Health-care providers, the gate keepers of health-care delivery, are important but understudied stakeholders in integrating digital health technologies (DHTs) into routine clinical practice. A doctor’s recommendation goes a long way in encouraging patients to use DHTs, and research increasingly suggests that a desire for DHTs exists among health-care providers. However, despite recent policy recommendations advocating the use of health apps in routine clinical practice, and publication of the National Institute for Health and Care Excellence (NICE) evidence framework for DHTs, such technologies are rarely recommended by health-care providers in practice.
We interviewed groups of digital health experts, practice nurses, allied health professionals, secondary care doctors, and general practitioners to determine the factors important to health-care providers for considering the recommendation of DHTs to patients. Combined with a pragmatic literature review, we identified nine factors which might affect digital decision making among health-care providers (table). From Feb 18, 2019, to April 1, 2019, we invited health-care providers from the UK to take part in a short survey (appendix), either online or via paper forms to identify factors instrumental in health-care providers advocating DHTs to patients. We invited 30 health-care providers in person, of which 28 agreed to take part (93% participation rate), with a further 57 accessing the survey online following advertisement via social media, which resulted in 85 responses in total. All respondents were based in the UK. A job title was collected for 60 (71%) of the 85 respondents; 25 did not answer the question. Among the individuals who answered, 24 (40%) were nurses, 17 (28%) were allied health professionals (such as speech and language therapists), 13 (22%) were general practitioners, 3 (5%) were secondary care doctors, and 3 (5%) answered “other”. An ordinal scale design was applied, where participants ranked themes from 1 (most important) to 9 (least important). Tests for statistical significance were done using the Kruskal-Wallis H test, followed by a post-hoc Dunn’s test when significant differences, defined at the conventional level of 5% (p<0·05), were identified (appendix).