By Carol A. Kemper, MD, FACP, Clinical Associate Professor of Medicine, Stanford University, Division of Infectious Diseases; Santa Clara Valley Medical Center, Section Editor, Updates; Section Editor, HIV, is Associate Editor for Infectious Disease Alert.

More on Benchmarking…

Source: Klompas M, Platt R. Ventilator-associated pneumonia — the wrong quality measure for benchmarking. Ann Intern Med. 2007;147:803-805.

Increasingly, hospitals are being required to make public their surveillance data on hospital-related infections as quality indicators. In addition, the Bush administration and CMS have stated that Medicare reimbursement and accreditation will be increasingly linked to these quality indicators, and that Medicare payment will not be provided for certain hospital-related infections and complications.

Tangible improvements in the quality of care is a laudable goal, but data clearly linking several so-called quality indicators to improved patient outcome is lacking. It's the worst possible medicine (Gee, this seems like it should be a good idea, so lets do it). If physicians actually practiced in such a non-evidence-based manner, patient care would be in serious trouble. Witness the ongoing debacle with community-acquired pneumonia "bundles," where every person hitting the emergency room door with cough and fever is given antibiotics, even before they are given a diagnosis. And this manipulation of good clinical practice is occurring against a backdrop of sky-rocketing C. difficile cases in many hospitals. (So, the very organizations that demand universal administration of antibiotics for every possible CAP are now refusing to pay for the consequences of antibiotic overuse?).

In addition, the measurement of some quality indicators may be fraught with variability and subjectivity (Yes, the CVP line was needed for one more day). For these reasons, these authors argue that ventilator-associated pneumonia (VAP) surveillance should not be used as a quality indicator for hospital care. The diagnosis of VAP requires a combination of several objective measures (eg, white blood count > 12 x 109 cells/L), multiple subjective measures (eg, change in character of sputum, increased respiratory secretions, worsening cough), and radiographic findings of persistent pulmonary infiltrates. In the ICU setting, pulmonary infiltrates can often be due to non-infectious causes; increased respiratory secretions could be for multiple reasons, including simply failure to adequately clear secretions; and bacterial cultures in intubated patients can easily represent colonization. For all these reasons, the clinical diagnosis of VAP is "notoriously inaccurate," and can lead to very different rates of VAP for different hospitals.

The original surveillance definition for VAP proposed by the CDC acknowledged the ambiguity of these clinical signs and symptoms, and the error rate inherent in that assessment. At least one-third of those patients diagnosed with VAP have no evidence of pneumonia at autopsy. Conversely, one-fourth of ventilated patients without a clinical diagnosis of PNA who come to autopsy are found to have pneumonia. In prospective assessments, only 30%-40% of intubated patients with fever, purulent secretions, and abnormal chest radiographs actually have VAP.

The intent of the CDC was to provide a mechanism for hospitals and their infection control staff to internally monitor their critical care over time, so that appropriate measures could be taken as needed. To make this a mandatory reportable to the public, and a basis for financial reimbursement, risks undermining that original objective turn this instructional epidemiologic tool into a numbers game.

Zinc Beneficial in PPD Testing

Source: Rao VB, et al. Zinc cream and reliability of tuberculosis skin testing. Emerg Infect Dis. 2007;13:1101-1104.

Zinc is known to have important affects on white blood cell function, antimycobacterial immunity, and cutaneous responses to intradermal PPD. Zinc deficiency is not uncommon in poorer communities, especially in those with malnutrition or active tuberculosis.

In order to improve the utility of PPD testing in poorer populations, Rao and colleagues examined skin test responses to PPD in 50 shanty town residents in Lima, Peru. PPD was injected intradermally into the volar surface of each forearm, one of which was covered with a placebo cream and the other was covered with a cream containing zinc sulfate (dissolved in an aqueous cream to a concentration of 1% elemental zinc). These areas were promptly covered with an occlusive dressing. Assessments of induration were made at 24, 48, and 72 hours. The skin test results were compared with plasma concentrations of zinc.

Ten percent of study subjects were underweight, and plasma zinc concentrations were deficient in 31%. Skin test responses were significantly less in patients with lower plasma concentrations of zinc. For those who were zinc deficient, the average skin test response in the control arm was 14 mm, compared with 27 mm in subjects with normal zinc levels (P = .03). Applications of zinc cream resulted in larger areas of induration by an average of 32% compared with the other forearm, and skin tests were more likely to be positive in zinc-treated arms. The affect on the skin test response was greatest in those with the lowest plasma zinc levels.

Zinc cream applied to PPD skin testing sites can augment the skin test response, yielding higher rates of positive results and enhancing the utility of skin testing. Topical zinc creams are an easy and inexpensive way to improve skin test responses, especially in populations at risk for zinc deficiency.

Female physicians at greater risk for suicide

Source: Peterson MR, Burnett CA. The suicide mortality of working physicians and dentists. Occupational Medicine Advance Access published October 27, 2007

Female physicians have more than twice the rate of suicide as other professional women and are proportionally at greater risk compared with their male physician counterparts. That was the unhappy conclusion of these authors who examined the National Occupational Mortality Surveillance Data for 26 states in the United States from 1984-1992. Age-standardized suicide rates were calculated for male and female physicians and male dentists (there were too few female dentists to assess); data for white vs non-white workers were also examined.

White male physicians > 45 years of age had two times the rate of suicide as their white female physician counterparts. But when proportional risk assessments to other working groups were made, women were at far greater risk relative to their working professional female colleagues. In contrast, because the overall rate of suicide for men in the general population is 5 times higher than that for woman, the proportional rate for male physicians relative to their male counterparts was significantly less than the proportional rate for female physicians relative to their female counterparts. (White male dentists had similar suicide rates to white male physicians). Suicide rates for male physicians was similar to that for other male professionals, but lower than non-professionals. In addition, suicide rates for man < 45 years of age were lower than then their older colleagues, and clearly increased with age. Suicide rates in women were not age-related.

Similar results were observed by the AMA in the 1960s-1970s. That earlier data also found a higher risk of suicide in female physicians, but also found that location may be an important factor, as may the physician speciality, neither of which was examined in the current study.

Raw milk Campy Outbreak

Source: ProMED-mail Post, December 4, 2007; www.promedmail.org

It is always fun to pimp medical students on infections possibly transmitted by raw milk, most of whom are too young to have seen or drunk unpasteurized milk. A partial list includes listeria, brucella, campylobacter, salmonella, Mycobacterium bovis, and E. coli 0157:H7. But raw milk is making a comeback in many states. At least 27 states in the United States now permit the sale of raw milk, including California, where it can now be purchased just down the road from Stanford Medical Center at the local family market. Previously, most of the cases of raw milk-related infections seen in this area were the result of soft cheeses ingested or illegally brought in from Mexico.

An outbreak of campylobacteriosis occurred this fall in Kansas, involving at least 87 people, which has been attributed to ingestion of raw milk from two different dairies in the state. One dairy supplied the milk for a soft cheese for a large community event, resulting in 68 cases, including at least 2 patients who were hospitalized. While most cases of campylobacter enteritis are self-limited, some patients may develop bacteremia, hemolytic uremic syndrome, or later complications, such as reactive arthritis and Guillain-Barré. ID physicians should be alert to the possible connection of this pathogen to ingestion of locally produced raw milk.