Objective To evaluate the role of the urgent 2-week referral pathway in facilitating the early diagnosis of childhood cancer.
Design Retrospective case notes review of all children referred via the pathway and all children diagnosed as having cancer via other means over a period of 3½ years (January 2007–July 2010).
Setting District general hospital with a shared care oncology service.
Patients 82 children aged 0–15 years.
Outcome measures Outcomes for children referred via the 2-week urgent referral pathway, including diagnosis, need for further investigations and need for further specialist input.
Results 35 children were referred via the 2-week urgent referral pathway with suspected cancer. Only one of these children had a malignancy. 49% of children required no further investigations and 43% were discharged after the first appointment with no follow-up thought necessary. Over the same time period, 47 children were diagnosed as having cancer via other means.
Conclusion The pickup rate of cancer among children referred via the 2-week urgent referral pathway is extremely low and the vast majority of cancer is diagnosed via different routes.
Statistics from Altmetric.com
The 2-week urgent referral pathway for suspected cancer has been in place for over a decade. It was introduced in 2000 with the publication of the Department of Health's NHS Cancer Plan1 which stipulated that all patients, including children, with suspected cancer must be seen by a specialist within 2 weeks of referral by their general practitioner (GP). This was followed by publication of National Institute for Health and Clinical Excellence guidelines2 advising on symptoms and signs suggesting the need for an urgent 2-week referral. The aim of both documents was to facilitate early diagnosis and treatment of cancer and hence improve survival.
Ten years later, concern remains that cancer mortalities in the UK appear higher than those in many other developed countries.3 In early 2011, the Department of Health published another document Improving Outcomes: A Strategy for Cancer.4 This supported an ongoing role for the 2-week urgent referral pathway in helping facilitate early diagnosis of cancer.
Despite government endorsement of the pathway, published evidence of benefit is limited. We found no published data on the use of the pathway in paediatric populations. Studies in adult populations have identified a number of areas of concern. The cancer pickup among adult patients referred via the pathway is variable depending on cancer type, but overall the yield is low (4–51%) and the majority of cancer cases (35–94%) are still being diagnosed outside of the 2-week referral pathway.5,–,8 A study looking at the role of the 2-week urgent referral pathway for suspected brain tumours (which included a number of young adults) found that only 9% of those referred had a malignancy and 94% of tumours were diagnosed outside of the pathway.9 A study looking at haematological malignancies found that only 14% of lymphomas were diagnosed via the 2-week pathway.10 Another published criticism is that the 2-week pathway is not impacting on overall time from onset of symptoms until definitive treatment is started. For example, since the introduction of the urgent referral pathway, time from GP referral to first appointment has decreased but there has been no change in the time from presentation to starting treatment.11 12 The most widely published and arguably most important concern is that, as a result of the resources needed to support the urgent referral pathway, there has been a significant increase in the waiting time for ‘non-urgent’ referrals. As the majority of cancer cases are still diagnosed via the non-urgent route, this results in a longer wait for the ‘average’ patient with cancer.5 13 14
What is already known on this topic
▶ All children with suspected cancer have to be seen by a specialist within 2 weeks of referral by their general practitioner.
▶ The 2-week urgent referral pathway has a low cancer yield in adult populations.
▶ The majority of adult cancer cases are diagnosed outside of the 2-week referral pathway.
What this study adds
▶ The 2-week urgent referral pathway has an even lower cancer yield in paediatric populations.
▶ The vast majority of cancers in children are not diagnosed via the 2-week urgent referral pathway.
We aim to evaluate the role of the 2-week referral pathway in a paediatric population.
A retrospective case notes review was performed at a single district general hospital. The hospital is based in West Hertfordshire and covers a population of about 600 000. It has a shared care oncology agreement with two local paediatric tertiary oncology centres and refers children as needed. In West Hertfordshire all children referred via the 2-week pathway are seen by a single paediatric consultant with training in paediatric oncology. If children are referred to another specialty, for example ear, nose and throat, skin or orthopaedics, they are automatically re-directed by the staff in the 2-week wait office to the paediatric consultant. A database is kept by the consultant of all children referred via the 2-week pathway.
The consultant also maintains a database of all children living within West Hertfordshire who are diagnosed as having cancer irrespective of route of referral. The tertiary oncology centres inform the district general hospital of all children living within the hospital catchment area who are newly diagnosed as having cancer and hence the database covers the entire population of children within the referral area of West Hertfordshire.
The database was retrospectively reviewed to identify all children referred via the urgent 2-week referral pathway over a period of 3½ years (January 2007–July 2010). The notes of those children were then reviewed and data collected on outcome following referral using predesigned data collection sheets.
The database was also reviewed to identify all children diagnosed as having cancer via other pathways over the same 3½-year period.
Thirty-five children were referred via the 2-week urgent referral pathway with suspected cancer over the 3½-year study period. The referral rate remained relatively constant throughout the study period. The characteristics of the children and the reasons for referral are shown in table 1. The most common reason for referral (45%) was concern about potentially abnormal lymph nodes.
