|
|
|
| The Vermont Health Care Quality Report 2008 |
Ch2: Measuring Quality of Care Ch3: Chronic Illness in Vermont
|
|
Methicillin-Resistant Staphylococcus Aureus (MRSA) in Vermont 2002-2006 (click for printable PDF) Introduction
Of note nationally, is the recent increase in the number of MRSA infections that are developing in community settings. In the past, MRSA infections typically occurred in healthcare settings like hospitals and skilled nursing facilities. At this time, MRSA infections are not necessarily reported to the Department of Health. As a result the only available information on the prevalence of MRSA in Vermont is hospital billing data. Hospital data allows us to trend the number of patients hospitalized that had positive cultures for MRSA. The data presented in this report does not differentiate between patients who developed MRSA infections during their hospital stay and those who entered the hospital already infected or colonized with the organism. This is a limitation of the detail available in hospital discharge data. Despite this limitation, VPQ feels that there is still value in monitoring the trend of MRSA prevalence in the hospitalized populations as it allows us some gauge of the prevalence of the bacteria in the state. Because of the rising incidence of community MRSA, some hospitals are screening certain patient groups as they enter the hospital to identify patients who are carrying MRSA without having the signs and symptoms of an infection. These patients are referred to as colonized or carriers. Two common spots where MRSA can be carried are on the skin or in the nasal passages. Active surveillance efforts can assist hospitals in their infection control practices, though the effectiveness of universal active surveillance (e.g. screening all patients on admission) is not clear at this time. As mentioned previously, administrative hospital discharge data cannot always distinguish a patient with a true MRSA infection from a colonized patient. Medical record coding practices vary between institutions, allowing this distinction to be made at some but not all hospitals in the state. The findings below shed some light on how prevalent hospitalizations with evidential documentation of MRSA may be in Vermont, while also allowing us to compare the characteristics of these hospitalizations with hospitalizations where there is no documentation of MRSA. Data source and Methodology Vermont Hospital Discharge Data sets for calendar years 2002 – 2006 were used to conduct all analyses. The unit of analysis is a hospital stay. Patients with more than one hospital stay will have each hospitalization counted separately.
Findings § Hospitalizations with evidential documentation of either MRSA infection or colonization have been steadily increasing over the past five years. In 2002, 0.5% of hospitalizations were coded with documented MRSA; in 2006 the rate had increased to 0.85%. § MRSA was more likely to be documented in male patients (54%). § MRSA was more likely to be documented in older adults. Over half (55%) of hospitalizations with an indication of MRSA were 65+ years of age. This was double the rate for other age groups: 27% for age 45 to 64; 13% for age 18 to 44; 2% for age 13 to 17; and 3% for age 0 to 12. § Consistent with the findings by age, 63% of individuals with documented MRSA had Medicare as the primary insurer followed by commercial insurance (18%) and Medicaid (13%). § On average, hospitalizations with documented MRSA were approximately twice as long than non-MRSA hospitalizations (14.2 vs. 5.2 days) and charges for these hospital stays totaled twice as much as non-MRSA hospitalizations ($40,672 vs. $18,103) when similar DRGs are compared.
