UC Berkeley

While Electronic Health Records (EHR) systems have been consistently promoted as a policy priority for improving the quality and efficiency of the American healthcare system, there is limited research evidence to inform policy-makers on how the organizational context may impact any potential benefits from use of the EHR. For my dissertation, I leveraged the staggered nature of the EHR implementation at Kaiser Permanente Northern California (KPNC) to conduct a quasi-experimental study with concurrent controls evaluating the impact of a certified EHR and primary care team member's working relationships on measures of care coordination and quality for patients with diabetes. The study included 110 primary care teams across seventeen medical centers within the KPNC delivery system, and clinical outcomes for over eighty thousand patients with diabetes. Measures of team cohesiveness were captured using a previously validated survey instrument collected from primary care team members in 2005, before the staggered implementation of the EHR. Clinical outcome measures included all LDL-C and HbA1c values captured using the health system's automated databases for laboratory tests results between 2005 and 2009.

I found that while the introduction of an outpatient EHR alone was associated with substantial improvements in care coordination across clinicians, it was not associated with improvements in coordination of care across delivery sites. However, use of the integrated outpatient-inpatient EHR system was associated with significant improvements in coordination of care across delivery sites and across clinicians.

I then examined how the specific effects of use of the integrated outpatient-inpatient EHR on coordination of care across delivery sites vary by primary care team member working relationships. I found that while use of the integrated EHR was associated with significant improvements in coordination for clinicians working in primary care teams with high cohesion, I did not find any benefits of EHR use on coordination for clinicians working in teams with poor cohesion.

Lastly, I analyzed how the effect of use of an outpatient EHR on clinical outcomes for patients with diabetes vary by primary care team member's working relationships. I found that while EHR use resulted in improved physiologic measures of disease control for all patients with diabetes, team functioning was an statistically significant moderator of this effect. Patients cared for by primary care teams with low cohesion experienced significantly reduced improvements in their HbA1c and LDL-C levels as a result of the EHR compared with patients cared for by primary care teams with high cohesion.

The organizational context, in particular primary care team members' working relationships, is critical to maximize any potential gains in care quality from EHR use. I found that clinicians work in teams with strong work relationships were able to leverage the tools available in the EHR to achieve significantly greater improvement in care. Policies aimed at increasing EHR use to improve care quality should consider including combined interventions that also target improvements in team working relationships.