All the children were seen within 2 weeks of referral (usually within 1–5 days) and assessed clinically. Seventeen children (49%) required no further investigations and a clinical decision was made that the child did not have a malignant condition. The rest of the children had at least one investigation with some children requiring multiple investigations.
Of the 35 children referred, only one child was diagnosed as having a malignant condition (Wilm's tumour). No significant abnormality was found in 26 children (74%). Fourteen of these had normal or reactive lymph nodes, five had self-resolving headaches, two had reactive changes on their blood tests which corrected with time and five had an entirely normal examination, including two babies referred due to concerns about their red reflex and one child referred with a normal prominent xiphisternum. Eight children (23%) were diagnosed as having a non-malignant diagnosis—two children with haemangiomas, one with migraine, one with sinusitis, one with inguinal hernia, one with a salivary duct stone, one with eczema and one with a benign osteoid osteoma.
The majority of children (28, 80%) were discharged after the first appointment or followed up locally by the paediatric team. Only three children were referred for a tertiary oncology opinion—the child with the Wilm's tumour, a child with extensive lymphadenopathy (which proved to be reactive) and the child with the benign osteoid osteoma. The outcomes are summarised in table 2.
Over the same 3½-year study period, 47 children were diagnosed as having cancer via other pathways—15 with acute leukaemia, nine with brain or spinal tumours, five with lymphoma, six with sarcoma, three with neuroblastoma, two with Wilm's tumour and seven with other rarer tumour types.
The referral pathways for this group varied greatly and often involved multiple visits to various healthcare providers outside of paediatrics. Time from first symptom to diagnosis varied from 1 day to 6 months. Referral pathways included attendance to the emergency department for symptoms related to the cancer or for other reasons with the cancer being picked up incidentally, routine GP referral when the waiting time for an appointment is 6–8 weeks, referral from other specialities and referral direct to the tertiary oncology centre from other tertiary specialists, for example, orthopaedics, ophthalmology, neurology or neurosurgery. The district general hospital operates a rapid access outpatient clinic where patients are seen within a week, but no children were diagnosed as having cancer via this route.
The pickup rate of cancer among children referred via the 2-week urgent pathway was 2.8% (only one patient) and 98% of cancers were diagnosed via a different pathway.
In our population the 2-week urgent referral pathway has not helped facilitate the early diagnosis of childhood cancer. The pickup rate is extremely low and the vast majority of cancer cases are being picked up via different means. The results from our study show a lower cancer yield than the studies in adult populations.5
If these results are representative of the general paediatric population one might be tempted to suggest that the 2-week urgent referral pathway is of no benefit in paediatrics. However, before one can make this conclusion one needs to consider if the pathway has any other benefits aside from facilitating early diagnosis of cancer. The pathway may be providing benefit in terms of decreasing parental or child anxiety. If a child or parent is worried about cancer, and the GP is unable to reassure them, the benefit of seeing a paediatric expert within 2 weeks may be huge. The converse may apply and parents who were not particularly concerned about their child may then become extremely anxious when informed that they are being referred for a presumed cancer. However, GPs do not usually inform parents about the reason for the referral and the fact that a cancer has been suspected. This does result in parents being unable to understand why they are being summoned to the hospital for an urgent appointment for their child. The psychosocial effect on a child and family of being referred via a 2-week urgent referral pathway has not been evaluated.
The pathway may benefit GPs in terms of providing them with support and easy access to tertiary care. Previous studies have found mixed opinions among GPs, with 50% thinking it improved communication and access to specialist care.15 Another potential benefit which is difficult to evaluate is the prevention of unnecessary urgent referrals to accident and emergency (A&E). One could postulate that if the 2-week pathway did not exist and a GP was concerned about a lymph node, they may refer the child to the on-call paediatric team in A&E. This could result in the child being seen by a more junior member of staff resulting in further anxiety and perhaps unnecessary investigations. These unseen benefits need to be further evaluated before one can be certain that the 2-week pathway has no role in paediatric populations.
The impact of the 2-week urgent referral pathway on non-urgent waiting times has not been evaluated in this study. This has been a major concern in adult populations, with many studies reporting an increase in non-urgent waiting times.5 13 14 Given that the number of 2-week urgent referrals is much smaller in the paediatric population and makes up a smaller percentage of our total referrals the impact may be less, but this needs further evaluation particularly given that almost all paediatric cancer is picked up via different pathways.
Finally, and probably most importantly, the study highlights ongoing concern that knowledge of early warning signs of childhood cancer are not well understood. Of all the children referred via the pathway, 49% required no further investigations and were reassured based purely on clinical grounds. Yet at the same time 47 children with early signs and symptoms of cancer were not being referred via the pathway and were presenting via non-urgent referrals or self-referral to A&E. GPs have previously expressed concern about missing childhood cancer based on evidence that they will likely only see one case every 20 years, the symptoms and signs of different cancers vary greatly and the predictive value of a single symptom or sign is extremely low.16 17
Early diagnosis of childhood cancer will be improved not by ongoing support and investment in the 2-week urgent referral pathway but by increased education and support for all healthcare providers in detecting the early symptoms and signs of possible childhood cancer.
Daniel Francis, medical student, helped with early data collection.
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.
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.