Sexual behaviour in Britain: early heterosexual experience

Background Relatively high rates of teenage conception and sexually transmi tted infection among young people in Britain have focused attention on earl y sexual behaviour and its determinants. We report the results of the secon d National Survey of Sexual Attitudes and Lifestyles (Natsal 2000). Methods We did a probability sample survey between 1999 and 2001, of men an d women aged 16-44 years in Britain. Participants were interviewed using a combination of computer-assisted face-to-face and self-completion questionn aires, and asked questions regarding first heterosexual intercourse, commun ication about sex, pregnancy, and sexually transmitted infections (STIs). Findings We recruited 11 161 men and women to the survey (4762 men, 6399 wo men). The proportion of those aged 16-19 years at interview reporting first heterosexual intercourse at younger than 16 years was 30% for men and 26% for women; median age was 16 years. The proportion of women reporting first intercourse before 16 years increased up to, but not after, the mid-1990s. There has been a sustained increase in condom use and a decline in the pro portion of men and women reporting no contraceptive use at first intercours e with decreasing age at interview. Among 16-24 year olds, non-use of contr aception increased with declining age at first intercourse; reported by 18% of men and 22% of women aged 13-14 years at occurrence. Early age at first intercourse was significantly associated with pregnancy under 18 years, bu t not with occurrence of STIs. Low educational attainment was associated wi th motherhood before 18 years, but not abortion. Interpretation The increase in the proportion of women reporting first inte rcourse before age 16 years does not appear to have continued throughout th e past decade. Only a small minority of teenagers have unprotected first in tercourse, and early motherhood is more strongly associated with educationa l level than with family background. Factors mast strongly associated with risk behaviour and adverse outcomes have considerable potential for prevent ive intervention.