models were tested. These included, first, the variable of alcohol consumption, and then socio-demographic and tobacco-use variables. After testing different models, we concluded that only tobacco consumption, age, and ENA region had more than a 10% effect on the crude estimates of the odds ratio obtained for alcohol consumption.
Table 3 summarizes the findings on the influence of alcohol consumption on spontaneous abortion, stillbirth, and congenital abnormalities, controlling for age (continuous variable), ENA region, and number of cigarettes smoked. The table presents the prevalence odds and 95% confidence limits. Alcohol consumption during pregnancy was only found to be associated with the prevalence of congenital abnormalities, with an odds ratio of 3.4. Regarding the following variables, alcohol consumption during the last 12 months, it was generally observed that the relationship between each variable at higher levels of consumption is increasingly evident. Alcohol consumption appears to be more important for congenital abnormalities than for stillbirth and lastly for spontaneous abortion. However, only in this latter complication are there cases with ADS. Consumption patterns seem to be the most important variable for distinguishing women at highest risk. It is also important to note that no relationship of dose-response in the variable of grams of alcohol consumed per year was found to exist.
Since publication of the first studies on FAS in 1973, the literature on this subject has grown considerably. However, studies seeking to relate alcohol consumption during pregnancy to growth measurements and development such as low birth weight are more frequent than ones attempting to relate problems such as spontaneous abortion or stillbirth. Most studies on spontaneous abortion performed in developed countries are cohort or case-control studies (Kline et al., 1980; Harlap & Shiono, 1982; Berkowitz, 1981). These first studies demonstrated a relationship between alcohol consumption and the incidence of spontaneous abortion. More recent studies, with similar methodologies, demonstrate that excessive alcohol consumption around the time of conception or during the first trimester of pregnancy are related to a series of congenital abnormalities, especially craniofacial ones (Ernhart et al., 1987; Day et al., 1989; Day et al., 1990; Rostand et al., 1990).
It is rarer to find studies based on cross-sectional survey designs establishing a relationship between the prevalence of these problems and alcohol consumption. However, two surveys done in the U.S. in 1980 and 1981 report such data. The 1980 study, based on the National Natality Survey and the National Fetal Mortality Survey, found only slight differences between alcohol consumption before and after pregnancy among mothers with underweight or aborted babies. Tobacco consumption is, nevertheless, more prevalent in the mothers of these cases (Prager et al., 1984). The 1981 study (a national survey on female alcohol consumption) found that women who consume six or more drinks at least five days a week, have a higher prevalence of babies with congenital abnormalities (Wilsnack et al, 1984). A more recent prevalence study found that alcohol consumption during the first trimester of pregnancy was related to spontaneous abortions in Canadian women, with an odds ratio of 1.82 for those drinking 21 and more drinks per week; nevertheless, the same study reported only null or weak associations between alcohol consumption and contenta abnormalities (Armstrong et al., 1992).
Our study found that only prevalence of congenital abnormalities is related to alcohol consumption during pregnancy. By analyzing consumption during the last twelve months, we found that women with the highest consumption were those with the highest prevalence of the three problems. However, use of across-sectional design whose primary objective was not the study of these questions meant major limitations in these findings, such as the small sample sizes for the analysis, especially for contenta abnormalities, and the possibility of important under-reporting of the three types of problems researched. Errors in classification for exposure (under-reporting consumption) and consequences (abortion, stillbirth, and abnormalities) are also possible. If these errors are non-differential, this would help explain the low risk found for the categories of lower alcohol consumption.
However, even with these limitations, this study is the first national epidemiological effort to determine the effects of alcohol consumption on the product of pregnancy. Follow-up cohort studies should be initiated in Mexico, specifically examining the effect of addiction, including alcohol, in order to obtain more conclusive findings than those making health recommendations for consumer groups. It is important to detect those women with excessive consumption or with ADS, so as to provide them with counseling during pregnancy or when planning a pregnancy on the possible effects of alcohol abuse on the fetus.
ABEL, E. L., 1984. Fetal Alcohol Syndrome and Fetal Alcohol Effects. New York: Plenum Press.