§ Even though the number of hospitalizations with indication of MRSA has increased, the percentage of in-hospital deaths among patients with an indication of MRSA decreased over the 5-year period. The decline was greatest between 2005 and 2006 (32%) and has decreased 47% since 2002. The percentage of deaths among patients without an indication of MRSA remained constant (approximately 2% each year). § The most common diagnoses in patients with evidential documentation of MRSA were: pneumonia (11%); skin and subcutaneous tissue infections (10%); complications of device, implant, or graft (8%); complications of surgical procedures (8%); septicemia (5%); and diabetes mellitus without complications (5%). § The most common procedures in patients with evidential documentation of MRSA were: other vascular catheterization (not heart) (12%); debridement of wound, infection, or burn (8%), respiratory intubation (8%), amputation of lower extremity (6%); and incisions and drainage, skin and subcutaneous (5%). Discussion The findings reported here are largely consistent with those reported in previous studies. The percentage of hospitalizations with evidential documentation of MRSA in recent years in Vermont (0.5% in 2002 to 0.85% in 2006) is similar to the rate reported nationally (0.7%)1 in a study employing similar methodology. Nationally this methodological approach has been corroborated by a recent study using a rigorous methodological approach and data extracted from reviewing medical charts4. At this time, it is unclear whether the same holds true on a smaller, local scale. Other research, however, suggests, administrative coding data is not a viable source for identifying healthcare-associated infections.5 The methodological approach employed in this study has a few limitations. First, it may overestimate MRSA infections since for some hospitalizations the patient may have been a carrier of the bacteria but not have had the signs and symptoms of an infection. Hospitals using active surveillance methods to screen for MRSA (i.e. testing patient groups such as all patients admitted to the intensive care units) may report more hospitalizations with documentation of MRSA. Many of these are likely to be cases where the patient is a carrier of MRSA rather than have an active infection. Some, but not all hospitals use modifying codes that allow carriers to be distinguish from infections, but not all institutions use such coding practices. Second, since the unit of analysis is a hospital stay, an individual who is admitted to the hospital more than once will have each hospitalization counted separately. This may also result in an overestimate of the actual number of MRSA cases if individuals have several hospitalizations in which MRSA is identified and coded. Third, the analysis assumes that all MRSA infections are identified and coded. This may result in either an overestimate of MRSA or an underestimate of MRSA and may vary from hospital to hospital. To fully understand the impact of the limitations discussed, a validation study with individual patient chart review at each hospital would be required. While this study has limitations, it does provide a platform for understanding potential data sources for identifying MRSA and a methodology for approximating the trend and severity of MRSA locally. Future research efforts could focus on validating these results through chart reviews. An in-depth validation study could also explore to what extent there are differences among hospital’s coding practices. Future research should also be focused on distinguishing between community onset MRSA infections and hospital onset MRSA infections. The prevention of healthcare associated infections is a priority of all Vermont hospitals. Several specific initiatives with defined process and outcome measures have been included in the Vermont Hospital Report Cards (www.vthospitalreportcards.info)6. The hospital comparison report for Prevention and Control of Antibiotic Resistant Infections identifies those actions each hospital has taken to prevent infection and transmission of multi-drug resistant organisms including MRSA. The range of strategies being employed varies and hospitals are using multiple approaches to address MRSA. Staff education, more aggressive hand hygiene programs, stricter use of contact precautions (e.g., gloves and gowns) for patients who test positive for MRSA, more emphasis on housekeeping/decontamination practices and targeted active surveillance testing are examples of interventions being used. In addition, recent work by the Vermont Infection Control Practitioners Association and the Vermont Department of Health have resulted in the publication of a patient education booklet, “Living with MRSA”7, that is being distributed to hospitals, nursing homes, school nurses, and veterinarians around the state. The publication is available as a hard copy or on-line from the Vermont Department of Health. Work to prevent healthcare associated multi-drug resistant infections including MRSA is ongoing. For more information about what is happening in your community we encourage people to contact their local hospital. References 1) Elixhauser, A and Steiner, C. Infections with Methicillian-Resistant Staphylococcus Aureus (MRSA) in U.S. Hospitals, 1993-2005. HCUP Statistical Brief #35. July 2007. Agency of Healthcare Research and Quality, Rockville, MD. http://www.hcup-us.ahrq.gov/reports/statbriefs/sb35.pdf 2) Panlilio AL, Culver DH, Gaynes RP et al., Methicillian-resistant Staphylococcus aureus in U.S. hospitals, 1975-1991. Infection Control and Hospital Epidemiology. 1992; 13:582-586. 3) Kuehnert MJ, Hill HA, Kupronis BA, Tokars JI, Solomon SL, Jemigan DB. Methicillin-resistant-Staphylococcus aureus Hospitalizations, United States. Emerging Infections Diseases. 2005; 11(6):868-872. 4) Klevens R. Monina et al., Invasive Methicillin-Resistant Staphylococcus aureus Infections in The United States. Journal of the American Medical Association. 2007;298(15):1763-1771. 5) Stevenson KB, Khan Y, Dickman J et al. Administrative coding data, compared with CDC/NHSN criteria, are poor indicators of healthcare-associated infections. American Journal of Infection Control. 2008; 36(3):155-164. 6) Vermont Department of Banking, Insurance, Securities, and Health Care Administration. Vermont Hospital Report Cards. 2008 Report. www.vthospitalreportcards.info. 7) For the “Living with MRSA” booklet and other MRSA related resources visit: http://www.vpqhc.org/MRSA.htm
|
|
