Article Text

Download PDFPDF

Patterns of healthcare use among adolescents attending an urban general practitioner-led urgent care centre
  1. Shamini Gnani1,
  2. Helen McDonald1,
  3. Saiful Islam1,
  4. Farzan Ramzan1,
  5. Michele Davison2,
  6. Tim Ladbrooke3,
  7. Azeem Majeed1,
  8. Sonia Saxena1
  1. 1Department of Primary Care and Public Health, Imperial College London, London, UK
  2. 2North End Medical Centre, London, UK
  3. 3London Central and West Unscheduled Care Collaborative, London, UK
  1. Correspondence to Dr Shamini Gnani, Department of Primary Care and Public Health, Imperial College London, 3rd Floor Reynolds Building, Charing Cross Campus, St Dunstan's Road, London W6 8RP, UK; s.gnani{at}


Introduction Adolescence is a time of increasing health and peak fitness, as well as increasing health risks. In the UK, primary care is free at the point of access, yet, adolescents aged 10–19 years are the lowest users of primary care services, and disproportionately high users of emergency services. The effect of new general practitioner (GP)-led urgent care centres in meeting the needs of adolescents are unknown.

Methods We used routinely collected data to describe the demographics and attendance pattern among adolescents at two new colocated GP-led urgent care centres at Hammersmith and Charing Cross Hospitals, London. We also compared attendance rates with those observed in routine general practice and emergency departments.

Results Adolescents formed 6.5% (N=14 038) of total urgent care attendances. 13.2% (95% CI 12.9% to 14.1%) was recorded as not being registered with a GP. Commonest reasons for attendance were musculoskeletal conditions and injuries (30.2%), respiratory tract infections (12.5%) and limb fractures (5.1%). Adolescents aged 15–19 years were more likely to attend the centres (30.6 vs 23.4, per 100, p<0.0001) than routine general practice. The opposite was true for adolescents aged 10–14 years.

Conclusions Adolescents aged 15–19 years are more likely to attend urgent care centres than general practice. The majority attended for conditions commonly seen in primary care including musculoskeletal conditions and injuries, and respiratory tract infections. Primary care services may need to be more responsive to needs of the older adolescent age, if their use of urgent care centres is to be reduced.

  • primary care
  • emergency departments
  • musculo-skeletal
  • emergency care systems, primary care

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Adolescence is a time of increasing health and peak fitness and also of increasing health risks.1 More than 40% of deaths among this age group are due to intentional and unintentional injuries.2 In the UK, primary care is free at the point of access,3 yet, adolescents aged 10–19 years are the lowest users of primary care services; (5.8% of primary care consultations) and disproportionately high users of emergency services; and account for 13.9% of total emergency department attendances.4 ,5

Comparisons of primary and emergency models of care are limited because they offer different services. Differences in patterns of service use between primary care and emergency departments may reflect different health needs among adolescents. For example, among 16–18-year-olds, road traffic accidents account for over 50% of mortality, and young people aged 15–24 years are at greatest risk of being victims of violent crime.2 Further, adolescents may prefer the immediate access available through emergency departments when compared with making and keeping appointments with UK general practitioners (GP).

In 2007, the National Health Service (NHS) in England set out a new model of GP-led health centres to reform access to urgent care.6 These urgent care centres were to provide walk-in access and have extended consulting times outside normal primary care working hours. NHS Hammersmith and Fulham commissioned an integrated urgent care model to establish a GP-led urgent care centre colocated with Charing Cross Hospital and Hammersmith Hospital emergency departments. The aims of the centres were to reduce accident and emergency attendances, short-stay hospital admissions, and the use of diagnostic tests. The centres also included a newly established general practice with a branch on each site.

The impact of the GP urgent care model of care is unknown. It may favour adolescents to be either high or low users of care when compared with other models of care. An urgent care setting may better meet the needs of adolescents by addressing responsiveness in access to care at times which may not be met by routine general practice. Our study aims to describe the demographics and attendance pattern among adolescents at two new colocated GP-led urgent care centres at Hammersmith and Charing Cross Hospitals, and compare attendance rates with those observed in routine general practice and emergency departments.


