Clinician Behaviour Shapes Shared Record Use

Access to shared patient information does not lead to consistent use across clinical teams. An analysis published in JAMIA Open examined how clinicians at UCSF Health, an academic medical centre in the United States, used an electronic health record interoperability tool before and during the COVID-19 pandemic. The tool gave authorised clinicians access to records held by participating healthcare organisations. Use increased during the later period, but substantial differences remained between clinical specialities, roles and experience levels. The findings suggest that technical connectivity must be considered alongside clinical workflow, information needs and the practical process of retrieving external patient records. 

Use Increased During the Pandemic Period 

Clinicians accessed external patient records more frequently in 2020–2021 than in 2018–2019. After adjustment for appointment volume and clinician characteristics, lookups per visit were 43% higher in the later period. A separate analysis following the same clinicians across both periods produced a similar result, indicating that the increase was not explained only by differences in the clinicians included at each stage. 

Must Read: EHR Data Could Strengthen Real-World Specialist Evidence 

The dataset covered 1,442 clinicians and included activity on days with scheduled appointments as well as days when clinicians used the electronic health record without scheduled visits. Work on these days could include documentation, responses to patient messages, review of test results and preparation for later appointments. This broader approach captured interaction with external information beyond direct consultations, although the principal comparison measured use relative to the number of scheduled visits. 

The analysis identified a clear change over time but could not determine its cause. Continuing adoption of the tool, changing clinical demands and the strain and uncertainty associated with the pandemic were possible influences. Appointment volumes were lower during the later period, yet the rate of external record access still increased. 

The findings therefore show a change in information-seeking behaviour rather than the effect of a specific policy or technical intervention. They also indicate that use can change substantially within the same organisation and among the same clinicians when the wider clinical context changes. 

Speciality and Experience Influenced Access 

Clinical speciality explained more variation in external record use than the other characteristics assessed. Nephrology and cardiology showed the highest levels, with external information accessed in more than half of visits. Internal medicine and haematology-oncology also used the tool relatively frequently. Dermatology and paediatrics recorded the lowest use. 

These differences may relate to variations in care complexity, coordination requirements and the nature of the relationship between clinicians and patients. Some specialities manage chronic conditions over time and coordinate care across multiple services. Others may have more limited or consultative contact. However, the dataset did not identify whether individual encounters involved new or established patients or whether care was consultative or ongoing. The observed differences could not therefore be attributed directly to these factors. 

Clinical role was also associated with use. Resident doctors accessed external records more often per visit than attending physicians, while advanced practice clinicians used them only slightly more than attendings. Use declined gradually with each additional year of professional experience. Gender showed little difference and was not a significant predictor after adjustment. 

The data do not establish whether these patterns reflect familiarity with digital tools, training, workflow or the patients managed by each group. Residents and less-experienced clinicians may have encountered more unfamiliar or complex patients, but patient-level characteristics were not available. Some differences attributed to clinicians may therefore reflect variations in their patient populations and responsibilities. 

Technical Connectivity Did Not Ensure Routine Use 

The findings distinguish the availability of shared records from their use in everyday clinical practice. The interoperability tool was integrated into the electronic health record and allowed clinicians to retrieve information from participating organisations. Despite access through a common system, engagement varied markedly across specialities, roles and levels of experience. 

Use was assessed through several measures. One captured the number of lookups relative to scheduled visits. Another recorded the proportion of visits in which external information was accessed at least once. A further measure captured the proportion of active clinic days involving any use. Together, these measures showed whether clinicians consulted external records across many encounters and whether they performed repeated lookups during individual visits. 

The results also indicate that higher use should not automatically be treated as evidence of better care. The analysis documented variation but did not assume that external records were required for every clinical encounter. It could not determine the amount or completeness of information available to each clinician. A decision not to search may therefore have reflected an absence of relevant data rather than reluctance to use the tool. 

The single-centre design further limits wider application. Other organisations may have different referral patterns, patient populations, technical infrastructures and levels of external data availability. Patient age, illness burden, insurance status and continuity of care were not included. The analysis also did not assess effects on outcomes, costs, duplicate testing or population health. Further work would need to connect record use with encounter characteristics, available information and subsequent clinical decisions. 

Use of shared records increased during the pandemic period but remained uneven across clinical settings. Clinical speciality accounted for the greatest variation, while residents and less-experienced clinicians accessed external information more frequently than attending and more-experienced colleagues. Gender was not a significant factor. The findings support attention to navigation, information retrieval, workflow and training alongside technical infrastructure. Broader evidence is still needed to determine how patient characteristics, encounter type and the availability of external data influence clinicians’ decisions to consult shared records. 

Source: JAMIA Open  

Image Credit: iStock