Paediatric contacts with the UK out-of-hours primary care service and contact outcomes: a regional service evaluation

Background Demand on hospital emergency departments for paediatric problems is increasing. However, the volume and nature of paediatric health demands placed on other parts of the urgent care system have not been explored. This understanding is an important first step in developing and improving out-of-hospital care. We aimed to describe the volume, nature, and outcomes of paediatric contacts with out-of-hours general practice (OOH GP). We performed a retrospective service evaluation using data from 12 months of paediatric patient contacts with the Oxfordshire OOH GP service. Methods A database of contacts with the Oxfordshire OOH GP service was created for a 12 month period from December 2014 to November 2015. Descriptive statistics were calculated using SPSS Version 25. Results 27,455 contacts were made by 18,987 individuals during a 12 month period. The majority of these were for children aged under 5. Over 70% of contacts were at the weekend. The peak contact period was between 18:30 and 21:30. Over 40% of contacts resulted in advice only (no onward referral, requirement for GP follow up, or prescription). 19.7% of contacts resulted in an antibiotic prescription, most commonly those linked with ear, chest, and throat infections. Discussion Paediatric contacts with the Oxfordshire OOH GP service were predominantly in younger age groups and in the evening, with 19.7% resulting in an antibiotic prescription. Almost half of the contacts had no follow up or prescription, suggesting non-prescribing health care professionals could be involved in providing care in OOH GP. Further research should consider how children and their parents can be best supported to optimise OOH consulting.


Background
Demand on hospital emergency departments for paediatric problems is increasing [1] , [2] , [3]. At one centre, attendances for medical problems in the Emergency Department (ED) rose 42% in the 10 years between 1997 and 8 and 2007-8 and the 0-4 years age group account for almost 70% of these contacts [3]. However, amongst children aged under 16, almost 30% of these contacts may be better managed in primary and out-of-hospital settings [4] , [5] , [6], including out-of-hours general practice (OOH GP).
In the UK, the provision of primary care services outside core contracted hours is an integral NHS service [7] , [8]. In 2013-2014, 5.8 million cases were handled by OOH GP in England resulting in 3.3 million face-to-face consultations, including 800,000 home visits [9]. This service provides access to urgent primary care between 18:30 and 08:00 on weekdays, and all day on weekends and on bank holidays. It is an important alternative to a visit to the emergency department. Appointments with the OOH GP service are booked vie the free-to-use NHS 111 telephone advice line, where trained call handlers use the NHS pathways algorithm to direct patients to self-care advice or the most appropriate service for their needs.
Understanding the nature and outcomes of contacts with NHS services is key to evaluating and improving these services. Paediatric contacts with NHS services have been described in relation to Emergency Departments (ED) or Urgent Care Centres (UCC) [10][11][12] . In OOH primary care services in eight European countries, up to a third of contacts were shown to be for children under the age of 18 [13] and a third of out of hours consultations amongst under 12 s were fever related in a Dutch OOH GP service [14]. Of these, 7.6% were referred to secondary care [14]. In Denmark 27% of children aged 0-5 presenting to OOH GP received a prescription, 74% of which were for antibiotics, and 7.4% of children were referred to a nearby hospital [15].
To date, there has been no description of paediatric contacts, or the outcome of these contacts, with OOH GP in the UK. This understanding is critical to inform new approaches to managing the rising demand for acute paediatric assessment across urgent and unscheduled care, and ensuring safe and appropriate management. We aimed to characterise the nature, timing, and outcomes of paediatric contacts with the OOH GP using a large dataset of patient contacts with the Oxfordshire OOH service, which provides care to a population of over 600,000 people.

