J. Gould et al., UTILITY OF PERCENTAGE OF BIRTHS TO TEENAGERS AS A SURROGATE FOR THE TEEN BIRTH-RATE, American journal of public health, 88(6), 1998, pp. 908-912
Objectives. The teen birth rate is commonly used in comparing regional
variation in teen pregnancies, but local teen birth rates are not alw
ays available. In this study the percentage of all births that are to
teens was evaluated for its utility as a surrogate for the teen birth
rate. Methods. Rank correlation and sensitivity and specificity analys
es were used. Results. The Spearman rank cor relations between percent
age of teen births (PTB) and teen birth rate (TBR) were .995,.906, and
.841 for the 3 age groups suggesting that it may be reasonable to emp
loy PTB to prioritize zip codes. Zip codes with upper quartile levels
of percentages of teen births identified zip codes with upper quartile
levels of TBR with a sensitivity of 83.8%, 68.8%, and 65%; a false-po
sitive rate of 2.1%, 8.6%, and 10%; and a positive predictive value of
89.3%, 67.6%, and 67.5% for the age groups 10 through 14, 15 through
17, and 18 through 19 years. Conclusions. The percentage of births to
teens is a useful surrogate for teen birth rate in California, especia
lly among younger teenagers.