The Borough of Hammersmith and Fulham is located in London. The population is around 176 000, and is comprised of 60% white British, 20% other white, and 20% non-white ethnic groups.7 In common with other inner London boroughs, the population is younger, on average, than that of England, with a high proportion of young adults. It is designated a spearhead area, therefore, in the top fifth nationally for socioeconomic deprivation.

Hammersmith Urgent Care Centre opened on 23 April 2009, and Fulham Urgent Care Centre on 7 September 2009. Data collection is on-going as part of the service. There are continuous checks and reviews for accuracy and completeness. No ethical approval was sought for this study, in line with National Research Ethics Service Guidance, as routinely available data was used and the project was considered a service evaluation.8

Model of care at Hammersmith and Fulham Urgent Care Centres

The centres offer a new model of ambulatory healthcare which comprises two types of services: open access urgent care services with no appointment necessary, and normal GP services with planned appointments for local residents registered with the service. Normal core GP services are from 8:00 to 18:30 Monday to Friday, however the centres offer GP services 7 days a week from 8:00 to 20:00. The urgent care facility is open 7 days a week and staffed by GPs and emergency nurse practitioners. The Fulham centre is open 24 h a day, while the Hammersmith centre is open between the hours of 8:00 and 22:00. Patients who attend either of the centres first register with reception. They are then seen by a GP streamer who makes a decision as to what stream the patients should be seen in.9 This streaming or triaging decision should normally take several minutes. The primary aim of streaming is for the GP to determine whether the patient needs immediate referral to the emergency department, or if they can be managed within the primary care setting of the centres.

Hence, a treatment stream is allocated to each patient after initial GP assessment. There are a number of treatment streams. The emergency stream is one where the patient requires emergency department facilities. The GP priority stream is where assessment requires a GP and, on average, 20 min is allocated for this assessment to take place. The minor illness stream is where patients may be seen by either a GP or an emergency nurse practitioner for conditions, such as urinary or upper respiratory tract infections. There is a minor injury stream, where sprains and soft tissue injuries are dealt with. The see-and-treat stream which especially operates at the Hammersmith site in the evening is where, at the point of streaming, the GP streamer also treats and discharges the patient. The reception navigation stream is where patients attend the urgent care centre, but would benefit from being seen by another specialist service, such as sexual health services, or are helped by reception to get an appointment with their own GP (see appendix 1).

Data sources

Urgent care centre patients’ data are entered in the Adastra database, which is used mainly by ‘out of hours’ GP providers.10 Data collected include demographics (date of birth, postcode and ethnicity); clinical information (diagnosis, treatment and ‘stream’ allocation); and outcome (discharge or referral). Clinical information is recorded using READ codes for the main clinical diagnosis.

Routinely collected data were analysed for attendances at the Hammersmith and Fulham Urgent Care Centres during a 30-month period from 1 October 2009 to 31 March 2012. We extracted demographic data including age, sex, ethnicity, attendance records and clinical information on the presenting complaint and treatment on all patients aged 10–19 years (adolescents).

Routine data were not included if the adolescent attended accompanied by a parent or a guardian. Attendances were considered as ‘out of hours’ unless the arrival time was within 8:00 and 18:30, Monday to Friday. Data excluded all patients brought in by ambulance, as these patients bypass the urgent care centres, attending the emergency department directly.

Statistical analysis

We examined the baseline characteristics of adolescent attendances at the urgent care centres by age and by type of clinical stream ‘in’ and ‘out of’ routine general practice hours, and the top 10 reasons for attendance based on clinical READ codes. We also compared the attendance rates between urgent care centres and routine general practice and emergency departments during 2010 and 2011. General practice consultation rates were taken from QResearch.4 We calculated urgent care centre attendance rates per 100 by age and sex, by standardising attendances among Hammersmith and Fulham residents to the Hammersmith and Fulham Borough population. Emergency department rates were calculated using Hospital Episode Statistics (HES) and the adolescent population of Hammersmith and Fulham Boroughs (rates were directly standardised to mid-2009-year English population by year of age).11 Emergency department data were only available for adolescents in combined age groups, that is, under age 19 years. All analyses were performed using Stata V.11.


