Statistical complexity of reasons for encounter in high users of out of hours primary care: analysis of a national service

Abstract

BackgroundManaging demand for urgent and unscheduled care is a major problem for health services globally. A particular issue is that some patients appear to make heavy use of services, including primary care out of hours. We hypothesised that greater variation (statistical complexity) in reasons for attending primary care out of hours services may be a useful marker of patients at high risk of ongoing heavy service use.MethodsWe analysed an anonymised dataset of contacts with the primary care out of hours care for Scotland in 2011. This contained 120,395 contacts from 13,981 high-using patients who made 5 or more contacts during a calendar year. We allocated the stated reason for each encounter into one of 14 categories. For each patient we calculated measures of statistical complexity of reasons for encounter including the count of different categories, Herfindahl index and statistical entropy of either the categories themselves, or the category transitions. We examined the association of these measures of statistical complexity with patient and healthcare use characteristics.ResultsThe high users comprised 2.4% of adults using the service and accounted for 15% of all contacts. Statistical complexity (as entropy of categories) increased with number of contacts but was not substantially influenced by either patient age or sex. This lack of association with age was unexpected as with increasing multi-morbidity one would expect greater variability in reason for encounter. Between 5 and 10 consultations, higher entropy was associated with a reduced likelihood of further consultations. In contrast, the occurrence of one or more contacts for a mental health problem was associated with increased likelihood of further consultations.ConclusionComplexity of reason for encounter can be estimated in an out of hours primary care setting. Similar levels of statistical complexity are seen in younger and older adults (suggesting that it is more to do with consultation behaviour than morbidity) but it is not a predictor of ongoing high use of urgent care.Electronic supplementary materialThe online version of this article (10.1186/s12913-019-3938-z) contains supplementary material, which is available to authorized users.

Death from cancer: frequent unscheduled care

“…There is no universally agreed definition of 'frequent' attenders; definitions range from 3 to 10 attendances per person per year. [6][7][8][9] We defined infrequent attenders as those with one to four attendances/year; frequent attenders had five to nine attendances/year and very frequent attenders had ≥10 attendances/year.…”

Heterogeneity of reasons for attendance in frequent attenders of emergency departments and its relationship to future attendance

“…There is no standard definition for high use of emergency departments. While several recent studies have used a cut-off of four or more attendances 3 , we chose five as this was the threshold used in a recent study of GP Out of Hours service attendance which used a similar approach to analysis 17 . The data contained some instances where patients had multiple encounters on the same day.…”

“…Shannon entropy extends this by describing the unpredictability of the distribution of categories (if most consultations are in one category it is quite likely the next one will be in that category too; if consultations are spread across several categories, it is harder to predict what the next one will be).. These measures which are widely used for instance in ecology have previously been used in a study of Primary Care Out of Hours services 17 . Figure 1 illustrates the relationship between these, using the example of patients with 8 ED attendances for up to 4 categories of presenting complaint (indicated by the letters A,B,C&D).…”

“…We used count and entropy of presenting complaint categories (split at the median into high and low) and attendance with a mental health problem (present or absent) over the first N contacts as predictors. We calculated the odds ratio adjusted for age and sex for each value of N 17 .…”

“…Our choice of 13 categories mirrored that in a study of a GP Out of Hours service. 17 The use of presenting complaint from ED data which was not linked to other medical records meant we were unable to include any measure of actual multimorbidity. This means we could not disentangle the effects on attendance of concern about multiple symptoms from the presence of multiple diseases.…”