In the January 2000 issue of the American Journal of Public Health, an article appears with new estimates of the number of children of alcoholics.  Using data from the 1992 National Longitudinal Alcohol Epidemiological Survey, the NIAAA authors found that 15% of all U.S. children were currently exposed to alcohol abuse and/or dependence in the family (which includes any adults living in the household), and 43% of all children were exposed to someone who, during their lifetime, satisfied a diagnosis of alcohol abuse or dependence.   Assuming that the best estimate lies between these two extremes, it was determined that approximately 1 in 4 children in the US is exposed to alcohol abuse and/or dependence in the family at some point before age 18.

This is just the latest in a long line of studies on this issue.  The first estimates of the number of children of alcoholics were developed in 1974 by Booz-Allen and Hamilton, who suggested extrapolating national survey data on problem drinking to the U.S. population and multiplying the resulting estimate of adult problem drinkers by the ratio of adults to children in the general population.  Such an approach assumes that alcoholic persons are randomly dispersed throughout the population and the ratio between alcoholic persons and children of alcoholic parents is the same as that of adults to children in general. This concept was then applied by the Children of Alcoholics Foundation in 1985, using data from the 1979 National Drinking Practices Survey and from the 1980 Census. In the survey, problem drinking, defined as a score of one or more on a loss of control scale and/or a score of two or more on an alcohol dependence scale, was reported by 10% of the sample, 15% of the males and 6% of the females.  Note that this definition does not imply clinical alcoholism; only one percent of the population reported having gone to someone about a problem related to their drinking.  Use of high cutting points or different alcohol problem scales would change the prevalence rates and corresponding estimates of problem drinking in the adult U.S. population. It should also be kept in mind that the data do not include alcoholics who are in the military, institutionalized, homeless, or who may have refused to participate in the survey. Given these caveats, it was estimated that there were approximately 6,600,000 children of alcoholics under the age of 18 years in the U.S. at that time.

Also included in the survey was a question as to whether, in the respondents’ judgment, their father, mother, sister(s) or brother(s) had been alcoholics or problem drinkers at any time in their lives. Parental alcoholism was reported by 11% of the males (fathers only, 9%, mothers only, 1%, and both, 2%). Applying these rates to the 1980 census data yields an estimate of approximately 22,000,000 children of alcoholics aged 18 or older. Note that reports of problem drinking in parents were obtained in face to face interviews, and were not validated by any other independent measure; therefore the data reflect the perceptions of the respondents only.

In 1995, Eigen and Rowden developed new figures. The researchers first put forth that there are many ways to define COAs:

This study used the DSM-III-R criteria for alcohol dependence and for alcohol to define alcohol problems, and attempted to address the issue of likelihood of parenting a child. The estimates derived for this study are:

While it is extremely difficult to estimate the number of children of alcoholics, coming up with a figure for children of other substance abusers is even more difficult. This is due, in part, to the fact that other drug use is illicit both legally and socially, and therefore is not likely to be disclosed in surveys. Moreover, because many of those who have drinking problems also have other drug problems, epidemiologists would also need to sort out COAs only from COSAs only to COA and COSA.