There were 215 424 patient attendances at the urgent care centres during the period 1 October 2009 to 31 March 2012, of which 6.5% (N=14 038) were among the adolescent age group (10–19 years); 9027 (4.2%) at Fulham, and 5011 (2.3%) at Hammersmith. There were no significant age and sex differences in attendances among the adolescent group at each site.

Baseline characteristics of adolescent attendances

Among adolescents, the number of urgent care attendances increased with age (table 1). Those aged under 15 years (n=2818) accounted for 20.1% of total attendances (95% CI 19.4% to 20.8%); 54.5% of adolescents attending the urgent care centres were girls (95% CI 53.7% to 55.3%). The proportion of attendees who were girls increased with age (p<0.0001), and 19-year-old girls accounted for 59.4% of attendances. Among the 10–14 year group, attendances by boys predominated (54.4%; CI 51.6% to 55.3%), and in the 15–19-year group more attendances were by girls (57.6%; CI 55.6% to 57.4%).

Table 1

Baseline characteristics of adolescent attendances at Hammersmith and Fulham Urgent Care Centres, by age N=14 038

Adolescent attendees, 53.2%, were from the white ethnic group, followed by black or black British adolescents (15.4% of attendances), and Chinese or other groups (9.6% of adolescent attendees); 7.9% of adolescent attendances did not have a recorded ethnic group.

Attendances, 13.2% (n=1853), were recorded as being among adolescents not registered with a GP (95% CI 12.6% to 13.8%). This varied significantly by age (p<0.001). Among adolescents not registered with a GP, 82.9% were aged 16 years and over. Of adolescent attendees, 53.3% gave a postcode of residence within the London Borough of Hammersmith and Fulham. Only 5.8% were from outside the London region, with 39 attendees classified as non-UK residents.

Types of adolescent attendance according to streaming

Adolescents aged 18–19 years accounted for 50.4% of adolescent attendances during core contract hours, and 49.6% of adolescent attendances ‘out of hours’ (table 2). A significant difference was observed among adolescents aged 10–15 years attending during GP contract and ‘out of hours’ (p<0.0001). They accounted for 22.2% (n=6684) of total attendances during GP contract hours (95% CI 20.1% to 24.3%), and for 29.9% (n=7354) of attendances ‘out of hours’ (95% CI 27.9% to 31.8%).

Table 2

Adolescent attendances according to type of streaming in and outside core general practitioner (GP) hours, by age (N=14 038)

The minor injury stream had the highest percentage attendance (31.6%), followed by 25.1% for GP priority stream, and 16.4% for minor illness stream during core contract hours. Out of hours saw a similar pattern of attendance by stream; however, the see-and-treat stream ranked third followed by minor illness stream. Adolescents aged 18–19 years were the most likely age group to be referred to the emergency department. Urgent care centre attendance varied significantly by stream and age group (p<0.0001). The 14–19-year age group was most likely to leave without waiting to be seen (p<0.001).

Top 10 clinical reasons for adolescent attendances

There were a total of 11 556 primary READ codes allocated to adolescent attendances. A further 231 adolescent attendances (2.0%) did not have a designated READ code (table 3). The commonest clinical reason for adolescents to attend the centres was for musculoskeletal conditions and injuries (excluding fractures), accounting for 30.2% (n=3486) of cases. The number of clinical diagnoses increased significantly as age increased (p<0.0001). We found that older children were more likely to have a musculoskeletal and injury problem. The next commonest reason for attendance was respiratory tract infection (12.5%), followed by limb fractures (5.1%). Other common clinical reasons for attendance included skin infection (3.8%) and abdominal pain (3.3%).

Table 3

Top 10 presenting clinical diagnosis (categorised by primary READ code allocated, N=11 556) among adolescent patients aged 10–19 years attending the Urgent Care Centres

Comparison of usage of urgent care with routine general practice

The combined urgent care centre attendance rate for 2010 and 2011 among 10–14-year-olds was 8.3 per 100 (95% CI 7.8 to 9.1) compared with the national GP attendance rate during the same time period 16.0 per 100; adolescents aged 10–14 years were less likely to visit the centres than their GPs (p<0.0001) (table 4). By contrast, boys and girls aged 15–19 years were more likely to visit the centres (30.6 per 100; 95% CI 29.9 to 31.3) for their healthcare needs than visiting their GPs (23.4 per 100; 95% CI 23.32 to 23.42), p<0.0001.

