[Quality assurance in heart surgery. General and personal concepts]

Abstract

In the era of managed health care quality assurance has become more and more important. In cardiac surgery immense costs have to be justified. Some of the patients might be treated alternatively with drugs or by cardiological interventions. Additionally, the operative risk is still not neglectable in cases with substantial comorbidities and advanced age. Therefore in Germany, a nationwide quality assurance system was established in 1992 covering more than 90% of the cases in more than 90% of the centers. The goals of quality assurance have to address the needs of the patients, insurance companies, and surgeons, and thereby define the primary endpoints for analysis: mortality, quality of life, and therapy costs including reinterventions. A precondition for a sufficient quality assurance system is the isolation of quality relevant data by multivariate analysis and its documentation. Weighing the different factors allows a risk stratification in order to compare the results of various centers with different patient populations. For every center an expected mortality is calculated, and compared with the observed mortality. By taking the average mortality into account a risk adjusted mortality is derived for every center, which is independent from the patient population. An automated continuous follow-up of the patients is essential and allows assessment of 30-day mortality, reinterventions and quality of life. In 1986, the German Society of Cardio-thoracic-surgery initiated the development of a multicentric method of quality assurance (Quadra). In 1992 the insurance companies agreed to cover the costs for the reduced documentation of 205 parameters for every patient undergoing CABG, valve surgery or repair of an aneurysm with the help of cardio-pulmonary bypass; a 30-day follow up is included. Isolation of risk factors has not been performed because the quality of the data was not considered valid due to incomplete documentation and non-rigorous data control. For the first time in 1990, New York State Department of Health published data on risk adjusted mortality in CABG separate for every hospital. Due to a lawsuit brought on by the journal "Newsday" the department was forced to publish the results of every surgeon. Each clinic reports 41 different patient parameters, which are controlled for completeness and accuracy by an independent committee. The efficacy of this system has been proved by a 41% reduction in mortality from 1989 to 1992 mainly due to changes in patient management. Raw mortality rates to different hospitals in CABG were published by the administration of the insurance companies in 1986 leading to misinterpretations by the public. In response, the Society of Thoracic Surgeons initiated a database to calculate risk adjusted data. Today, more than 50% of all US-American centers participate and more than 700,000 patients have been reported. The data is not controlled and the majority of participating hospitals are small community centers. We have developed a methodology of quality assurance which estimates the operative risks of an individual patient by calculating the survival curve of all patients with the same risk profile who have already been operated on in our center. An automated long-term follow-up at one and five years after surgery provides the data for the calculation. The development of an effective quality assurance in the US was forced by media and insurance companies. Presentation of incorrect or misleading data in Germany has already damaged the image of cardiac surgery and surgeons. Therefore, there are no alternatives to public presentation of risk-adjusted mortalities in order to regain trust. Fears of surgeons and hospitals with results below the average are serious and patients with a high mortality risk may be afraid of not being operated on, although the New York System shows that these fears have not become real...