Patient Satisfaction With Providers: Do Patient Surveys Give Enough Information to Help Providers Improve Specific Behaviors

Background: Patient satisfaction surveys ask patients specific questions about provider behavior such as whether they were satisfied with the provider’s instructions about medications or time spent with the patient. It’s unclear how responses to these surveys can help providers focus on specific behaviors to improve. Methods: In a primary care setting, we analyzed Press Ganey patient experience survey responses. We examined the 10 questions dealing with satisfaction specific to the care provider experience. We used the “Top Box” counts (counts of most favorable responses) and Top Box% (percentage of most favorable response) for categorical and continuous measures of patient satisfaction. Results: For 12 consecutive months, 652 providers of 1014 accumulated at least 300 total responses from patients for the 10 provider-related questions. Only 8 of the 652 providers had significant differences (P < .05) in Top Box% for the 10 questions. Correlation of responses between the questions were between 0.86 and 0.96. Analysis of variance showed that 87% of the total variation in the Top Box% of the 10 questions was between providers and only 13% within providers. Factor analysis found no independent factors within the 10 questions (ie, a one factor model was sufficient; P < .0001). Conclusion: Patient survey questions appear to ask about specific provider behaviors that contribute to patient experience. However, the responses to 10 different questions are highly correlated and may not give providers or management enough statistically significant information to focus patient experience improvement efforts for individual providers.