Table 4

Comparison of urgent care centre attendances with general practice attendances, by adolescents during 2010 and 2011

The proportion of adolescent attendances at the urgent care centres in 2010 was higher (18.5%; 95% CI 17.1% to 19.8%) than emergency departments (13.43%; 95% CI 13.41 to 13.44), p <0.0001. A similar relationship of adolescent urgent care centre and emergency department attendance was observed in 2011.


Main findings

We found an attendance rate of urgent care of 19 per 100 adolescents, which accounted for 6.5% of all attendances. Older adolescents were more likely to attend the centres than routine general practice; whereas the opposite finding was observed among younger adolescents. We also found 13.2% of adolescents reported they had no GP. The majority attended for conditions commonly seen in primary care including musculoskeletal conditions and injuries, and respiratory tract infections, and also fractures. Girls aged 15–19 years were most likely to attend the centre than any other adolescent group.

Musculoskeletal conditions and injuries accounted for 30% of attendances at the urgent care centres, and are likely to reflect the previous profile of the emergency departments. This is because the urgent care centres are colocated with the hospital's emergency departments and, therefore, are not perceived by patients as being different. Furthermore, adolescents cannot directly access the emergency department; they can only do so if referred by urgent care staff or if they are brought in by ambulance.

Therefore, there is still unmet need among this age group in accessing ongoing health advice in managing such injuries. This could be provided in primary care, or a more community-based setting.

Limitations of study

To our knowledge, this is the first study of adolescent attendance at GP-led urgent care centres, a new service model that is now being commissioned. We used 30 months of data with good data completeness to examine time trends and, therefore, mitigate the effect of seasonal variation in patterns of urgent care attendance.

We used QResearch, an electronic database of around 660 general practices that use the Egton Medical Information Services (EMIS) GP clinical system, to determine general practice consultation rates. The QResearch database is representative of the UK general population and, therefore, unlikely to introduce age–sex bias. We used HES data to calculate the national emergency department attendance rate, and this may be a potential limitation. HES data are incomplete when compared with Quarterly Monitoring of Accident and Emergency data reports. However age–sex bias is unlikely to explain the differences in rates of urgent care and emergency department use, as missing HES data are mainly from a group of emergency departments that do not report any data, while those departments that do report, report fully. We assumed the catchment population for urgent care and the emergency department was the London Borough of Hammersmith and Fulham. It is likely that calculated attendance rates are an underestimate due to density of services in London, and that people are more likely to travel further for emergency care than urgent care.

Although we are unable to compare attendance figures between emergency departments and the urgent care centres by similar age groups (data was only available in combined age groups), we did find that overall attendances were higher at the urgent care centres in both 2010 and 2011. We were unable to undertake a control study, which would enable a comparison of attendances in areas where there were no urgent care centres to fully evaluate the effect size. However, at the time, different models of urgent care centres, for example, stand alone, colocated with hospital and community centres, were being set up rapidly and without full evaluation.

Evidence of attendance with primary care problems

Adolescents are traditionally perceived as a healthy group with little need for primary care. Their mortality is low compared with other age groups, and has fallen over the last 10 years.2 However, nearly half of all adolescents report at least one current health concern.12 In a survey of 14–16-year-olds, 45% felt they had a health problem at the time, and in the previous 4 weeks, 37% had taken time off from school, and 70% had taken medications.13 Teenagers also report that they would like to discuss their health concerns with a doctor, although few have done so. Common concerns include sex, sexually transmitted diseases, contraception and pregnancy, acne, smoking, diet, exercise, weight and nutrition.14 ,15

However, within the current urgent care system, there is a lack of opportunity for adolescents to address these concerns, and only through access to GP services can they access medication, and preventive interventions, otherwise, they may present late with a more advanced condition, and have no opportunity to access health education.

