reports

Burden of illness in defined populations

Stockholm, Sweden: Karolinska Institute.

Karolinska Institute, Stockholm, Sweden

This thesis addresses possibilities of describing the burden of illness in a population by elucidating morbidity and comorbidity patterns in various groups of patients using information on the individual level from health care registers.

In the first study, a cost-of-illness analysis of one type of disease was performed to assess the annual direct and indirect costs of skin diseases caused by ultraviolet radiation. Data on the individual level were retrieved from health care registers in one county council. Direct health care costs for diagnosing, treatment and secondary prevention, as well as indirect costs caused by morbidity and mortality, were included in an economic burden analysis. In 1999 the total annual cost of illness for skin diseases caused by ultraviolet radiation exposure in Stockholm county was approximately 162.4 MSEK, with indirect costs comprising about 56% of total costs.

In the second study, data were retrieved from electronic patient records from one primary health care centre in a Swedish municipality. Patients utilizing primary health care during 1998 and 1999 were categorized into 81 groups. Grouping was carried out using the Johns Hopkins Adjusted Clinical Groups^® (ACG) risk adjustment system. The resulting pattern provided a view of the scope of primary care that differs from that yielded by statistics on diagnoses. Changes over time in terms of morbidity patterns for a population could be described and analysed. The ACG system seemed to be a feasible and relevant tool for describing the outcome of work done at a primary health care centre.

The third study retrospectively processed encounter data from all publicly managed primary health care centres in one county council in Sweden. The objective was to elucidate types of morbidity and categories of patients in a large population, and the ACG system was applied. Types of morbidity in primary care seemed to be dominated by nearly equal proportions of “time limited”, “likely to recur”, “chronic” and “signs/symptoms”. The predominant categories of patients were those with only one type of morbidity. About one third of all patients studied had a constellation of two or more types of morbidity during a one-year period.

This thesis illustrates that retrieval of clinical data on the individual level is appropriate for grouping diseases as well as patients. The three studies demonstrate the possibilities of using clinical data on an individual level to indicate the economic as well as the clinical burden of illness in defined populations. The ACG system yields a new view of the burden of illness in a population and may be a relevant tool for supporting quality improvement in primary care focusing on clinically meaningful categories of patients.