Among the 7,612,087 patients under 18 years in CPRD Aurum, there were 416,137 patients with 687,753 coded events for foot and ankle problems from 1st of January 2015 to 31st of December 2021 from 1448 practices (see online resource 1). Descriptive characteristics for the cohort and their total number of coded events are presented in Table 1. The mean age of the study population was 10.7 years (standard deviation, 4.6), and the age category with the highest frequency of first (42%) and total (44%) primary care events was 10 to 14 years. There were more males (52%) than females (48%) in the cohort and across all the age categories apart from category 5 to 9 years where 70,090 (53%) were females compared to 62,087 (47%) males. Most participants (67%) had only one coded event during the study period. The most common category was “musculoskeletal” (34%), followed by “unspecified pain” (22%), “dermatological” (21%) and “infection” (11%). Participants were mostly in the white ethnic group (77%), followed by Asian (7%). Based on the indices of multiple deprivation, slightly more practices were categorised in the least deprived regions (24%). The most frequently recorded pre-existing diagnoses were autism (4%) and ADHD (4%), although those with intellectual disability had the highest proportion of total coded events (5%). Most participants did not have a BMI (z) recording in the years pre and post their index date or were below the age of 3 during this recording (90%). Whilst the BMI (z) was unknown for most of the total coded events (42%), there were more BMI (z) records overall, indicating that over a third of consultations were with patients with a normal BMI (z), 13% overweight and 11% obese.

The frequency of the ten most recorded foot and ankle consultation codes are shown in Table 2—“ingrowing great toenail” was observed 110,624, representing 16% of total codes, followed by “foot pain” (10%) and “paronychia of toe” (7%). The frequency of the ten most recorded foot and ankle events for males and females is reported (see online resource 2).

Rates of foot and ankle health consultations peaked at 601 consultations per 10,000 patient-years among males aged 10 to 14 years in 2018 and 641 consultations per 10,000 patient-years among females aged 10 to 14 years in 2015 (Fig. 1). The average rate across the study period was 343 (SD = 178) per 10,000 patient-years overall and 352 (SD = 179) and 333 (SD = 179) for males and females, respectively.

Rate per 10,000 patient-years of foot and ankle consultations in CPRD (2015 to 2020)

Table 3 shows incident rate ratios (95% confidence intervals) for foot and ankle consultations, unadjusted and adjusted for age group, gender, year and region. Being female was associated with lower consultation rates for foot and ankle health than being male (adjusted rate ratio (ARR) 0.96; 95% CI 0.95 to 0.96). Most regions were associated with a higher rate of consultations compared to the South East, apart from London, which was associated with a lower rate (ARR 0.74; 95% CI 0.68 to 0.81). Years 2020 and 2021 were associated with lower rates compared to the 2015 reference year: 2020 (ARR 0.62; 95% CI 0.61 to 0.63) and 2021 (ARR 0.71; 0.70 to 0.72).

There were 83,197 (21%) out of 398,952 with repeat consultations for foot and ankle problems within 6 months (Fig. 2). Those in black, Asian and other ethnic groups had lower odds of repeat consultations compared to those in the white group, as did females compared to males (odds ratio 0.95, 95% confidence interval 0.93 to 0.96) (Fig. 2). There were increased odds of repeat consultations for CYP with pre-existing diagnoses: autism (1.12, 1.08 to 1.17), diabetes (1.21, 1.14 to 1.28), intellectual disabilities (1.13, 1.07 to 1.20) and juvenile arthritis (1.73, 1.48 to 2.03). The DAG is reported (online resources 3) and depicts assumed relationships between the exposure and outcome and all variables included in the fully adjusted analysis model. Factors associated with repeat consultations according to subgroups of code categories of the index consultation are reported (see online resources 4–7).

Logistic regression model of variables associated with the outcome of repeat consultations for all foot and ankle health encounters within six months during the study period

The average general practice in our sample had approximately 10,000 patients and 3500 patients aged 18 or younger (Table 4). In 1 year, such a general practice could expect to see 103 CYP (95% CI 83 to 122) with a first consultation for foot and ankle health, 41 patients (95% CI 28 to 53) in the musculoskeletal category, 21 dermatological (95% CI 12 to 30) and 25 unspecified pain (95% CI 15 to 35).