Among adolescents aged under 16 years, nearly all would have been seen accompanied by an adult (not always necessarily by their parent or guardian). There are several reasons that influence parents choosing to take their child to the emergency department. These include access to and quality of primary care16; convenience and previous experience of healthcare services17 and parental education.18

We found 13.2% of adolescent attendances were not registered with a GP. It is rare for patients to deregister themselves from GP services unless they move area. Given that 98% of younger children are registered with UK GPs, it may be that adolescents are unaware of whom their GP is, or are choosing not to disclose this fact to avoid being sent away back to their GP. Furthermore, not being registered with a GP does not mean that they are unable to access primary care; they can do so as a temporary resident, or can be seen if requiring immediate treatment. Alternatively, this high figure may reflect a mobile population, especially among older adolescents, for example, those in full-time education who move, and may avoid, or lack knowledge of, the local primary care system and, consequently, choose to attend urgent care services. We found the comparative figure for adults attending the urgent care centre to be higher at 14.9%.

Nevertheless, the high percentage of adolescent centre attendees who are registered with a GP practice, suggests that adolescent attendances are not simply due to lack of GP registration. The proportion of adolescent attendances streamed as ‘GP priority’ was just as high during GP core contract hours (25.1%) as out of hours (22.8%), supporting the view that barriers to primary care access may be more complex among this group.

Although we found that musculoskeletal conditions and injuries were the commonest clinical reason for attendance, ranking of clinical conditions is not the same as evidence of appropriateness of attendance. Attendance with musculoskeletal conditions along with limb fractures could be entirely appropriate for an urgent care centre. Similarly, there would be more merit in asking GPs to consider how they met the needs for conditions, such as respiratory, skin, gut and urinary infections which, in total, account for over one in five of adolescent urgent care attendances. However, caution is required in interpreting these as ‘primary care type’ attendances, as urgent care centres have greater access to investigative and treatment facilities and, therefore, more capacity than is available in traditional primary care settings.

Implications for health policy

The investment in urgent care is not seen by all as a solution to rising demand for unplanned care in emergency departments.19 If GP-led urgent care models continue, it will be important to monitor changes in numbers and reasons for attendance over time, especially if there is a move to substitute urgent care centres for emergency departments, and evaluate their effectiveness in reducing demand. If policy focuses on reducing urgent care need because of escalating costs, then further work is required to promote self-management of musculoskeletal conditions and infections among adolescents and responsiveness of primary care.


We found that adolescents aged 15–19 years had higher rates of attendance at urgent care centres than general practices. Musculoskeletal conditions, respiratory tract infections and fractures were the main reason that adolescents attended. Primary care services in urban areas may need to be more responsive to the needs of adolescents if use of urgent care centres by this group is to be reduced.


We thank all the staff who have worked for, and currently work for Partnership for Health in setting up the services; and staff at Hammersmith and Charing Cross Hospital emergency departments. Partnership for Health is a consortium of London Central and West Unscheduled Care Collaborative, Imperial College London Healthcare Trust (and Central London Community Health Services. The Department of Primary Care & Public Health at Imperial College London is grateful for support from the NW London NIHR Collaboration for Leadership in Applied Health Research & Care (CLAHRC), the Imperial NIHR Biomedical Research Centre, and the Imperial Centre for Patient Safety and Service Quality (CPSSQ). SS holds an NIHR postdoctoral fellowship.



  • Contributors SG, HM and SS were responsible for the study conception and design. SI and FR were responsible for data integrity and analysis. All authors have had full access to study data, and take responsibility for the accuracy of data analysis. All authors were responsible for critical manuscript revision and approval. SG is the study guarantor.

  • Funding The new service model was funded by NHS Hammersmith & Fulham. The Department of Primary Care & Public Health at Imperial College London received funding from the Imperial College Healthcare NHS Trust to help evaluate the new model. SS holds an NIHR post-doctoral fellowship.

  • Competing interests SG, SI, FR, AM and SS are employed by Imperial College London, which received funding to help evaluate the new model of care.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement Primary data associated with the research paper is available on request to the corresponding author. Further unpublished data is available to researchers on request.