insertion of tubes in children’s eardrums in the absence of clinically appropriate indications; and
performance of chiropractic spinal manipulation for certain back conditions for which there is no evidence of benefit.
Lack of insurance is a major contributing factor to underuse. Even with comprehensive insurance coverage, however, much of the population fails to receive recommended preventive services, and many patients do not receive the full range of clinically indicated services for acute and chronic conditions. Examples include the following:
Cardiac care In a study of 3,737 Medicare patients with a diagnosis of heart attack who were eligible for treatment with beta blockers, only 21 percent were found to have received beta blockers within 90 days of discharge. The adjusted mortality rate for patients with treatment was 43 percent below that of patients without treatment (Soumerai et al., 1997).
Pneumococcal vaccine In 1989, the U.S. Preventive Services Task Force recommended that people 65 years and older receive a one-time vaccination for pneumonia, and in 1996, this recommendation was modified to apply to all immunocompetent people aged 65 and older. Yet studies of the proportion of elderly who had been vaccinated produced estimates in the range of only 28 to 36 percent (CDC, 1995; Kottke et al., 1997).
Acute care for pneumonia Two studies of hospitalized patients with pneumonia found serious shortcomings in the proportion of patients receiving appropriate components of care (Kahn et al., 1990; Meehan et al., 1997).
In recent years, increased attention has been focused on misuse. Studies of misuse are particularly challenging because actual or potential adverse events often go undocumented and unreported. But studies of preventable deaths and adverse drug events point to frequent and sometimes serious errors. For example, one study of over 4,000 hospitalized patients found that there were 19 preventable or potential adverse drug events per 1,000 patient days in intensive care units and 10 preventable or potential adverse drug events per 1,000 patient days in general care units (Cullen et al., 1997).
The existing literature does not allow a comprehensive estimate of the burden of harm due to poor quality. The literature on health care quality covers only a portion of the full range of quality concerns. For the most part, published studies focus on individuals who come into contact with the health care system. From a population perspective, the opportunity cost of poor quality must also