Because of the low survey response rate of the Hospital Consumer Assessment of Healthcare Provider and Systems (HCAHPS) patient satisfaction survey, currently administered by more than 4500 hospitals, patient satisfaction scores (PSSs) may not reflect true satisfaction with health care delivery. Specialty hospitals (SHs) are a subgroup of hospitals that advertise high PSSs. Although several factors have been suggested as reasons for differences in PSSs between SHs and general medical hospitals (GMHs), the difference and effect of survey response rates on PSSs has not been explored. Therefore, we designed this study to examine whether there are differences in survey response rate between SHs and GMHs and if these differences affect PSSs.
HCAHPS results were obtained for July 2007–June 2010. SHs were identified using American Hospital Association data, the Physician Hospital Association's hospital directory, and online search. The HCAHPS survey item “definitely recommend the hospital” was considered to represent overall PSS. Linear regression was performed to examine the relationship between overall satisfaction and hospital specialty status (yes or no), survey response rate, and selected subdomains of PSS. To avoid colinearity, a variance inflation factor of less than 4 was used as a cutoff to include subdomains of PSS in the regression.
We identified 168 SHs; the majority were concentrated in states that do not require certification of need (27% in Texas, 11.7% in Louisiana) and the majority did not have emergency services (43.4% of SHs vs. 93.1% of GMHs). Compared with GMHs, SHs had higher overall PSS (86.2 vs. 69.1; P < 0.0001) and survey response rate (39% vs. 66%; P < 0.0001). Once adjusted for survey response rate, the difference in PSS decreased by >50% (from 17.1 to 7.7) but remained statistically significant (P < 0.0001). Similar results were obtained for PSS subdomains (Table 1). Survey response rate was significantly associated with overall PSS, as each 0.01 increase in response rate was associated with a 0.35 increase in PSS. Survey response rate remained a significant predictor of overall PSS even after adjustment for satisfaction with physician communication, cleanliness, and discharge preparedness (Table 2).
SHs have a significantly higher overall HCAHPS PSS than GMHs, although more than half of this difference disappears once adjusted for survey response rate. Implementation of strategies to increase survey response rate will make PSSs more reliable and may identify factors that improve health care delivery and experience of hospitalized patients.