Lists of hospitals have been compiled showing which hospitals have patient death rates exceeding the national average. The data have been adjusted to allow for differences in the ages of patients.

Each of the following, if true, provides a good logical ground for hospitals to object to interpreting rank on these lists as one of the indices of the quality of hospital care EXCEPT:

• ARank order might indicate insignificant differences, rather than large differences, in numbers of patient deaths  
• BHospitals that keep patients longer are likely to have higher death rates than those that discharge patients earlier but do not record deaths of patients at home after discharge  
• CPatients who are very old on admission to a hospital are less likely than younger patients to survive the same types of illnesses or surgical procedures  
• DSome hospitals serve a larger proportion of low-income patients, who tend to be more seriously ill when admitted to a hospital  
• EFor-profit hospitals sometimes do not provide intensive-care units and other expensive services for very sick patients but refer or transfer such patients to other hospitals