Lifestyle and risk of premature sexual activity in a high school population of Seventh-Day Adventists: Valuegenesis 1989

Adolescence, Summer, 1996 by Miriam L.M. Weinbender, Annette MacKay Rossignol

Within the data subset of 40 questions, an epidemiologic, case/control study was designed to identify the level of risk for adolescent sexual activity associated with participating in each behavior of interest. Cases were defined as individuals who had "ever had sexual intercourse"; controls were individuals who had never had intercourse. (Question 415 of the Valuegenesis test questionnaire asked "Have you ever had sexual intercourse ("gone all the way," "made love") and included four levels of "yes": "once," "twice," "3 times," and "4 or more times," and a "no" response. Any student answering "yes" to any of these responses was categorized as a case. Students answering "no" were categorized as controls.)

The exposure odds ratio (EOR) for early sexual activity associated with each behavior studied was estimated by dividing the odds of adolescent sexual activity among "exposed" students by the odds among "unexposed" students for each level of each behavioral variable evaluated (Hennekens & Buring, 1987). Thus, the exposure odds ratio provides an estimate of the risk of early sexual activity for students engaging in a particular behavior divided by the risk among students who do not engage in the behavior.

To prevent confounding by age and sex, the data were analyzed separately for males and females, and for each grade. The gender-specific data were later combined for grades for which the exposure odds ratio were similar. This process yielded four groups for analysis: ninth grade males, tenth through twelfth grade males, ninth and tenth grade females, and eleventh and twelfth grade females. Confidence limits for the exposure odds ratios were estimated by the method proposed by Miettinen (Rothman & Boice, 1979).

RESULTS

Overall, 21.6% of the subjects reported being sexually active (24.3% of males and 19.6% of females). See Table 1.

Figures 1 through 7 summarize the exposure odds ratios for being sexually active by gender and school grade group for each behavioral risk factor evaluated. To simplify the data presentation, only the highest exposure category is shown - either "more than 20 hours per week" or "several times each week or more often."(1) Each of the increased or [TABULAR DATA FOR TABLE 1 OMITTED] decreased exposure odds ratios shown in the figures is associated with a two-sided 90% confidence interval that excludes the null effect.

Ninth and tenth grade girls showed the highest risks for premature sexual activity associated with legal drug use in the form of beer or liquor while ninth grade boys showed the lowest risk [ILLUSTRATION FOR FIGURE 1 OMITTED]. Tenth through twelfth grade boys were at considerably higher risk than were eleventh and twelfth grade girls for the use of beer and liquor. An additional question about alcohol use, which appeared in a different section of the questionnaire from the questions forming Figure 1, yielded risks of early sexual activity associated with "drinking alcohol (beer, wine, or liquor) alone or with friends 40 or more times during the last 12 months" ranging from 14.0 for ninth grade boys to 75.6 for ninth and tenth grade girls. The highest risks for early intercourse associated with tobacco use were lower than the risks associated with beer or liquor use (exposure odds ratio = 19.1 vs. 29. "Beer or liquor drinking" and "tobacco smoking" were higher risk behaviors in this population than were the use of wine or the use of caffeinated beverages. However, drinking wine was associated with a two-fold increase in the risk of ninth and tenth grade girls' participation in early sexual activity. Drinking caffeinated beverages was associated with a nearly three-fold increase in this risk for the ninth and tenth grade girls and for the tenth through twelfth grade boys.