[Hospital volume and quality of care: Health technology assessment and international literature review]

PURPOSE: In this report, the importance of the patient count for the treatment quality at hospital and at the level of the doctor is evaluated. The report also assesses the relevance of the results of international scientific literature to the Norwegian healthcare system. METHODS: The report of York ("Concentration and choice in the provision of hospital services" (1997)) has been the basis for the work. There are further searches for primary studies of scientific literature from medical databases for the period 1997-2000. Studies that satisfy the given criteria (see the main report) are then examined. Special attention has been paid to a satisfactory assessment of the patient mix ("case mix"). Moreover, information from the Norwegian Patient Register has been obtained for a selection of procedures conducted at Norwegian hospitals in the period 1995-98. The method assessment was carried out by an expert group at the Center for Medical Methodology Assessment under management by Professor Per Teisberg and with Dr. Philos Inger N.Norderhaug as Project Coordinator. RESULTS: 112 scientific articles published after the York report release are considered. Of 87, it was found that hospital or leg volume has a bearing on treatment quality. Quality is most often expressed with mortality or complication rates. The research activity around volume quality has mainly included surgical interventions. The documentation of a volume-quality context is best for patients with cancer or cardiovascular disease treated with surgery. Likewise, it is documented the quality of organ transplantation and AIDS treatment is better at the hospital's high volume. However, in the case of trauma and at orthopedic surgery, no consistent relationship between volume and mortality or complications has been demonstrated. (The studies concerning injuries are from the United States The treatment is organized in ITRA centers with different levels of competence.) In other diseases or operations studied, there is insufficient basis for drawing safe conclusions. The report also presents an overview of the activity level at Norwegian hospitals for surgical procedures and medical diagnoses. The survey shows that many Norwegian hospital practitioners planned surgical procedures and medical treatment with a volume that was very low. COMMENTS: Scientific work on volume-quality relationships is exposed to multiple possible sources of error. An important aspect of this report is that such sources of error have been discussed and the studies are evaluated critically to reduce the effects of the error sources on the results. Knowledge of volume quality is mainly based on information from studies conducted in IUSA. Only five of the studies in this and the York report are from Norway or include Norwegian hospitals. Knowledge of conditions at Norwegian hospitals should ideally be used as a starting point for assessments of volume-quality dimensions in Norwegian health services. Such knowledge is to some extent available, or may be, through clinical quality registers. However, the findings in the literature that are considered are relevant to Norwegian conditions. Although the evaluated scientific evidence indicates that higher volumes provide better quality-like quality is defined here, this is not a conclusion that can be automatically transferred to other areas of medical treatment. The volume-quality relationship has been studied extensively in the treatment of patients with chronic disorders and non-surgical diseases. The size of the volume effect is also not evaluated against other factors that are crucial to the quality. There are also studies that show no connection between volume and quality, and for some medical areas, the study is contradictory. The quality of treatment in this and York report is most often expressed as mortality and complications associated with or after treatment. There are very few scientific workers dealing with long-term effects, for example. function level, long-term survival rate of implant and reoperation rates (eg by joint prosthesis). In some areas where there is scientific evidence, the relevance for Norwegian relations is clear. Some hospitals have such a small volume that one must fear that the quality is not optimal. In addition, the volumes are so small that it is doubtful whether hospitals can say something about their own quality in a conciliatory way. Creating quality records for areas where the volume dimension is important will therefore be an important quality enhancement measure in Norway. This report will provide a basis for discussion in both the academic and social policy circles. The report is a challenge for the professional communities in relation to the requirement to document results in the work performed. It challenges health politicians in relation to planning, task allocation and quality assurance. The volume quality dimension is one of many factors that must be emphasized and must be balanced against other factors, such as accessibility, geographical equality and patient satisfaction. SMM would like to thank the members of the Expert Group for an impressive work.