ARMSTRONG, B. G.; McDONALD, A. D. & SLOAN, M., 1992. Cigarette, alcohol and coffee consumption and spontaneous abortion. American Journal of Public Heaith, 82:85-87.
BERKOWITZ, G. S., 1981. An epidemiological study of preterm delivery. American Journal of epidemiology, 113:81-92.
BORGES, G., 1990. Consumo moderado de bebidas alcohólicas por mujeres embarazadas: Una controversia epidemiológica. Salud Pública de México, 32:507-22.
CAHALAN, D. & CISIN, J. H., 1968. American drinking practices: Summary of findings fram a national probability sample. Quarterly Journal of Studies on Alcohol, 29:130-5 1.
DAY, N. L.; JASPERCE, D.; RICHARDSON, G.; ROBLES, N.; SAMBAMOORTHI, U.; TAYLOR, P.; SCHER, M.; STOFFER, D. & CORNELIUS, M., 1989. Prenatal Exposure to alcohol: Effect on infant growth and morphologic characteristics. Pediatrics,84:536-41
DAY, N. L.; RICHARDSON, G.; ROBLES, N.; SAMBAMOORTHI, U.; TAYLOR, P.; SCHER, M.; STOFFER, D.; JASPERCE, D. & CORNELIUS, M., 1990. Effect of prenatal alcohol exposure on growth and morphology of offspring at 8 Months of age. Pediatrics, 85:748-52.
ERNHART, C. E.; SOKOL, R. B.; MARTIER, S.; MORON, P.; NADLER, D.; AGER, J. W. & WOLF, A., 1987. Alcohol teratogenicity in the human: A detailed assessment of specificity, critical period and threshold. American Journal of Obstetrics and Gynecology, 156:33-9.
HARLAP, S. & SHIONO, P. H., 1982. Alcohol, smoking and incidence of spontaneous abortions in the first and second trimester. The Lancet, 2:173-6.
JONES, K. L. & SMITH, D. W., 1973. Pattern of malformation in offspring of chronic alcoholic mothers. The Lancet,1:1267-71.
KLEINBAUM, D. G.; KUPPER, L. L., & MORGENSTERN, H., 1982. Epidemiologic Research. California: Wadsworth, Inc.
KLINE, J.; SHROUT, R; STEIN, Z.; SUSSER, M. & WARBURTON, D., 1980. Drinking during pregnancy and spontaneous abortion. The Lancet, 2:176-80.
KLINE, J.; STEIN, Z. & SUSSER, M., 1989. From Conception to Birth: Epidemiology of Prenatal Development. NewYork: Oxford University Press.
MEDINA-MORA, M. E.; TAPIA, C. R.; SEPULVEDA, J.; OTERO, M. R.; RASCON, M. L.; SOLACHE, G.; LAZCANO, F.; VILLATORO, J.; MARIÑO, M. C. & LOPEZ, E. K., 1989. Patrones de consumo de alcohol y síntomas de dependencia en una región del centro de la República Mexicana. Revista Latinoamericana de Alcohol y Drogas, 1:47-56.
PRAGER, K.; MALIN, H.; SPIEGLER, D.; VAN NATTA, P. & PLACEK, P., 1984. Smoking and drinking behavior before and during pregnancy of married mothers of live-born infants and stillborn infants. Public Health Reports, 99: 1 17-22.
ROSTAND, A.; KAMINSKI, M.; LELONG, N.; DEHAENE, P.; DELESTRET, I.; KLEIN-BERTRAND, C.; QUERLEU, D. & CREPIN, G., 1990. Alcohol use in pregnancy: craniofacial features and fetal growth. Journal of Epidemiology and Community Health, 44:302-6.
TAPIA-CONYER, R.; MEDINA-MORA, M.,E.; SEPÚLVEDA; J.; DE LA FUENTE, R. & KUMATE, J., 1990. La encuesta nacional de adicciones de México. Salud Pública de México, 32:507-22.
WILSNACK, R. W.; WILSNACK, S. C.& KLASSEN, A. D., 1984. Womens'drinking and drinking problems: patterns from a 1981 national survey. American Journal of Public Health, 74:1231-8.