Background: The overall national patient satisfaction scores have improved since the HCAHPS reporting began. However, it is unknown whether the improvement trends are different between teaching and non-teaching hospitals. Therefore, our objective was to compare long-term changes in patient satisfaction between teaching and non-teaching hospitals.
Methods: We used three datasets; HCAHPS data (2008 to 2013) from the CMS website, socioeconomic data for hospital service area (HSA) from the US Census Bureau, and HSA information from the Dartmouth Atlas of Health Care. To evaluate satisfaction scores, we used three domains; patient satisfaction with physician communication (PC), patient recommendation of the hospital (RC), and patient rating of the hospital (RT). Satisfaction scores were determined by the percentage of responses to ‘doctors always communicated well’, ‘definitely recommend the hospital’ and percentage of patients who rated hospital as 9 or 10 on a 10-point scale respectively. A hospital was considered a teaching hospital if it obtained graduate medical education funding. The time-trend in satisfaction score was determined by using 2-level multilevel models (to account for correlation between observations from the same hospital) with and without adjusting for median income, population size, racial composition, number of physicians, and insurance status of the HSA for each hospital and survey response rate and hospital ownership.
Results: Of the 3746 hospitals with HCAHPS data in 2008, 1000 were teaching hospitals. Two of the three satisfaction scores for the teaching hospitals were significantly lower in 2008 (mean difference: PC= -3.9%, P<0.001; RT= -3.0%, P<0.001) and remained so in 2013 as compared to non-teaching hospitals (PC= -3.3%, P<0.001; RT= -2.0%, P<0.001). There was no difference in RC scores between teaching hospitals and nonteaching hospitals in 2008 (-0.8%; P=0.06) or in 2013 (0.7%; P=0.07). However, the teaching hospitals had greater improvement in all three scores during the 6 years than nonteaching hospitals (slope difference/year: PC=0.15%; RC=0.38%; RT=0.32%; all interaction P<0.001; Figure). These differences remained significant after adjusting for potential confounders (Table). Survey response rate was positively associated while HSA number of physicians was negatively associated with all scores (both P<0.05).
Conclusions: As compared to nonteaching hospitals, teaching hospitals had lower scores for patient satisfaction with physician communication and lower hospital ratings than nonteaching hospitals but similar scores for hospital recommendation throughout the 6 years. On the other hand, teaching hospitals had higher rates of improvement in all three domains than nonteaching hospitals. These results suggest that teaching hospitals may be more responsive to patient feedback and/or more innovative in addressing shortcomings in patient experience.
| Satisfaction Domains | Unadjusted (P-value) |
Adjusted* (P-value) |
| Physician Communication | 0.15 ; P <0.001 | 0.16 ; P <0.001 |
| Hospital Recommendation | 0.38 ; P <0.001 | 0.40 ; P <0.001 |
| Hospital Rating | 0.32 ; P <0.001 | 0.34 ; P <0.001 |
