Key message:
Nearly 60% of Emergency Presentations resulting in a new diagnosis of cancer come through A&E, with 30% being emergency referrals from GPs.Emergency referrals to outpatients are higher for certain cancers. Survival estimates for this group are higher than other emergency subgroups and more comparable to survival from “managed” Routes.
Routes to Diagnosis: Exploring Emergency Presentations
Background
The Routes to Diagnosis study showed that 24% of newly diagnosed cancers in 2006-2008 (excluding non-melanoma skin cancer) first presented into secondary care as an Emergency Presentation. Relative survival estimates for Emergency Presentations were significantly lower than for other Routes across all sites.
The Emergency Presentation Route comprises different emergency pathways into secondary care, including A&E attendance, emergency GP referrals to an inpatient setting (non – two week wait referrals) and emergency admissions to either an inpatient or outpatient setting.
This data briefing looks at the breakdown of Emergency Presentations by the different emergency pathways and explores whether there are differences by cancer site and whether outcomes differ by each emergency subgroup.
Emergency subgroups
In the Routes to Diagnosis study Emergency Presentations are assigned using inpatient and outpatient HES data. When assigning a Route several pathways are grouped together into the Emergency Presentation Route.
These different pathways are assigned as emergencies based on the source of referral (for pathways which started as a referral to an outpatient setting) or admission method (for pathways which started as a referral to an inpatient setting). Table 1 shows these different emergency pathways. Some pathways are similar in nature and contain a very small proportion of patients and they have therefore been grouped together into emergency subgroups. For all cancers combined, nearly 60% of Emergency Presentations were inpatient admissions following an A&E attendance (A&E), 30% were inpatient admissions following an emergency referral from a GP (GP), 4% were other emergency admissions to inpatients (IP emergency) and 6% were emergency referrals to outpatients (OP emergency).
Table 1: Breakdown of Emergency pathways by admission method or source of referral
Table 2: Emergency presentations and subgroups by site, all ages, England, 2006-2008
Emergency subgroups by cancer site
Table 2 shows that for the majority of sites, a similar proportion of Emergencies present through each emergency subgroup. The highest proportion of A&E emergencies was seen for cancers of the central nervous system (70% of emergencies) whilst the highest proportion of GP emergencies was seen for pancreatic cancer (38%). Differences from the overall proportions are observed for head and neck cancers, sex specific cancers, melanoma and sarcomas. These sites have a significantly higher proportion of emergency referrals to an outpatient setting compared to all cancers combined, ranging from 12% (testis) to 35% (oral cavity). For both oral cavity and oropharynx, a greater proportion of emergencies were diagnosed through the OP emergency subgroup than as emergency GP referrals.
Children, teenagers and young adults
Whilst a higher proportion of childhood (aged 0-14) cancers were classified as Emergency Presentations (54% compared to 24% for all ages), the proportion by emergency subgroup doesn’t differ greatly from the all age distribution with the exception of sarcoma (connective and soft tissue) which showed a noticeably lower proportion of GP emergencies (16%) and a higher proportion of OP emergencies (31%).
For cancers diagnosed in teenagers and young adults (TYA, aged 15-24), a similar distribution by emergency subgroup was seen for each cancer site compared to the distribution for that site for all ages combined. Differences were seen for a small number of cancers; Colorectal cancer in TYA does have a slightly higher proportion of A&E emergencies (71% compared to 58% for all ages) with fewer GP emergencies, acute myeloid leukaemia has a slightly lower proportion of A&E emergencies for TYA (47% compared to 59% for all ages) with a lower proportion of GP emergencies and ovarian cancer showed a higher proportion of OP emergencies (17% compared to 5% for all ages).
Survival by emergency subgroup
The Routes to Diagnosis study showed that Emergency Presentations have poorer one-year relative survival than other Routes. Figure 1 shows that for lung cancer there is also variation in relative survival estimates between different types of emergency subgroup. The OP emergency pathway has higher relative survival estimates than other emergency subgroups whereas relative survival estimates for the A&E and GP emergency pathways produce similar survival estimates at both one month and one year.
Figure 1: 1-month and 12-month relative survival estimates for lung cancer for non-emergency Routes and Emergency subgroups, England, 2006-2008
The higher relative survival estimates for the emergency OP pathway shows survival estimates that are closer to the “managed” presentation Routes of Two Week Wait, other GP referrals and non-emergency inpatient and outpatient Routes. Further work will be undertaken to try and understand more about these emergency referrals that do not lead to the poorer survival seen from other emergency pathways.
One year survival estimates are shown in table 3 by broad age group for four selected cancer sites. Across age groups and sites, the general pattern of survival for OP Emergency subgroup being higher and more similar to non-emergency Routes is repeated.
Table 3: 12-month relative survival estimates by emergency subgroup and TWW Route by site and age group, England, 2006-2008 followed up to 2009
Conclusions and summary
Almost 90% of patients classed as an Emergency Presentation enter secondary care on their way to being diagnosed with a cancer through two emergency subgroups; A&E (59% for all cancers combined) or an emergency referral to inpatients from a GP (30% overall). A similar distribution to these two subgroups is seen for most individual sites. Relative survival estimates for A&E and GP emergencies were similar for four different cancer sites examined. Producing results by these different emergency pathways would allow for further exploration to see if differences exist for specific sites or by equality groups.
Overall, 6% of emergencies are diagnosed through the OP emergency pathway although for head and neck cancers, melanoma and sex-specific sites a much higher proportion of patients first present through this pathway than for other cancers. Relative survival estimates for OP emergencies are higher and more comparable to survival from more managed routes such as TWW, non-emergency GP referrals and other outpatient and inpatient elective Routes. This may imply that these cancers are at an earlier stage than other emergencies yet presenting symptoms have caused an emergency referral. Further work will be undertaken to investigate the presenting symptoms of these cancers in order to understand this further.
i “Routes to diagnosis for cancer – determining the patient journey using multiple routine data sets” Br J cancer, doi:10.1038/bjc.2012.408
ii The Bed Bureau is the centre for day-to-day communication between General Practitioners, the Patient Services Co-ordinators and clinical staff. It supports the timely admission of emergency and GP admissions to the Trust.
FIND OUT MORE:
Routes to diagnosis
Other useful resources within the NCIN partnership:
Cancer Research UK CancerStats