JMIR Publications

ABSTRACT

Background:

The low levels of control of hypertension and diabetes mellitus are a challenge that requires innovative strategies to surpass barriers of low sources, distance and quality of healthcare.

Objective:

To develop a clinical decision support system (CDSS) for diabetes and hypertension management in primary care, to implement it in a resource-constrained region, and to evaluate its usability and healthcare practitioner satisfaction.

Methods:

This mixed-methods study is a substudy of HealthRise Brazil Project, a multinational study designed to implement pilot programs to improve screening, diagnosis, management, and control of hypertension and diabetes among underserved communities. Following the identification of gaps in usual care, a team of clinicians established the software functional requirements. Recommendations from evidence-based guidelines were reviewed and organized into a decision algorithm, which bases the messages. Following pre-testing and expert panel assessment, pilot testing was conducted in a quasi-experimental study, which included 34 primary care units of ten municipalities in a resource-constrained area in Brazil. A Likert-scale questionnaire evaluating perceived feasibility, usability and utility of the application and professionals’ satisfaction was applied after six months. In the endline assessment, two focus groups with primary care physicians and nurses were performed.

Results:

A total of 159 reminders and suggestions was created and implemented for the CDSS. At the six-month assessment, there were 1939 patients registered in the application database and 2160 consultations were performed by primary care teams. Of the 96 healthcare professionals who were invited for the usability assessment, 26.0% were physicians, 45.8% were nurses and 28.1% were from other health professionals. The questionnaire composed by 24 items about impressions of feasibility, usability, utility and satisfaction, presented global Cronbach's alpha of 0.93. As for feasibility, all professionals agreed (median scores from 4 to 5) that the application can be used in primary care setting and it can be easily incorporated in work routine, but physicians claimed that the application might have caused significant delays on daily routine. As for usability, overall evaluation was good and it was claimed that the application was easy to understand and use. All professionals agreed that the application was useful (score 4-5) to promote prevention, assist to treat and might improve patient care, and they were overall satisfied with the application (median scores between 4 and 5). In the endline assessment, there were 4211 patients (94,8% with hypertension and 24,4% with diabetes) registered in the application’s database and 7960 consultations were performed by primary healthcare teams. The 17 participants of the focus groups were consistent to affirm they were very with the CDSS.

Conclusions:

The CDSS was applicable in the context of primary health care setting in a low income region, with good user’s satisfaction and potential to improve adherence to evidence-based practices.