Patterns of healthcare utilisation in children and young people: a retrospective cohort study using routinely collected healthcare data in Northwest London

Abstract

ObjectivesWith a growing role for health services in managing population health, there is a need for early identification of populations with high need. Segmentation approaches partition the population based on demographics, long-term conditions (LTCs) or healthcare utilisation but have mostly been applied to adults. Our study uses segmentation methods to distinguish patterns of healthcare utilisation in children and young people (CYP) and to explore predictors of segment membership.DesignA retrospective cohort study.SettingRoutinely collected primary and secondary healthcare data in Northwest London from the Discover database.Participants378 309 CYP aged 0–15 years registered to a general practice in Northwest London with 1 full year of follow-up.Primary and secondary outcome measuresAssignment of each participant to a segment defined by seven healthcare variables representing primary and secondary care attendances, and description of utilisation patterns by segment. Predictors of segment membership described by age, sex, ethnicity, deprivation and LTCs.ResultsParticipants were grouped into six segments based on healthcare utilisation. Three segments predominantly used primary care, two moderate utilisation segments differed in use of emergency or elective care, and a high utilisation segment, representing 16 632 (4.4%) children accounted for the highest mean presentations across all service types. The two smallest segments, representing 13.3% of the population, accounted for 62.5% of total costs. Younger age, residence in areas of higher deprivation and the presence of one or more LTCs were associated with membership of higher utilisation segments, but 75.0% of those in the highest utilisation segment had no LTC.ConclusionsThis article identifies six segments of healthcare utilisation in CYP and predictors of segment membership. Demographics and LTCs may not explain utilisation patterns as strongly as in adults, which may limit the use of routine data in predicting utilisation and suggest children have less well-defined trajectories of service use than adults.

Social deprivation and the use of healthcare services over one year by children less than 18 years of age in 2018: A French nationwide observational study

“…We found that approximately 24% of children < 18 years of age had at least one ED visit, which is highly congruent with other estimations for 0- to 15-year-old children in France (more than 25%) [ 41 ] and in in Northwest London (24%) [ 42 ]. However, there are large differences between countries.…”

Comparing natural language processing representations of disease sequences for prediction in the electronic healthcare record

“…16 Second, the recurrence of codes in the medical record is dependent on factors including the GP practice, patient demographics, and disease-related factors, such as whether the disease is part of a quality assessment metric for practices. 12 Previous research has highlighted that those in some ethnic groups, and living in areas of higher socioeconomic deprivation have a greater frequency of recurrent codes, which may result in models having more data, and therefore learning sequences better for some patients than others. 12 In our study, we focussed on the performance across the population, but further work could explore the relative differences stratified by patient demographics, such as ethnicity.…”

“…EHR-BERT adapts and extends the BEHRT model to include additional sociodemographic factors known to be associated with code frequency in the EHR. 12 We added embedding layers for gender, ethnicity, socioeconomic deprivation decile and calendar year of the observation (Table 2 and Supplementary Information Figure S4). Compared to BEHRT, we removed the segment token, and the [SEP] tokens between visit, as most visits contained only one code, and we believed that visit number should suffice to capture sequential visit information.…”

“…Previous work has highlighted the potential biases in code frequency in primary care data, which may relate to patient demographics, organisational policies, and coding incentives, rather than being an objective marker of a person's health status. 12 However, the better performance of BERT models, which also utilise recurrent codes, suggests that an ability of the attention mechanisms in these models to regulate the impact of repeated codes.…”

“…12 Previous research has highlighted that those in some ethnic groups, and living in areas of higher socioeconomic deprivation have a greater frequency of recurrent codes, which may result in models having more data, and therefore learning sequences better for some patients than others. 12 In our study, we focussed on the performance across the population, but further work could explore the relative differences stratified by patient demographics, such as ethnicity.…”