Implementing an electronic health record dashboard for safe anticoagulant management: learning from qualitative interviews with existing and potential users to develop an implementation process

Abstract

Abstract

Background Facilitating appropriate care delivery using electronic health record (digital health) tools is increasing. However, frequently used determinants frameworks seldom address key barriers for technology-associated implementation.

Methods Semi-structured interviews were conducted in two contexts: the national Veterans Health Affairs (VA) following implementation of an electronic dashboard, a population health tool, and the Michigan Anticoagulation Quality Improvement Initiative (MAQI2) prior to implementation of a similar electronic dashboard. The dashboard is designed for pharmacist or nurse use to monitor safe outpatient anticoagulant prescribing by physicians and other clinicians We performed rapid qualitative inquiry analysis and selected implementation strategies. Through a stakeholder focus group session, we selected implementation strategies to address determinants and facilitate implementation in the MAQI2 sites.

Results Among 45 interviewees (32 in VA, 13 in MAQI2), we identified five key determinants of implementation success: (1) clinician authority and autonomy, (2) clinician self-identity and job satisfaction, (3) documentation and administrative needs, (4) staffing and work schedule, and (5) integration with existing information systems. Key differences between the two contexts included concerns about information technology support and prioritization within MAQI2 (prior to implementation) but not VA (after implementation) and concerns about authority and autonomy that differed between the VA (higher baseline levels, more concerns) and MAQI2 (lower baseline levels, less concern).

Conclusions The successful implementation of electronic health record tools requires unique considerations that differ from other types of implementation, must account for the status of implementation, and should address the effects of the tool deployment on clinical staff authority and autonomy. Interviewing both post-implementation and pre-implementation users can provide a robust understanding of implementation determinants.