Health care rationing in Germany

In the years to come demographers expect the elderly population in Germany to grow due to both low birth rates and rising longevity. It has generally been assumed that population aging leads to increasing health expenditures. Data usually exhibit a linear increase in health care expenditures with age. But studies from scientists in the U.S.A question this assumption. For example, Lubitz and colleagues from the American Health Care Financing Administration show that older patients in hospitals induce lower costs than younger patients. In 1988, Medicare costs for deceased patients 65 to 69 years of age averaged $15.436, whereas costs for deceased patients 90 years of age or older averaged only $8.888. "Unfortunately there were hardly any comparable studies in other countries," says Dr. Hilke Brockmann from the Max Planck Institute for Demographic Research in Rostock. Thus, it was impossible to say whether these decreasing medical expenditures at higher ages are an American pattern due to the specific feature of a private health insurance regime. "In addition, it was completely unknown whether sex plays also a role in discriminating against older people in hospitals", says Brockmann. That is why she started a research project based on data obtained from Germany's largest health insurer (AOK) in two different regions; Westphalia-Lippe (Western Germany) and Thuringia (Eastern Germany). The data include the costs of hospital stays of AOK members between 1996 and 1997 who were at least 20 years old. The descriptive analyses reveal the total annual expenses for deceased and non-deceased patients and the age-specific distribution of diseases among the deceased patients. The influence of age discrimination was tested with multivariate regression models. The study examines the severity of the illness and multi-morbidity, as well as the length of stay, the kind of treatment and the place of treatment.

Patients that died in 1997 and survivors were analyzed separately. It turns out that nearly 40 percent of total hospital expenditures went towards the care of 10 percent of the patients. The most expensive patients were females in Westphalia-Lippe aged 20 to 49 years who died of cancer. Average costs of 36,300 DM (18,560 EUR) exceed treatment expenses for the younger group four to five times. In Thuringia costs of treatment were highest for those 55 to 59 year old women who died within one year. Expenses for life-threatening diseases drop decisively after the age of sixty. In general, hospitals in Eastern and Western Germany provide more expensive treatment for non-surviving women than for non-surviving men of the same age group through very old ages. In contrast, treatment for surviving male patients is costlier than for surviving woman. The only exception is women from Thuringia who were 85 and older.

It is surprising that, despite different health care systems in U.S.A. and Germany, reductions in age-dependent health care spending are similar. "The cost for treating patients 90 years and older in both German regions is somewhat half of that of treating 65 to 69 year-olds," Brockmann explains. These differences can only be partially explained by age-specific disease patterns.

Health expenses for women aged 20 to 60 are usually due to breast cancer or other tumours, whereas older women suffer more from cardio-vascular diseases. Men's cancer rates increase after the age of 65. Older men and women usually die from cardio-vascular or respiratory diseases.

Using the ICD-classification (International Classification of Diseases), a widely accepted norm in health care, it is most interesting to compare cost of illness with the therapy costs according to age. "Here we can see that significantly less is being spent on patients 80 and older who have the same ICD-classification," says Brockmann. The cost of hospitalisation were 50 percent lower than those of clinic residents who had either not reached the legal retirement age or who were only a few years past it.

Old and terminally ill patients were responsible for the lowest costs overall. Declining expenditures indicate health rationing, where expenses and outcome are weighed against one another, rather than striving for an optimal medical outcome. One explanation for health care rationing could be that patients do not want intensive care. "It is conceivable that the most elderly in-patients want less medically aggressive, i.e. less costly, treatment," explains Dr. Brockmann. According to some surveys, the majority would rather die at home than in hospital. The reality is somewhat different, as nearly 50 per cent of all people die in hospitals. Brockmann supposes that age-specific cost patterns are caused by clinical decision making. For example: age is an indicator of probability of death, and a death that occurs during treatment might be considered a medical failure. Furthermore, there is less knowledge about the treatment of older patients and the most elderly. Like babies, the elderly have been largely excluded from clinical trials in order to minimize analytical problems caused by co-morbidity or an increased risk that participants die during the course of the study.

The question remains as to whether lives at advanced ages could be saved if age rationing were discontinued and maximum medical treatment were to be applied to everyone, irrespective of their age. Brockmann plans to carry out a follow-up study using data from different health insurers in Germany and other countries.

Figure 1: Unevenly distributed expenses of all hospital patients: the Lorenz-Curve. Diagram: Max Planck Institute for Demographic Research

Figure 2: Yearly hospital expenses of non-surviving and surviving patients. Diagram: Max Planck Institute for Demographic Research

---

---