Search

ABSTRACT Analysis of ballistics evidence (spent cartridge casings and bullets) has been a staple of forensic criminal investigation for almost a century. Computer-assisted databases of images of ballistics evidence have [...] more...

Health care can be improved either by identifying and educating physicians who provide substandard care or by a plan of continuous quality improvement. The solution depends on whether a few practitioners make most of the errors, or whether a few errors are made by many different practitioners. If the latter is true, targeting individual physicians is a less efficient strategy than investigating clinical processes. Other studies have found that the risk of malpractice claims varies among specialties and among individuals within specialties. Malpractice data from 1,371 physician claims were used to find out if error-prone physicians could be identified, and what factors were associated with errors. In addition, hospital characteristics were studied to determine if they were associated with types of clinical errors. Past claims were only moderately predictive of future malpractice claims, although physicians with many past claims were likely to incur many more. Other ways to determine negligence-prone physicians must be found, such as sampling office records of practitioners who have made especially serious errors or who are in high-risk specialties. Part of the difficulty with using malpractice claims is that they are actually very rare. Within hospitals, specialty was the most important predictor of physician error pattern, followed by ratio of hospital staff to admissions. Negligence claims do not provide a good basis for targeting individual physicians, but may identify some very incompetent physicians. They may be even more useful on a national basis for identifying clinical errors, and leading to improved clinical processes. (Consumer Summary produced by Reliance Medical Information, Inc.) more...

Accountability in health care is an important issue, but measurement of quality is difficult. Medical malpractice claims records provide important information about a valuable set of experiences: cases where poor care has caused bad outcomes. An evaluation of 1,371 malpractice claims was undertaken to identify areas of medical care where poor outcomes were preventable. Four specialties were included: obstetrics and gynecology, general surgery, anesthesiology, and radiology. There are two problems with using malpractice claims alone for this purpose. First, most cases of physician negligence never generate claims; and second, many malpractice claims are not a result of physician negligence. Therefore, malpractice cannot be the only criterion used to measure health care quality, but it is a start. It was found that patient management errors were more frequent than other errors, and caused more deaths. These errors also resulted in higher indemnity payments. Error types varied by specialty, indicating that specialties should not be combined when looking at error profiles. Omission of diagnostic tests and patient monitoring resulted in only a small fraction of negligence claims. Problems in communication were rare, but were more common in obstetrics and gynecology. Problem areas and some potential solutions are briefly outlined for each of the four specialties. In reviewing the cases, system failures, such as poor coordination or communication among health care personnel, were responsible for many of the errors that led to malpractice. (Consumer Summary produced by Reliance Medical Information, Inc.) more...

Peter SKERRY. Washington, DC: Brookings Institution Press, 2000. ISBN 0-8157-7964-X. viii 4261 pp. $25.95 (P) Two books on census-taking are reviewed in this issue. One, Who Counts? The Politics of [...] more...

Presents a stochastic model of the employment process in which both the worker's search for jobs and the employer's search for workers are simple Markov processes. Use of the dynamic programming methods to find the optimal hiring policy; Discussion of white employment and transient behavior of workers; Derivation of the steady-state distribution of worker unemployment by type. more...