Methods
A database of all intended patient contacts with the Oxfordshire OOH GP service, which provides care to a population of over 600,000 people, over 1 year from 1 December 2014 to 30 November 2015 was created from the OOH Electronic Record System used by clinicians (Adastra).
Service data included contact type, contact outcome, date, clinical codes assigned and prescriptions issued. Demographic data included patient sex, age, and a deprivation index (the Index of Multiple Deprivation (IMD)). Lower IMD scores indicate lower levels of deprivation. Contacts were coded with their final contact type: a 'telephone consultation', a 'base visit' (patient assessed at OOH base), a 'home visit' (patient assessed in their home), or a "111 appointment to book" for patients the 111 service have determined an appointment is needed but cannot be booked (for example due to appointments being full, or the triaging service lacking access to appointment system). This results in a contact with the OOH GP which may be a base visit, telephone consultation, or home visit.
Timings of calls were classified as evening 18:30-23: 59 h, night 00:00-07:59 h, and daytime (on weekends and bank holidays) 08:00-18:29 h. The weekend period was classified as 18:30 h Friday until 08:00 h Monday, whilst contacts classified as 'Bank Holiday' were those occurring during daytime hours on each Bank Holiday.
At the end of each consultation, clinicians assign at least one clinical code (for example 'H05z. Upper respiratory infection. NOS'), which were used in this study to determine clinical presentation for that contact. These clinical codes were validated by two members of the direct care team based on previous coding validity studies [16]. The positive predictive value (PPV) of the clinical code for medical diagnosis or conclusion was 92.6% [8].
In order to facilitate analysis these codes translated into International Classification of Primary Care (ICPC) codes by two clinically qualified authors (GH, RB) (supplementary Table 1). Codes which may indicate an infection were noted. An additional code, not available in the ICPC, was added for contacts indicating the patients did not attend (DNA). A small number of contacts (n = 264, 1%) had "NULL" as their clinical code. These contacts were removed from the dataset.
Clinicians also detail any prescriptions issued during the consultation in up to six fields. Prescriptions were grouped into 21 categories and 79 sub-categories for the purposes of analysis. These are detailed in supplementary Table 2.
Each contact is coded with an 'Outcome'. There were 19 outcome codes, which were condensed into 9 outcome categories. These are detailed in supplementary Table 3.
Descriptive statistics were calculated in SPSS version 25. Analyses were undertaken at the contact level unless otherwise stated.
The median age of children contacting the OOH GP service was 3.1 (IQR 1.1-7.0). In the twelve month period, 6136 contacts (22.3%) were for children under the age of one, and 11,852 (43.2%) of contacts were for children aged between 1 and 4 years. Children aged 5-11 account for 6118 contacts (22.3%), whilst there were 3349 contacts (12.2%) for children aged 12 and over. The median index of multiple deprivation (IMD) score was 10.79 (IQR 6.15-18.24).
48.7% of contacts were for female patients, and of these the median age in years was 3.5 (IQR 1.2-8.1). The median age of male patients was 2.8 (IQR 1.1-6.0). Figure 2 shows the age distribution, in whole years, of all paediatric contacts with the Oxfordshire OOH GP service during the 12 month period. At younger ages, there were slightly more male patients (54.5 and 54.0% in the 0-1 and 1-5 age groups respectively). In contrast, in the 12-17 age group almost twice as many female patients contacted the service (63.5% of contacts in this age group).

Contact types and timings
The month with the greatest number of contacts was December, and July had the lowest number of contacts (Table 1). 19,260 (70.2%) contacts were at the weekend and 870 (3.2%) were on a Bank Holiday. 9103 (33.2%) contacts were on a Saturday and 8048 (29.3%) were made on a Sunday. During the week, Tuesdays and Wednesdays had the lowest proportion of contacts (6.4% in both cases), followed by Thursday (7.0%), Monday (8.8%) and Friday (9.0%).
There were 20,692 (75.4%) base visits whilst 6603 (24.1%) contacts were managed purely by telephone. There were 123 (0.4%) home visits and 37 (0.1%) "111 appt to book contacts" (contacts for which appointments cannot be booked within the time frame dictated by 111 and are passed onto the OOH GP service separately). Figure 3 demonstrates the pattern of attendance during weekdays (top) and at the weekend or on Bank Holidays (bottom) divided by type of contact (excluding "111 appt to book"). Peak attendance was between 18:00 and 21:00. On non-bank holiday weekdays, 5089 contacts (73.4%) were in the evening and 1749 (25.2%) contacts  were in the night. In contrast at the weekend, 10779 (56.0%) contacts were made during the day, 2021 (10.5%) were during the night and 6457 (33.5%) were during the evening.
Reasons for contact Table 2 shows the frequency of contacts including at least one clinical code falling into each ICPC Chapter. The most common clinical category overall and in each age-category was 'Respiratory' (Chapter R), with related codes in 9280 (33.8%) contacts. Other commonly arising chapters 'General and Unspecified' (Chapter A) (n = 5627, 20.5%), 'Digestive' (Chapter D) (4181, 15.2%), 'Skin' (Chapter S) (2635, 9.6%), and 'Ear' (Chapter H) (2209, 8.0%). 472 (1.7%) of contacts were coded with a musculoskeletal code and this was most prevalent in children aged 12 and older. 82 (2.4%) contacts for children aged 12 and over were coded with a psychiatric condition. Ninety-four of the 101 of children aged less than 1 with a psychological code had problems feeding.

Outcomes of contacts
After 16,153 (58.8%) contacts there was no documented requirement for follow up. After 11,359 (41.4%) contacts there was no documented requirement for follow up and no prescription issued, considered as 'advice only' appointments. In 8394 (30.6%) of contacts the patient was advised to contact their own GP, and there was a referral to secondary care in 1754 (6.4%) of contacts. Table 3 shows the proportion of each contact outcome by age group.

