PURPOSE: To quantify women’s preferences for information when making the decision to have non-invasive prenatal testing (NIPT) to understand the risk of a genetic condition in their baby.
METHODS: An online discrete choice experiment (DCE) quantified the preferences of a purposive sample (recruited using an online panel; Dynata) of current and future mothers aged 18 to 45 years living in Sweden. Respondents chose their preferred alternative from three non-labelled options representing two modes of information provision or an opt-out (no NIPT or information). A systematic review supported by think-aloud interviews informed the selection of three attributes with four levels (when information is provided; format of information provided; out-of-pocket cost associated with receiving information) and one attribute with two levels (level of detail in information provided). The final experimental design consisted of four blocks of eight choice sets generated to minimise the D-error. Respondents were asked background questions about themselves, previous pregnancies, number of children and genetic illness in the family. Additional questions were asked to assess respondents’ general attitude towards information provision for healthcare decision-making. Choice data were analysed using a conditional logit model and willingness-to-pay (WTP) values were calculated.
RESULTS: Of the 1000 women completing the survey, 660 (66%) had children, 86 (9%) were currently pregnant and 162 (16%) reported having a prenatal test. A quarter of respondents (n=263; 26%) reported having a genetic illness in the family. Respondents indicated a high-level of engagement with information provision in general with a mean score of 2.74 on a pre-defined scale from zero (no engagement) to four (high engagement). Estimated coefficients from the conditional logit indicated that all attributes were statistically significant predictors of choice with signs in line with a priori expectations. In general, women preferred to receive extensive levels of information before/at nine weeks of pregnancy using either leaflets, internet or an app. Strong preferences were indicated for not receiving information in an antenatal group discussion. Women were willing-to-pay SEK736 for a scenario in which the attribute levels were set at the mean effect indicating a preference towards NIPT.
CONCLUSIONS: This study suggests that women living in Sweden value appropriate information provision, given before nine weeks of the pregnancy, when deciding whether to have NIPT.
