2. Sub-sample Analysis
We conduct a sub-sample analysis to deepen our analysis of patient engagement’s role. We estimate the ordered logistic regression models for two groups of observations – one with an established PFAC and the other without it. The model specifications are identical to our principal analysis, except for patient engagement as a predictor variable. Table 4 shows the estimation results for the quality of care. One notable difference between the two groups is the impact of human resources. For hospitals without patient input, only the number of doctors significantly improves the quality of care. However, for hospitals using patient engagement as their resource, the number of nurses per bed significantly predicts performance. A hospital with one more registered nurse per bed has 79.5% greater odds of receiving a higher quality of care score.
Table 4. Sub-sample Results – Quality of Care