Journals Watch: Cancer mortality, diabetes complications and risks of working long hours

The latest research relevant to GPs.

Practices should review their use of urgent referral pathways
Practices should review their use of urgent referral pathways


It is widely assumed that cancers that are diagnosed earlier are likely to have a better prognosis and improved survival.

Urgent referral pathways allow for rapid access to diagnostics and specialist opinion for patients with possible symptoms of cancer.

Study design
This cohort study of 8,049 general practices and 215,284 patients with cancer in England looked at time to death of patients diagnosed with cancer in 2009, followed up until 2013. Three measures were derived from the data.

  • Practice referral ratio: number of urgent referrals for suspected cancer (standardised according to list size and age/sex distributions of people on the list) against expected number of urgent referrals (according to list size data and national age and sex specific rates)
  • Practice conversion rate: proportion of urgent referrals resulting in cancer diagnosis (PPV for cancer among urgent referrals)
  • Practice detection rate: number of cancers resulting from an urgent referral by a GP Practices with a small list or a large change in their list size were excluded, to minimise year by year fluctuations seen with small numbers.

Key findings
A high referral ratio and/or detection rate were associated with reduced mortality. The risk of death at four years where there was a high referral ratio and a high detection rate was 47%, compared with 52% where there was both a low or intermediate referral ratio and low or intermediate detection rate.

This study involved a large number of patients, so the figures represent a sizeable group for whom survival might have been improved with more use of the urgent referral pathway.

Implications for GPs
Referring via an urgent referral pathway may be beneficial to survival, but their use varies. Practices could consider reviewing their use of urgent referral pathways, particularly where referral rates are consistently low, to ensure clinicians make the best use of local pathways.


Each year in the UK, approximately 20,000 patients are diagnosed with bladder or renal cancer. Most of these patients present first to primary care with symptoms.

The aim of this study was to quantify the risk of bladder and/or renal cancer in patients presenting to primary care with possible symptoms.

Study design
The authors carried out a systematic review of papers published between 1980 and 2014, looking for retrospective, prospective or case-control diagnostic accuracy studies of symptomatic patients presenting to primary care with one or more symptoms, for whom follow-up data were available. Eleven studies were included, covering a total of 3,451,675 patients. Studies were appraised using the QUADAS-2 tool.

QUADAS-2 consists of four key domains to evaluate the risk of bias and applicability of primary diagnostic accuracy studies:

  • Patient selection
  • Index test
  • Reference standard
  • Flow and timing

For each symptom of possible cancer, the studies were analysed to find the number of patients with that symptom who had renal or bladder cancer (true positives) and the number of patients with the symptom who did not have renal or bladder cancer (false positives). PPVs were then calculated for each symptom.

Symptoms included haematuria, UTI, dysuria, fatigue, abdominal pain, back pain, loss of appetite, constipation, nausea, DVT and anaemia.

Key findings
The PPV for visible haematuria was 5.1%. This was the only high-risk symptom for renal or bladder cancer, with other symptoms studied having a PPV of 1.4% or less.

Having both haematuria and additional symptoms increased the risk of cancer, as did having unresolving haematuria.

Implications for GPs
This review supports the current practice of investigating symptoms of possible renal or bladder cancer together.

Patients with haematuria warrant careful assessment and investigation, which should include checking for persistent non-visible haematuria in patients presenting with UTI.

This study demonstrates that the absence of haematuria does not rule out the possibility of renal or bladder cancer in the presence of other symptoms, but that the risk is lower.


Patients with diabetes can be assessed for risk of blindness

This research provides risk tools to be used with patients who have diabetes, to assess their 10-year risk of developing blindness or of having a lower limb amputation.

Study design
Data from the QResearch database were used to develop risk prediction equations for estimating the risk of blindness and lower limb amputation over 10 years in men and women (aged 25-84 years) with diabetes.

Data from 763 general practices, involving 454,575 patients with diabetes, were used and the risk prediction equations were subsequently validated in another 357 practices (206,050 patients).

Key findings
The algorithms developed by the authors of this study can be used to calculate the absolute risk of developing these complications over 10 years for patients with type 1 or type 2 diabetes.

They are available to use online at

Implications for GPs
These complications lead to a large financial and time burden on NHS services and can also provoke fear and anxiety in patients.

Being able to quantify the risks for individual patients can now form part of patient education and care planning discussions.


This study assessed the prevalence of, and variation in, potentially hazardous prescribing in general practice.

Study design
This cross-sectional study involved 526 surgeries contributing to the Clinical Practice Research Datalink up to 1 April 2013.

Only adult patients were included in the study and the resulting data were analysed for either potentially hazardous prescriptions, for example, drug combinations, or inadequate monitoring procedures.

The research covered prescriptions for drugs such as anticoagulants, NSAIDs, oral hypoglycaemics and beta-blockers, and the monitoring procedures for patients prescribed these drugs, for example, blood tests for patients taking ACE inhibitors or amiodarone.

Key findings
There was a wide variation between practices in the prevalence of different types of potentially hazardous prescribing (almost zero to 10.2%) and for inadequate monitoring of patients (10.4% to 41.9%). Risks were higher for older patients and those with multiple repeat medications.

Implications for GPs
This study produced some unsettling data involving well-known hazards, for example, prescribing NSAIDs for patients with chronic kidney disease, or monitoring thyroid function with amiodarone prescription.

The current high workload in practices can make it difficult to keep on track with repeat prescriptions and monitoring requirements.

The presence of practices with very low rates of potentially hazardous prescribing or inadequate monitoring implies that it is possible to minimise such hazards to our patients, and we should be doing more to encourage targeted practice-level interventions.


Long hours can raise stroke risk, study found

Long working hours have been implicated in the development of various adverse health outcomes. Here the authors show an increased risk of stroke in those working long hours, something many doctors are used to doing.

Study design
Data were extracted from published and unpublished prospective cohort studies to produce this large meta-analysis.

Studies were identified using the search terms work hours, working hours, overtime work and coronary heart disease, ischaemic heart disease, acute myocardial infarction, angina pectoris, chest pain, stroke, cerebrovascular and cerebro­vascular disease. Long working hours were defined as more than than 55 hours per week.

Key findings
Working 55 hours or more per week raises stroke risk to 1.3 times that of a person working standard hours. This did not vary between men and women or geographical area. The association with cardiovascular disease was weaker.

Implications for GPs
Discussing the risk of cardiovascular or cerebrovascular disease with patients has become a normal part of life as a GP.

This study implies that asking about working hours is also important when considering risk reduction strategies. In addition, it serves as a useful reminder to ensure that GPs look after their own health.

  • Dr Collier is a GP in Kent

Reflect on this article and add notes to your CPD Organiser on MIMS Learning


  1. BMJ 2015; 351: h5102
  2. Br J Gen Pract 2015; DOI:10.3399/bjgp15X687421
  3. BMJ 2015; 351: h5441
  4. BMJ 2015; 351: h5501
  5. Lancet 2015; 386: 1739-46

Before commenting please read our rules for commenting on articles.

If you see a comment you find offensive, you can flag it as inappropriate. In the top right-hand corner of an individual comment, you will see 'flag as inappropriate'. Clicking this prompts us to review the comment. For further information see our rules for commenting on articles.

comments powered by Disqus