Use of methodological standards in diagnostic test research: getting better but still not good

Article Abstract:

Diagnostic test research in selected medical journals does not strictly adhere to established methodological standards. Researchers selected 112 studies on diagnostic tests published in the New England Journal of Medicine, Journal of the American Medical Association, British Medical Journal, and Lancet over a 16-year period. Only one of the studies, published in the Journal of the American Medical Association, fulfilled all seven standards set forth by researchers. These standards involved spectrum composition, analysis of pertinent subgroups, avoidance of workup bias, avoidance of review bias, precision of results for test accuracy, presentation of indeterminate test results, and test reproducibility. Adherence to the standards has improved over the 16 years, but between 1990 and 1993, less than half the studies used three or more of the standards. Diagnostic tests should be evaluated by the methodological standards so that substandard tests do not become widely used, which in turn could increase health care costs and impact patient care.

Evaluation, Medical research, Methodology, Research methods

---

A clinimetric approach to the components of the patient-physician relationship

Article Abstract:

Physicians can break the physician-patient encounter into separate components to understand how they can interact better with the patient. The encounter includes the hopes and wishes of both patient and physician as well as any reactions during the encounter itself. The hopes of the patient may conflict with the hopes of the physician. Physicians should also adopt an open and tolerant communication style and ask patients directly what they expect to get out of the visit. Physicians should comply with the patient's request if possible or explain their reasons for refusing and try to negotiate a compromise.

Social aspects, Physician and patient, Physician-patient relations, Medical examination, Periodic health examinations

---

Context bias: a problem in diagnostic radiology

Article Abstract:

Radiologists may be more likely to identify an abnormality on a diagnostic image if the abnormality is present on a large percentage of the films being reviewed. Researchers gave 6 radiologists 2 sets of 40 lung arteriograms to review. One set had arteriograms with a low percentage of abnormal images and the other had a high percentage of abnormal images. Twenty-four images were common to both sets and one set was reviewed 8 weeks after the other. The radiologists were more likely to change their opinion about the diagnosis when they reviewed the set with a high percentage of abnormalities.

Analysis, Practice, Diagnostic imaging, Radiologists

Similar abstracts:

• Abstracts: Relining, rebasing partial and complete dentures. Making better removable partial dentures. Abnormal occlusal conditions: a forgotten part of dentistry
• Abstracts: Managing diabetes: knowledge for practice. Diabetes care: getting your patients on a sure footing. Spreading the word on insulin therapy
• Abstracts: Expanding the skills needed in clinical supervision. Conducting a research interview. Taking counsel
• Abstracts: Middle man in the prostate puzzle. Rapid research response emphasized in unique prostate cancer program. Aging heart adapts to change; researchers begin to unravel some mechanisms involved
• Abstracts: Nocturnal hyperventilation in pregnancy - reversal with nasal continuous positive airway pressure. Sleep in multiple pregnancy: breathing patterns, oxygenation, and periodic leg movements

This website is not affiliated with document authors or copyright owners. This page is provided for informational purposes only. Unintentional errors are possible.