Improving the health of young African American women in the preconception period using health information technology: a randomised controlled trial
Brian W Jack, Timothy Bickmore, Leanne Yinusa-Nyahkoon, Matthew Reichert, Clevanne Julce, Nireesha Sidduri, Jessica Martin-Howard, Zhe Zhang, Elisabeth Woodhams, Juan Fernandez, Mark Loafman, Howard J Cabral, Brian W Jack, Timothy Bickmore, Leanne Yinusa-Nyahkoon, Matthew Reichert, Clevanne Julce, Nireesha Sidduri, Jessica Martin-Howard, Zhe Zhang, Elisabeth Woodhams, Juan Fernandez, Mark Loafman, Howard J Cabral
Abstract
Background: Preconception care focuses on improving women's health before pregnancy as a means to improve their health and future pregnancy outcomes. How to effectively deliver such care is unknown. The aim of this research was to assess the impact of an embodied conversational agent system on preconception risks among African American and Black women.
Methods: We did an open-label, randomised controlled trial of women aged 18-34 years, self-identified as African American or Black, or both, and not pregnant, recruited from 35 states in the USA. Sealed allocation envelopes (in permuted blocks of six and eight, prepared using a random number generator) were opened after enrolment. Intervention participants received an online conversational agent called Gabby that assessed 102 preconception risks and delivered 12 months of tailored dialogue using synthesised speech, non-verbal behaviour, visual aids, and health behaviour change techniques such as motivational interviewing. The control group received a letter listing their preconception risks and encouraging them to talk with a clinician. The primary outcome was the proportion of identified risks at the action or maintenance stage of change at months 6 and 12. The study is registered with ClinicalTrials.gov, NCT01827215.
Findings: From March 11, 2014, through July 8, 2018, 528 women recruited from 35 states and 242 cities across the USA received the Gabby intervention (n=262) or were assigned to the control group (n=266). Participants identified a mean of 21 preconception risks per woman (SD 9·9). In the intention-to-treat analysis, at 6 months, intervention women reported reaching the action or maintenance stage of change for 50·0% (SD 28·9) of those preconception risks identified compared with 42·7% (28·3) in the control group (incidence rate ratio 1·16, 95% CI 1·07-1·26; p=0·0004). This result persisted at 12 months.
Interpretation: The Gabby system has the potential to improve women's preconception health. Further research is needed to determine if improving preconception risks impacts outcomes such as preterm delivery.
Funding: National Institute for Minority Health and Health Disparities.
Copyright © 2020 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY-NC-ND 4.0 license. Published by Elsevier Ltd.. All rights reserved.
Source: PubMed