Developing citizen report cards for primary health care in low and middle-income countries: Results from cognitive interviews in rural Tajikistan

• Published in: PloS one 2017-10, Vol.12 (10), p.e0186745-e0186745
• Main Authors: Bauhoff, Sebastian, Rabinovich, Lila, Mayer, Lauren A
• Format: Article
• Language: English
• Subjects: Accountability; Adolescent; Adult; Biology and Life Sciences; Cards; Cognitive ability; Community; Developing Countries; Engineering and Technology; Evaluation; Female; Health care; Health care policy; Health Services Accessibility; Humans; Income; Internet; Interviews; Interviews as Topic; Iterative methods; Low income groups; Male; Maternal & child health; Medicine and Health Sciences; Middle Aged; Patient safety; People and Places; Primary care; Primary health care; Primary Health Care - utilization; Quality; Research and Analysis Methods; Rural areas; Rural Health Services - utilization; Social Sciences; Studies; Tajikistan; Young Adult
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0186745

Citizen report cards on health care providers have been identified as a potential means to increase citizen engagement, provider accountability and health systems performance. Research in high-income settings indicates that the wording, presentation and display of performance information are critical to achieve these goals. However, there are limited insights on developing effective report card designs for middle- and low-income settings. We conducted cognitive interviews to assess consumers' understanding, interpretation of and preferences for displaying information for a health care report card in rural Tajikistan. We recruited a convenience sample of 40 citizens (20 women and 20 men aged 18-45) from rural areas of two provinces of Tajikistan (Soghd and Khatlon oblasts). The interview protocol was adapted from the model of cognitive interviews used in social science research to improve survey questionnaires. We used multivariate regression to assess understanding and interpretation of the report card; chi2 tests to assess differences in preferences for displaying information; and tests of proportions to assess the preferred comparison group. Respondents understood the main idea of the report card and are not confused by the indicators or display. However, many respondents had difficulties making comparisons, and when asked to identify worst-performing services. Respondents preferred detailed rankings using school grades, comparisons of their local clinic with the regional or national average performance, and the use of color in the report card. We found some heterogeneity across the two provinces. Overall, our findings are promising regarding the citizens' comprehension of health care report cards in rural Tajikistan, while underscoring the challenges of effectively providing health care performance information to communities. Cognitive interviews and iterative testing can support an effective implementation of reporting initiatives.