Usability and Acceptability of a Conversational Agent Health Education App (Nthabi) for Young Women in Lesotho: Quantitative Study

BACKGROUND

Young women in Lesotho face myriad sexual and reproductive health problems. There is little time to provide health education to women in low-resource settings with critical shortages of human resources for health.

OBJECTIVE

This study aims to determine the acceptability and usability of a conversational agent system, the Nthabi health promotion app, which was culturally adapted for use in Lesotho.

METHODS

We conducted a descriptive quantitative study, using a 22-item Likert scale survey to assess the perceptions of the usability and acceptability of 172 young women aged 18-28 years in rural districts of Lesotho, who used the system on either smartphones or tablets for up to 6 weeks. Descriptive statistics were used to calculate the averages and frequencies of the variables. χ2 tests were used to determine any associations among variables.

RESULTS

A total of 138 participants were enrolled and completed the survey. The mean age was 22 years, most were unmarried, 56 (40.6%) participants had completed high school, 39 (28.3%) participants were unemployed, and 88 (63.8%) participants were students. Respondents believed the app was helpful, with 134 (97.1%) participants strongly agreeing or agreeing that the app was "effective in helping them make decisions" and "could quickly improve health education and counselling." In addition, 136 (98.5%) participants strongly agreed or agreed that the app was "simple to use," 130 (94.2 %) participants reported that Nthabi could "easily repeat words that were not well understood," and 128 (92.7%) participants reported that the app "could quickly load the information on the screen." Respondents were generally satisfied with the app, with 132 (95.6%) participants strongly agreeing or agreeing that the health education content delivered by the app was "well organised and delivered in a timely way," while 133 (96.4%) participants "enjoyed using the interface." They were satisfied with the cultural adaptation, with 133 (96.4%) participants strongly agreeing or agreeing that the app was "culturally appropriate and that it could be easily shared with a family or community members." They also reported that Nthabi was worthwhile, with 127 (92%) participants reporting that they strongly agreed or agreed that they were "satisfied with the application and intended to continue using it," while 135 (97.8%) participants would "encourage others to use it." Participants aged 18-24 years (vs those aged 25-28 years) agreed that the "Nthabi app was simple to use" (106/106, 100% vs 30/32, 98.8%; P=.01), and agreed that "the educational content was well organised and delivered in a timely way" (104/106, 98.1% vs 28/32, 87.5%; P=.01).

CONCLUSIONS

These results support further study of conversational agent systems as alternatives to traditional face-to-face provision of health education services in Lesotho, where there are critical shortages of human resources for health.

TRIAL REGISTRATION

ClinicalTrials.gov NCT04354168; https://www.clinicaltrials.gov/study/NCT04354168.

Change in Sexual and Reproductive Health Knowledge among Young Women Using the Conversational Agent "Nthabi" in Lesotho: A Clinical Trial

Background

Young women worldwide face problems like unwanted pregnancy and sexually transmitted infections. Providing sexual and reproductive health education to this population remains a priority. It is unknown if using digital health interventions to deliver health education in human resource-constrained settings is effective.

Methods

We conducted a clinical trial of the Nthabi intervention to determine participant's knowledge before and after discussion of family planning, folic acid and healthy eating among young women aged 18-28 years in two rural districts of Lesotho who used the Nthabi conversational agent system on either smartphones or tablets for up to six weeks. The number of correct pre- and post-test responses were compared using generalized linear models that directly estimated the proportions and percentages of correct responses.

Results

Of the 172 participants enrolled, the mean age was 22.5 years, 91% were unmarried, 69% completed high school, 23% were unemployed and 66% were students. The mean number of interactions with Nthabi was Family planning was chosen to be discussed by 82 (52.2%), of the 172 participants and of those, 49 (59.8%) completed the content on this topic, and 26 (53.1%) completed the post-test. For the 11 questions about family planning, there were 717 (76.6%) correct responses on the pre-test and 320 (89.9%) on the post-test (p = 0.0233). Folic acid was chosen to be discussed by 74 (47.1%) of 172 participants, and of those, 27 (36.5%) completed the content on this topic, and all 27 (100%) completed the post-test. For the 5 questions about folic acid use, there were 181 (45.3%) correct responses on the pre-test and 111 (71.6%) on the post-test (p < 0.0001). The number of correct responses on the post-test was positively associated with the number of sessions that the participant engaged with Nthabi.

Conclusion

The Nthabi conversational agent system increased knowledge of family planning methods and folic acid use among young women in Lesotho. Digital health interventions like Nthabi offer new opportunities to deliver reproductive health information in countries that have limited human resources for health.

Trial Registration

ClinicalTrials.gov ID: NCT04354168.