Children as frequent attenders in primary care: a systematic review

Using machine learning to identify frequent attendance at accident and emergency services in Lanarkshire

Background

Frequent attenders to accident and emergency (A&E) services pose complex challenges for healthcare providers, often driven by critical clinical needs. Machine learning (ML) offers potential for predictive approaches to managing frequent attendance, yet its application in this area is limited. Existing studies often focus on specific populations or models, raising concerns about generalisability. Identifying risk factors for frequent attendance and high resource use is crucial for effective prevention strategies.

Objectives

This research aims to evaluate the strengths and weaknesses of ML approaches in predicting frequent A&E attendance in NHS Lanarkshire, Scotland, identify associated risk factors and compare findings with existing research to uncover commonalities and differences.

Method

Health and social care data were collected from 17,437 A&E patients in NHS Lanarkshire (2021-2022), including clinical, social and demographic information. Five classification models were tested: multinomial logistic regression (LR), random forests (RF), support vector machine (SVM) classifier, k-nearest neighbours (k-NN) and multi-layer perceptron (MLP) classifier. Models were evaluated using a confusion matrix and metrics such as precision, recall, F1 and area under the curve. Shapley values were used to identify risk factors.

Results

MLP achieved the highest F1 score (0.75), followed by k-NN, RF and SVM (0.72 each), and LR (0.70). Key health conditions and risk factors consistently predicted frequent attendance across models, with some variation highlighting dataset-specific characteristics.

Conclusions

This study underscores the utility of combining ML models to enhance prediction accuracy and identify risk factors. Findings align with existing research but reveal unique insights specific to the dataset and methodology.

Foot and ankle problems in children and young people: a population-based cohort study

The aim of this research was to describe the epidemiology, presentation and healthcare use in primary care for foot and ankle problems in children and young people (CYP) across England. We undertook a population-based cohort study using data from the Clinical Practice Research Datalink Aurum database, a database of anonymised electronic health records from general practices across England. Data was accessed for all CYP aged 0-18 years presenting to their general practitioner between January 2015 and December 2021 with a foot and/or ankle problem. Consultation rates were calculated and used to estimate numbers of consultations in an average practice. Hierarchical Poisson regression estimated relative rates of consultations across sociodemographic groups and logistic regression evaluated factors associated with repeat consultations. A total of 416,137 patients had 687,753 foot and ankle events, of which the majority were categorised as "musculoskeletal" (34%) and "unspecified pain" (21%). Rates peaked at 601 consultations per 10,000 patient-years among males aged 10-14 years in 2018. An average practice might observe 132 (95% CI 110 to 155) consultations annually. Odds for repeat consultations were higher among those with pre-existing diagnoses including juvenile arthritis (OR 1.73, 95% CI 1.48 to 2.03).    Conclusions: Consultations for foot and ankle problems were high among CYP, particularly males aged 10 to 14 years. These data can inform service provision to ensure CYP access appropriate health professionals for accurate diagnosis and treatment. What is Known: • Foot and ankle problems can have considerable impact on health-related quality of life in children and young people (CYP). • There is limited data describing the nature and frequency of foot and ankle problems in CYP. What is New: • Foot and ankle consultations were higher in English general practice among CYP aged 10 to 14 years compared to other age groups, and higher among males compared to females. • The high proportion of unspecified diagnoses and repeat consultations suggests there is need for greater integration between general practice and allied health professionals in community-based healthcare settings.

Consultation patterns and frequent attenders in UK primary care from 2000 to 2019: a retrospective cohort analysis of consultation events across 845 general practices

OBJECTIVE

To describe the distribution of consultations at the practice level and examine whether increases are uniform or driven by people who consult more frequently.

DESIGN

Retrospective cohort study.

SETTING

UK general practice data from the Clinical Practice Research Datalink (CPRD) GOLD database.

PARTICIPANTS

1 699 709 314 consultation events from 12 330 545 patients, in 845 general practices (1 April 2000 to 31 March 2019).

METHODS

Consultation information was aggregated by financial year into: all consultations/all staff; all consultations/general practitioners (GPs); face-to-face consultations/all staff; face-to-face consultations/GPs. Patients with a number of consultations above the 90th centile, within each year, were classified as frequent attenders. Negative binomial regressions examined the association between available practice characteristics and consultation distribution.

RESULTS

Among frequent attenders, all consultations by GPs increased from a median (25th and 75th centile) of 13 (10 and 16) to 21 (18 and 25) and all consultations by all staff increased from 27 (23-30) to 60 (51-69) over the study period. Approximately four out of ten consultations of any type concerned frequent attenders and the proportion of consultations attributed to them increased over time, particularly for face-to-face consultations with GPs, from a median of 38.0% (35.9%-40.3%) in 2000-2001 to 43.0% (40.6%-46.4%) in 2018-2019. Regression analyses indicated decreasing trends over time for face-to-face consultations and increasing trends for all consultation types, for both GPs and all staff. Frequent attenders consulted approximately five times more than the rest of the practice population, on average, with adjusted incidence rate ratios ranging between 4.992 (95% CI 4.917 to 5.068) for face-to-face consultations with all staff and 5.603 (95% CI 5.560 to 5.647) for all consultations with GPs.

CONCLUSIONS

Frequent attenders progressively contributed to increased workload in general practices across the UK from 2000 to 2019. Important knowledge gaps remain in terms of the demographic, social and health characteristics of frequent attenders and how UK general practices can be prepared to meet the needs of these patients.

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

OBJECTIVES

With 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.

DESIGN

A retrospective cohort study.

SETTING

Routinely collected primary and secondary healthcare data in Northwest London from the Discover database.

PARTICIPANTS

378 309 CYP aged 0-15 years registered to a general practice in Northwest London with 1 full year of follow-up.

PRIMARY AND SECONDARY OUTCOME MEASURES

Assignment 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.

RESULTS

Participants 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.

CONCLUSIONS

This 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.