Assessment of Substance Abuse: T-ACE

views updated


Heavy maternal drinking is a major pregnancy risk and a significant public health problem. Fetal Alcohol Syndrome (FAS) was first reported as a recognizable clinical syndrome nearly thirty years ago. It is characterized by

  1. prenatal and/or postnatal growth restriction
  2. central nervous system (brain) abnormalities and
  3. facial dysmorphology, i.e., an abnormal appearing face characterized by underdevelopment of the midface with small eyes, a short nose and a long simple (flat) philtrum, the area below the nose and above the upper lip.

As these children grow up, they are often mildly mentally retarded, with average IQs of about 70 and disabling behavioral abnormalities. In addition, there is a continuum of abnormalities among offspring exposed before birth to alcohol, but without the full syndromeabnormalities that are much more common than full FAS. There are anatomic anomalies, called alcohol-related birth defects (ARBD) and alcohol-related neurobehavioral disorder (ARND), a set of behavioral abnormalities in offspring prenatally exposed to substantial levels of alcohol. Other adverse pregnancy outcomes related to maternal drinking during pregnancy include miscarriage and stillbirth.

A national goal to reduce the prevalence of FAS by one half by decreasing maternal drinking was set in Healthy People 2000. Unfortunately, the reported prevalence did not decrease through the 1990s, but, in fact, increased, possibly because of improved case finding. Regardless, it is likely that heavy drinking in pregnancy did not decrease, despite warning labels on all alcoholic beverages since 1989.

There is evidence that pregnant women are receptive to advice from their health care providers, particularly their physicians, to quit or at least cut down both drinking alcohol and smoking cigarettes. Given that generalized warnings, such as the warning label, have not proven effective, a more focused approach would seem reasonablethis would focus prevention efforts on women who drink or are likely to drink enough during pregnancy to damage their offspring. Such drinking has been labeled "risk drinking."

The precise level of drinking that might damage the embryo/fetus is unknown, but is probably variable because of differing susceptibility and differing exposures depending on exactly which adverse effect is considered and when during pregnancy the exposure occurs (critical period). Solid estimates of risk drinking have decreased over the years, as better interviewing and statistical techniques have become available. It is now reasonable to use a figure of about seven drinks per week, typically massed on one or two days, but averaging about one drink per day or 0.5 ounces of absolute alcohol per day. This is the amount of absolute alcohol in one can of beer, one glass of wine or one mixed drink of standard size. This amount of alcohol intake, while unlikely to pose any health risk to the mother, is enough to adversely affect the embryo/fetus. There is not convincing evidence of clinically important effects on the offspring from an occasional drink during pregnancy.

There are, as of yet, no laboratory tests, i.e., biological markers, which will reliably identify risk drinking women. The only way to identify them is

THow many drinks can you hold (TOLERANCE)?>=62
AHave people ANNOYED you by criticizing your drinking?Yes1
CHave you felt you ought to CUT DOWN on your drinking?Yes1
EHave you ever had a drink first thing in the morning to steady your nerves or get rid in blackjack, each ACE is worth one. A total score of two or more is positive.

to obtain an appropriate history of drinking, but this is complicated by D Enialthe woman doesn't want to admit drinking to herself or to her doctor. Further, time is distinctly limited during prenatal visits and there are many problems to identify and address. Thus, a brief, simple questionnaire was needed. Most brief questionnaires, such as the CAGE, were developed and tested almost entirely in male populations, and do not function well for reproductive age women.

The T-ACE questions were developed specifically as a screening test for risk drinking. They have been tested and validated over the last decade in women of multiple ethnicities, including white, African American and Native American, and in a range of socioeconomic statuses and geographic locations. The original questionnaire included the question, "How many drinks does it take to make you feel high ?" as the (T)olerance question. An answer of greater than two standard drinks was considered positive. Several studies have now shown that substituting the hold question, included in Table 1, instead of the high question, gives better results, improving the sensitivity of the T-ACE questions.

T-ACE is a screening test, so it was designed to pick up as high a proportion of risk drinkers as possible. This version picks up about nine in ten women who drink enough in pregnancy potentially to damage their baby. If the score is less than two, i.e., the T-ACE is negative, it will correctly identify about seven in ten women who are not risk drinkers. It has a substantial false positive rate, i.e., warning the clinician, though the patient is not, in fact, a risk drinker. It has been speculated though that any woman who scores positive might, in fact, be at risk to drink too much during pregnancy and should be counseled.

A final pointscreening for risk drinking is not enough. At the minimum a brief intervention to support the patient in becoming abstinent during pregnancy or at the minimum cutting way down is warranted, as is close follow-up. If alcohol abuse or dependence is present, consultation or referral may be warranted.


Chang, G., et al. (1998). Alcohol use and pregnancy: Improving identification. Obstetrics and Gynecology, 91, 892-898.

Russell, M., et al. (1996). Detecting risk drinking during pregnancy: A comparison of four screening questionnaires. American Journal of Public Health, 86, 1435-1439.

Russell, M. et al. (1994). Screening for pregnancy risk-drinking. Alcoholism: Clinical and Experimental Research, 18, 1156-1166.

Sokol, R. J., et al. (1989). The T-ACE questions: Practical prenatal detection of risk drinking. American Journal of Obstetrics and Gynecology, 160, 863-870.

Robert J. Sokol