UCSF

OBJECTIVES: To evaluate the psychometric performance, including predictive validity, of a UK version of the Desire to Avoid Pregnancy (DAP) scale. DESIGN: Prospective cohort study for psychometric evaluation. SETTING: UK. PARTICIPANTS: Women in the UK aged 15 years to menopause, who were not pregnant at the time of recruitment in October 2018, were eligible. 994 women completed the baseline survey and 90.2% of women eligible for the 12-month survey participated. PRIMARY AND SECONDARY OUTCOME MEASURES: The DAP scale was assessed according to key measurement properties of validity (construct (structural and hypothesis testing) and criterion (predictive)), reliability (internal consistency using Cronbachs alpha and test-retest using intraclass correlation coefficients, ICC) and differential item functioning. Item response and classical test theory methods were used. RESULTS: The scale was acceptable, understandable and showed good targeting with the full range of scores captured. Construct validity was demonstrated on hypothesis testing, with odds of contraceptive use increasing threefold with each increasing DAP point (range: 0-4). Eighty per cent of women with the lowest DAP score became pregnant within 12 months, compared with <1% of those with the highest DAP score. Reliability, both in terms of internal consistency (Cronbachs α 0.96) and test-retest (ICC 0.95), was excellent. Some tests of structural validity, in relation to model fit with the item-response model, were not met, and investigations suggest further exploration of the factor structure of the DAP is required in other samples. Item 5, regarding relationship with a partner, showed differential item functioning by age, number of children and relationship group. CONCLUSIONS: The UK DAP is a valid and reliable measure of womens DAP and is highly predictive of pregnancy within the next 12 months. Further evaluations should continue the assessment of the factor structure and the performance of the item relating to the partner.