In the News

SGIM research targets health disparities, drug errors

MIAMI—Health care disparities were a theme among research abstracts selected for presentation at the Society of General Internal Medicine's annual meeting last week.

In one study, researchers used data from the Women's Health Initiative to evaluate the association between a woman's risk of coronary heart disease (CHD) and the socioeconomic status (SES) of her residential neighborhood. After adjusting for a number of factors (including hypertension, hyperlipidemia and obesity), they found that a 1-point drop in SES equated to 1% increase in heart disease risk. For example, a woman living in a poor neighborhood in southeast Washington D.C. would have a 50% higher risk of CHD than a woman living in the wealthier northwest end of the city.

Another study looked at the impact of different patient populations on physician quality measures. Researchers in the Massachusetts General Hospital system rated 159 physicians' performance on several cancer screening and chronic care quality measures. They then looked at differences in the characteristics of the physicians and in 87,000 of their patients. The study found that physicians who rated in the bottom quartile for performance treated significantly more non-English speakers, minorities and nonpaying patients, leading them to conclude that physician quality measures may represent the composition of a doctor's patient population, rather than just the quality of his or her care. The study also found that female physicians met more measures in the treatment of female patients, while male physicians scored better treating men.

During the same session, researchers from Brigham and Women's Hospital showed how implementation of bedside barcode scanning affected medication errors at their hospital. The study had clinical nurses record administration of medications given to patients on the medical, surgical and intensive care wards and then compare the drugs actually administered to the orders given to check for errors. Before the barcodes to match patients to medications were implemented, 11.6% of administrations resulted in errors. With barcodes, the error rate declined to 7%. Pre-barcode, potential adverse drug events occurred in 3.2% of cases, compared to 1.6% after implementation. The study authors noted that although their results are not necessarily generalizable to other hospitals, the study does indicate that barcode systems have the potential to dramatically reduce adverse drug events.

Pain management guidelines updated for geriatric use

The American Geriatrics Society updated guidelines nearly eliminating non-steroidal anti-inflammatory drugs (NSAIDs) in drug management of persistent pain in elderly patients.

The society focused on drug therapy and those age 75 or older in this latest revision because of new drugs and treatment approaches available, and because drugs are the most common strategy for managing persistent pain among the elderly.

The major change is the near elimination of NSAIDs. Newer information suggests increased cardiovascular risk and gastrointestinal toxicity usually outweigh the benefits. Based on newer clinical trials as well as clinical observation, the panel recommends that NSAIDs and COX-2s be considered rarely, and with extreme caution, in highly selected individuals (high quality of evidence, strong recommendation).

The guidelines also recommend that all patients with moderate-severe pain or diminished quality of life due to pain should be considered for opioid therapy, which may be safer than long-term NSAIDs. The authors provide some discussions and recommendations about the use of adjuvant and other drugs for older persons with recalcitrant pain problems.

The guidelines and a patient tip sheet are online, and the guidelines will be published in the August issue of the Journal of the American Geriatrics Society.

Hysterectomy, ovary removal tied to mortality

Women who have their ovaries removed at the time of hysterectomy are more likely to die than those who don't, a new study found.

In a prospective, observational study, researchers evaluated 29,380 women from the Nurses' Health Study, 56% of whom had hysterectomy with bilateral oophorectomy, and 44% of whom had hysterectomy with ovarian conservation. Researchers evaluated incident death or events from coronary heart disease (CHD), stroke, breast cancer, ovarian cancer, lung cancer, colorectal cancer, total cancers, hip fractures, pulmonary embolus and all-cause mortality. The study is in the May 2009 Obstetrics & Gynecology.

Bilateral oophorectomy patients were 12% more likely to die during 24-year follow up. They also had a 17% higher risk of CHD, a 14% higher risk of stroke, a 26% higher risk of lung cancer and a 17% higher risk of total cancer. Their risk of breast cancer, however, was 25% lower, and their risk of ovarian cancer was 96% lower. The risk of death, stroke and CHD was higher for those who had their ovaries removed before age 50 and never used estrogen therapy.

Prophylactic oophorectomy, with the goal of improving survival by reducing ovarian cancer, is not supported by the study, the authors noted. About 300,000 U.S. women per year undergo elective oophorectomy, they said.

IOM calls for change on conflicts of interest

Physicians, institutions and physician organizations need to make significant changes to reduce conflicts of interest, according to a new report from the Institute of Medicine.

The report calls on physicians to forgo gifts of any amount from medical companies and to decline to publish or present any material ghostwritten or controlled by industry. Interactions with pharmaceutical reps should be limited and samples should be used only for patients who cannot afford medication, the report said.

The recommendations also call for greater disclosure of industry ties. Physicians should notify medical organizations of their financial connections and Congress should require device and drug manufacturers to publicly disclose payments made to physicians, patient groups, academic health centers and professional societies.

The professional societies and other groups that develop guidelines should not accept direct industry funding for the work, the report said.

Anyone with a financial conflict of interest should not participate in guideline development. There's also a need for an overhaul of continuing medical education to make it free of industry influence, according to the IOM. The medical profession should adopt these changes voluntarily, or risk having them implemented by legislation, warned the expert panel which wrote the report.