Health Promotion and Disease Management in Primary Care: A Comparison of Nurse Practitioner and Physician Practice Patterns on Four Selected HEDIS Outcomes

Defense Date


Graduation Date

Spring 1-1-2008


Campus Only

Submission Type


Degree Name





School of Nursing

Committee Chair

Joan Such Lockhart

Committee Member

Elizabeth Corwin

Committee Member

Luann Richardson


disease management, health promotion


Nurse practitioners (NPs) have been providing primary care since 1965. Despite the longevity of their practices, there is sparse new data in the area of NP and physician comparability on health promotion and disease management outcomes. This study compared NPs and physicians on four primary care measures from the Healthcare Effectiveness Data and Information Set (HEDIS): breast cancer screening; cervical cancer screening; diabetes management; and hypertension management. Carper's Fundamental Patterns of Knowing in Nursing served as the study framework.

A retrospective, non-experimental design was utilized to compare the NPs and physicians. Data extraction from an electronic health record data base yielded 3001 women for the breast cancer screening outcome; 7167 women for the cervical cancer screening outcome; 2097 adults for the diabetes management outcomes; and 5133 adults for the hypertension management outcomes. General Estimating Equations were used to calculate odds ratios and 95% confidence intervals. The effects of age, gender, provider and clinic were controlled for where appropriate.

No statistically significant differences were detected between NPs and physicians for the breast cancer screening outcome (adjusted odds ratio 0.86; 95% Confidence Interval= [0.48, 1.53], p=0.61). No differences were detected between NPs and physicians for the cervical cancer screening outcome (adjusted odds ratio 1.28; 95% Confidence Interval= [0.65, 2.51], p=0.47). A significant difference was found between NPs and physicians on the diabetes measure for hemoglobin A1C (HgbA1C) ordering, with physicians demonstrating a stronger performance (adjusted odds ratio 0.48; 95% Confidence Interval= [0.26, 0.91], p=0.025. No differences were detected for the diabetes outcome of HgbA1C results, with NPs and physicians performing comparably for diabetics in good HgbA1C control (adjusted odds ratio 1.07; 95% Confidence Interval= [0.71, 1.61], p=0.73). Finally, no significant differences were detected between NPs and physicians for the hypertension outcome of blood pressure control (adjusted odds ratio 0.83; 95% Confidence Interval= [0.58, 1.17], p=0.29).

The findings suggested NP outcomes were not different from physician outcomes for the selected HEDIS measures, except for HgbA1C ordering. Future studies should investigate this difference in HgbA1C ordering, including the potential influences of a physician incentive program for HgbA1C ordering that existed during the study year.





This document is currently not available here.