Comparison of a Timing-Based Measure of Unintended Pregnancy and the London Measure of Unplanned Pregnancy.

CONTEXT: Unintended pregnancy is a universal benchmark for reproductive health, but whether variations reflect differences in measurement and how well measures predict pregnancy outcomes warrant further examination. U.S. and British measures of unintended and unplanned pregnancy offer a useful comparison.

METHODS: Some 220 women seeking pregnancy testing at the Columbia University Medical Center in 2005 responded to three pregnancy measures: a binary timing-based measure of unintended pregnancy (TMUP); a multi-item measure of timing-based intentions and planning behaviors, the London Measure of Unplanned Pregnancy (LMUP); and a measure combining intentions (from the TMUP) and how women would feel about a positive pregnancy test. Six-month pregnancy status was assessed among 159 respondents. Estimates of unintended and unplanned pregnancy were calculated using the TMUP and the LMUP, and receiver operating characteristic (ROC) curves were generated to assess congruence.

RESULTS: According to the TMUP, 76% of pregnancies were unintended; by contrast, LMUP scores categorized 39% as unplanned. The ROC curve indicated that expanding the range of scores for classifying pregnancies as unplanned on the LMUP would achieve greater congruence between these measures. At six months, the proportion of pregnancies that had ended in abortion was 42% of those classified as unintended using the TMUP, 60% of those classified as unplanned using the LMUP and 71% of those that women said they had not intended and were very upset about.

CONCLUSIONS: U.S. and British measures of unintended pregnancy are not directly comparable, and a measure combining intentions and feelings may better predict pregnancy outcomes.