Background

Longitudinal studies on clinical categories of patients in primary health care are rare, as are studies of the individual co-morbidity. In Sweden diagnoses have been recorded and stored in electronic patient records for a long time but so far have not been used for follow-up on categories of patients.

Aim

The aim of this paper was to test the feasibility of retrieving diagnosis data from a long period of time and to elucidate the patterns of clinical categories of patients over time by applying the Adjusted Clinical Groups® case-mix system to encounter data from primary health care units in one limited geographic area.

Methods

Longitudinal data from 17 years wereretrieved from the primary health care centre in Gagnef in Dalarna county council in Sweden. The Swedish primary health care version of the diagnosis codes were mapped to the full version of the ICD-10 classification. The 7.1 version of the Adjusted Clinical Groups software was utilised.

Results

It was feasible to elucidate the longitudinal pattern of morbidity during all 17 years. The categories of patients according to the case-mix system showed great stability over time. On the diagnosis level some distinct shifts over timewere found.

Conclusion

Diagnoses fromelectronic patient records in primary health care during seventeen years could be used to elucidate the morbidity pattern among people in a defined geographic area. From a patient perspective thecase-mix system used showed great stability over time in terms of categories of patients with similar morbidity pattern. From the diagnosis perspective a slow shift in some types of diagnoses could be seen. The need for rules of coding and registering diagnoses in primary health care has to be emphasized.