Health care organizations are leveraging machine-learning techniques, such as artificial neural networks (ANN), to improve delivery of care at a reduced cost. Applications of ANN to diagnosis are well-known; however, ANN are increasingly used to inform health care management decisions. We provide a seminal review of the applications of ANN to health care organizational decision-making. We screened 3,397 articles from six databases with coverage of Health Administration, Computer Science and Business Administration. We extracted study characteristics, aim, methodology and context (including level of analysis) from 80 articles meeting inclusion criteria. Articles were published from 1997–2018 and originated from 24 countries, with a plurality of papers (26 articles) published by authors from the United States. Types of ANN used included ANN (36 articles), feed-forward networks (25 articles), or hybrid models (23 articles); reported accuracy varied from 50% to 100%. The majority of ANN informed decision-making at the micro level (61 articles), between patients and health care providers. Fewer ANN were deployed for intra-organizational (meso- level, 29 articles) and system, policy or inter-organizational (macro- level, 10 articles) decision-making. Our review identifies key characteristics and drivers for market uptake of ANN for health care organizational decision-making to guide further adoption of this technique.

BackgroundDespite research demonstrating the potential effectiveness of Telehomecare for people with Chronic Obstructive Pulmonary Disease and Heart Failure, broad-scale comprehensive evaluations are lacking. This article discusses the qualitative component of a mixed-method program evaluation of Telehomecare in Ontario, Canada. The objective of the qualitative component was to explore the multi-level factors and processes which facilitate or impede the implementation and adoption of the program across three regions where it was first implemented.MethodsThe study employs a multi-level framework as a conceptual guide to explore the facilitators and barriers to Telehomecare implementation and adoption across five levels: technology, patients, providers, organizations, and structures. In-depth semi-structured interviews and ethnographic observations with program stakeholders, as well as a Telehomecare document review were used to elicit key themes. Study participants (n = 89) included patients and/or informal caregivers (n = 39), health care providers (n = 23), technicians (n = 2), administrators (n = 12), and decision makers (n = 13) across three different Local Health Integration Networks in Ontario.ResultsKey facilitators to Telehomecare implementation and adoption at each level of the multi-level framework included: user-friendliness of Telehomecare technology, patient motivation to participate in the program, support for Telehomecare providers, the integration of Telehomecare into broader health service provision, and comprehensive program evaluation. Key barriers included: access-related issues to using the technology, patient language (if not English or French), Telehomecare provider time limitations, gaps in health care provision for patients, and structural barriers to patient participation related to geography and social location.ConclusionsThough Telehomecare has the potential to positively impact patient lives and strengthen models of health care provision, a number of key challenges remain. As such, further implementation and expansion of Telehomecare must involve continuous assessments of what is working and not working with all stakeholders. Increased dialogue, evaluation, and knowledge translation within and across regions to understand the contextual factors influencing Telehomecare implementation and adoption is required. This can inform decision-making that better reflects and addresses the needs of all program stakeholders.Electronic supplementary materialThe online version of this article (doi:10.1186/s12913-015-1196-2) contains supplementary material, which is available to authorized users.

Objectives: To assess the cost-effectiveness of the pilot Toronto tele-retina screening program in comparison with existing standard of care (SOC) diabetic retinopathy (DR) screening for patients with diabetes mellitus and in a simulated Pan-Ontarian cohort. Methods: Decision trees were constructed to compare tele-retina to SOC in the pilot and Pan-Ontarian cohort. Cost-effectiveness was assessed as cost per case detected (true-positive) and cost per case correctly diagnosed (true-positive and true-negative results).