Power database: 3rd in a series
ICU quality being benchmarked by SCCM’s Project Impact effort
By BETTY GASCH
Healthcare InfoTech Contributing Writer
The recent Society of Critical Care Medicine (SCCM) Symposium held Jan. 23-27 in San Francisco included an update of the SCCM’s Project Impact, the society’s critical-care database product launched in 1995. Being developed in conjunction with Tri-Analytics (Bel Air, MD), the project intends to measure and describe the care given to Intensive Care Unit (ICU) patients.
Project Impact is quantifying practice patterns and patient outcomes within individual ICUs and then providing a comparison of local results for benchmarking of similar ICUs across the nation and, ultimately as the number of international users grows throughout the world.
This benchmarking capability is one of the strengths of Project Impact and a primary reason why hospital ICUs are drawn to using the database. Additional drivers are the current environment created by managed care, healthcare reform and cost containment efforts that require continuous quality improvement, and quality reports for accreditation. Finally, Project Impact is less expensive than implementing many commercial systems, such as Apache III or HCIA.
Each hospital that participates in Project Impact can choose to send a core data set to the national database on a quarterly basis. The data is compiled at the national level and then quarterly "benchmark" reports are generated, and sent back to the hospitals, thus allowing evaluation of their individual performance versus the national benchmarks. The reports allow the individual ICU to compare severity-adjusted statistics and answers questions such as the following:
• For each disease category, what percentage of the ICU patients experienced cardiac arrest or infections as complications?
• What was the length of stay for such patients?
• How do the patients in our local ICU, adjusted for severity of illness, compare to the national trends?
This benchmark information, plus the trends in each individual ICU’s performance, provides most of the continuous quality improvement data required by JCAHO. The Project Impact sample of hospitals is broad enough to reflect the "typical" ICU’s characteristics. Specific types of ICUs will differ, of course, depending upon a number of variables.
The SCCM believes that, on a typical day, 55,000 patients are actually treated in ICUs and that 33% are cared for by "intensivists," that is, physicians with a specialty in critical medicine who manage their ICUs. Other ICUs can be managed by either "generalists" or, in some cases, the patient’s attending physician. Project Impact is currently being used by 62 hospitals, including two in Brazil, one in Australia and one in Hong Kong. The majority of these hospitals submit data to be included into the national repository.
Research by SCCM physicians and others is beginning to suggest that better outcomes happen when an intensivist manager and a team of doctors, nurses, therapists and others deliver care to patients in ICUs, rather than when it is delivered by a single "specialist" or an attending physician plus lots of "consultants." SCCM has been encouraging this team approach, and it appears to have created a closer working relationship between SCCM and the American Association of Critical Care Nurses, as evidenced by the two groups’ joint presentations at the recent SCCM Symposium.
The Project Impact data further show that about half of the intensivists in practice are working in small- to medium-sized hospitals (100 to 300 beds) and that two-thirds of intensivists have a subspecialty of pulmonary medicine. Intensivists provide between 6 million and 7 million hours of patient care in ICUs per year. Intensivists are generally younger physicians, who unfortunately often leave their positions in critical care in their mid-40s. The higher the managed care environment, the more likely it is that "intensivists" are heading up the care in ICUs.
Knowing where ICU patients come from, their conditions and their disease severity is important in controlling ICU resource allocation. This is easily determined from Project Impact reports (as shown in Table 1).
The great majority of patients admitted to ICUs (91.6%) have pre-existing medical conditions at the time of their admission. And most patients admitted have more than one pre-existing condition, as shown in Table 2, also taken from the Project Impact database.
When time of day, severity and other data are added, a picture of the ICU practice and patient population at a particular ICU emerges. This is useful when determining what capitated payments are sufficient and when negotiating contracts with managed care payer groups for an individual hospital’s particular patient population.
Gathering and entering the data for Project Impact requires a 0.5 FTE employee for a 12-bed ICU. To ensure the entering of quality data, the data-entry person must know the ICU environment. The core product includes 20 built-in standard reports, but the system is able to be extended to include 150 more elements that may be defined by the individual hospital.
The SCCM is currently looking for funds or partnerships in order to bring Project Impact up to current database standards. The SCCM chairman of project impact is Milton McPherson, M.D., who is the Director of Critical Care Medicine at Northeast Medical Center (Concord, NC). Funds from user fees are apparently insufficient to support robust product development. If a hospital submits data to Project Impact, the system cost is $2,000 a year and $350 per licensed bed. If a hospital does not want to submit information, the cost is $5000 annually, but it does not receive the quarterly reports.
The Project Impact global database is still relatively small, including only 25,000 registered patients. In comparison to Project Impact, the National Registry for Myocardial Infarction (NRMI), a database of heart attack patients, started in 1990 and has 1,500 hospitals participating and over 1 million patients enrolled.
Once a hospital purchases the Project Impact database, it receives a training video. A one-day training course in Maryland is available for a fee, but the SCCM suggests that the hospital use the database on 30 or so patients prior to the training. Technical support is available Monday through Friday 8 a.m. to 5 p.m. EST, and after hours by e-mail, answered the next day. The SCCM reported the system to be Y2K compliant as of January ’99.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.