Developing and Validating a Patient Advocacy Engagement Scale for Health Professionals in Acute Care Hospitals

Background. Codes of ethics of nursing, social work, and medicine require members of these professions to engage in advocacy to help patients resolve problems and ensure the highest quality of care. Yet the extent they do so is unknown, largely due to a lack of a broadly applicable instrument with which to measure these healthcare professionals’ actual engagements in advocacy. A major goal of the “Improving Healthcare Outcomes Through Advocacy” study was to develop and validate a scale to measure the extent healthcare professionals engage in patient advocacy.

Methods. Development of the Patient-AES scale began with seven categories of problems identified by Jansson (2011): 1) failure to honor patients’ rights, 2) lack of quality care, 3) lack of culturally-competent care, 4) lack of preventive care, 5) lack of affordable care, 6) lack of attention to mental problems, and 7) lack of community-based care. The project obtained data from a sample of 100 nurses, 100 social workers, and 100 medical residents randomly drawn from rosters at eight acute-care hospital in Los Angeles County. The Patient-AES scale asked these respondents to rate their engagement on each of 33 specific manifestations of the seven core problems during the last two months on a Likert scale from “never” to “almost always.”

To establish content validity, a panel of experts reduced 117 original manifestations to 33. Item-level content validity index (CVI), subscale-level CVI and scale-level CVI, which reflect the proportion agreement among expert raters, was calculated. Exploratory factor analysis (EFA) followed by confirmatory factor analysis (CFA) for both an originally hypothesized seven-factor model and a six-factor model that emerged from EFA was conducted. Cronbach’s α was calculated to establish internal consistency reliability. To establish test-retest reliability, 50 healthcare professionals were recruited to complete the repeat questionnaire (they did so, on average, 41 days from their initial response). Pearson’s correlation coefficient r was calculated. The research team decided to name this scale the Patient Advocacy Engagement Scale (Patient-AES).

Results. Results of index, subscale, and scale-level CVI showed good to excellent content validity. Results of EFA resulted in a six-factor solution, which revealed a clear structure with each item loading substantially (>0.4) on only one factor. Altogether these six factors accounted for 67% of the variability of patient advocacy. A comparison of the model fit indexes resulting from the CFA from both the hypothesized seven-factor model and the six-factor model that resulted from the EFA indicated that both models fit the data equally well. Hence, CFA of the seven-factor model was selected. The test-retest Pearson correlation coefficient r for the entire scale was 0.81, demonstrating that the scale has very good test-retest reliability, and Cronbach’s α for the scale was 0.95, showing very good internal consistency reliability.

Conclusion. The PATIENT-AES should be useful in measuring patient advocacy engagement by social workers, nurses, and medical residents in acute-care hospitals.  It does not measure the duration or effectiveness of these specific engagements.  Some respondents may have over-reported their patient-advocacy engagements even though their identities, data, and hospitals are confidential.