Summary of findings
Over a quarter (26.69%) of contacts with the Oxfordshire OOH GP service between December 2014 and November 2015 were for children under the age of 18. Of these, the majority (65.5%) were for children aged under 5 years. There were more male than female patients amongst the under 5 s, but amongst teenagers almost two thirds of the contacts were for female patients.
In over 40% of contacts there was no onward referral, requirement for GP follow up, or prescriptions, indicating that the primary clinical output of the contact was advice or reassurance. The proportion of these contacts was highest amongst the youngest age group (47.9%) and fell in each subsequent age group. Antibiotics were prescribed in 19.7% of contacts, most commonly linked with ear, respiratory, and skin conditions. During the week, the majority of contacts were in the evening, with a peak between 18:30 and 21:30. At the weekend, there was an early peak of contacts at 09:00 and a second, higher peak between the hours of 19:00 and 21:00.
Children presented most commonly with respiratory conditions. The most common ICPC code was an upper respiratory tract infection. Other common codes were unspecified viral diseases, tonsillitis and acute otitis media. There were few contacts with musculoskeletal or psychological complaints.

Comparison with the literature
In the UK research into paediatric use of NHS services has focused on unscheduled urgent care in Urgent Care

Centres (UCCs) and the Emergency Department (ED).
Use of health services is highest amongst the youngest children [10,17]. In 2011-12, one in three children aged under five visited the ED compared with 17% of children aged 5-14 [10]. This use peaks at 18:00 on weekdays, although not at weekends [10]. Amongst children under 5 the most common complaints to a UCC are respiratory system, infectious, or parasitic diseases [11]. In contrast, older children, especially those aged over 15 [11], use primary care services less, but present more commonly to UCCs with musculoskeletal complaints [12]. Our findings suggest that the profile of paediatric demand on OOH GP is more similar to that of the ED than the UCC. In Europe the proportion of paediatric contacts with OOH GP ranged from 1.9% in Switzerland to 33.3% in Denmark [13]. In Belgium, Denmark, the Netherlands, and Norway, contacts were highest for those in the 0-4 age group [13]. This work found that musculoskeletal and respiratory problems were very common amongst patients aged 0-17, as were 'general and unspecified' and 'ear related'. Our study also found that respiratory and ear conditions were common but musculoskeletal conditions were less common. This may be due to the co-location of over half of the Oxfordshire bases with a minor injuries unit which would typically assess musculoskeletal problems. Similarly, urgent patient problems related to psychological illnesses may be handled by parallel out-of-hours services. These differences may be interesting to explore further in the context of the remit of other OOH GP available in these countries.
There is limited research into the outcomes of paediatric consultations with OOH GP. One study of consultations for fever amongst children consulting with an OOH GP service in the Netherlands found that 92% of contacts were managed without referral to secondary care [14]. In our study the equivalent figure was 93.6%.
Antibiotics were prescribed in 19.7% of contacts in this study, primarily for respiratory, ear, and skin conditions. Antibiotic prescribing has previously been associated with 15% of all contacts with the UK OOH GP service, and 18.0% of contacts for individuals under the age of  18 [18]. Although our data were from the same population, it was from a later time period; reasons for a higher antibiotic prescription rate are unclear. Qualitative work has revealed that GPs have a lower threshold for prescribing OOH [19]. They raised concerns including safety, a lack of background knowledge about patients in an OOH GP setting and a lack of availability of diagnostics [19].

Strength and limitations
To our knowledge this is the first study to explore in detail paediatric demand on a UK OOH GP service. It uses a large dataset of 27,455 contacts. However, as the dataset is limited to Oxfordshire, our results may not be generalisable to the UK population as a whole. Oxfordshire is, on the whole, less deprived than the UK, and our methods should be replicated in other regions with different populations and care models. These data also precede recent extended access schemes which may alter how people navigate the health services.

Implications for research and practice
We found that 41.4% of contacts were 'advice only'. Whilst these contacts may have been appropriate, this highlights the need for research exploring the content and main function of these consultations, and reviewing whether alternative sources of advice could offer a safe substitute. This also suggests that nurse practitioners  and paramedics who do not prescribe could be part of the team offering paediatric assessment with appropriate training. Antibiotics were prescribed in 19.7% of contacts. UK OOH GP is associated with a disproportionately high rate of antibiotic prescribing [20], representing 5% of prescriptions from primary care despite only 1% of GP consultations occurring out-of-hours. In the Netherlands the appropriateness of antibiotic prescribing in OOH GP is comparable to in-hours care [21], but the extent to which this is true in the UK is unclear and could be a focus of future research. Point of care CRP testing has been shown to be feasible in a UK OOH primary care service [22], however, its role in reducing prescribing for acute respiratory illness in paediatrics remains unclear in this setting [23].
Finally, our study took a clinician perspective on contacts with OOH GP linking the reason for contact with the clinical codes applied during the consultation. Qualitative research exploring the process by which parents make decisions to consult OOH GP specifically will deepen our understanding of these consultations and how patients can be best supported. This is of specific interest for general and respiratory illnesses amongst younger children.

Conclusions
We found that paediatric contacts with the Oxfordshire OOH GP service were predominantly in younger age groups and in the early evening with 19.7% resulting in an antibiotic prescription. Almost half of the contacts were advice-only appointments, and further research should consider how patients making these appointments can be best supported.