This site is intended for health professionals only

Clinical round-up: November

GPs ‘twice as good as apps’ at diagnosis

GPs outperformed symptom-checker apps by two to one when diagnosing illnesses, according to new research.

The US study found that doctors were far more accurate than digital symptom-checker apps at diagnosis, making the correct diagnosis 72% of the time, compared with a 34% success rate for the apps. And 84% of clinicians listed the correct diagnosis in their top three possibilities, compared with 51% for digital symptom checkers.

Researchers used 45 standardised patient cases to test the accuracy of 234 GPs versus 23 symptom checkers. The cases included symptoms and history items, but not exam or test findings.

Lead author Professor Ateev Mehrotra, associate professor of health at Harvard Medical School, said: ‘While the computer programs were clearly inferior to doctors in terms of diagnostic accuracy, it will be critical to study future generations of programs that may be more accurate.’

JAMA Int Med 2016, online 10 Oct

NSAIDs ‘raise heart failure risk in elderly’

NSAIDs are associated with an increased risk of heart failure in elderly patients, according to a new study in the BMJ.

The research looked at 10 million people from the UK, Holland, Italy and Germany, with an average age of 77, who took NSAIDs such as ibuprofen.

Those taking any NSAID in the previous 14 days had a 19% increased risk of admission with heart failure, compared with others who had used NSAIDs at any point in the previous six months.

The increase in risk of admission varied by dose and NSAID. Naproxen was linked with a 16% increased risk, while ketorolac increased risk by 83%. No absolute risks were included.

But the study focused on prescription NSAIDs, and not over-the-counter NSAIDs, which the authors admitted are ‘typically used at lower doses, by younger people and for shorter durations’.

Professor Peter Weissberg, medical director at the British Heart Foundation, claimed the study reinforced previous research establishing a link between NSAIDs and developing heart failure.

He said: ‘This study serves as a reminder to doctors to consider carefully how they prescribe NSAIDs, and to patients to take the lowest effective dose for the shortest possible time.’

BMJ 2016, online 28 Sep 

Talking to receptionist ‘deters patients’

Having to discuss symptoms with practice receptionists is one of the main factors that puts patients off from seeing a GP.

A survey of almost 2,000 people, investigating commonly perceived barriers to seeing a GP, reported that 36.6% of men and 42.6% of women said they ‘don’t like having to talk to the GP receptionist’ about symptoms.

The analysis, conducted by Cancer Research UK, revealed that another key barrier to seeing a GP was difficulty getting an appointment with a particular doctor or at a convenient time.

GPC deputy chair Dr Richard Vautrey added: ‘All receptionists receive training to help ensure that when a patient calls they are given the most effective advice about what appointment they may need, but it is always made clear they are under no obligation to disclose information they are not comfortable with.’

J Pub Health 2016, online 8 Sep

Fifth of patients ‘on drugs they don’t need’

Taking too many types of medication is likely to lead to hospitalisation in the over-75s, new research has suggested.

Patients on 10 or more medications are three times more likely to be hospitalised, according to the research, presented at the RCGP’s annual conference last month.

Pharmacists conducted a review of the medical care of more than 1,800 patients aged over 75 over a year, and found that on average each patient had 14 GP consultations and was prescribed more than six different medicines.

More than 20% of patients were found to be on medication they did not need. As a result of the review, 35% of patients had their medication stopped and 4% had their doses reduced.

The research said that ‘medication reviews are a key strategy to help practices and patients address [the problem of hyperpolypharmacy]’, concluding that considering the cost of the programme and the reduction in admissions ‘the clinical programme would pay for itself 1.8 times in a year’.

RCGP annual conference 2016, poster

  • Managing multimorbidity, from page 58

Naproxen ‘relieves pain in hip or knee OA’

Low-dose naproxen is effective in reducing pain in hip and knee osteoarthritis, according to a new trial.

Patients with moderate pain from were randomised to receive either naproxen 660mg (440mg in patients over 65 years) or placebo for a week.

Those receiving naproxen were found to have significantly improved pain scores at rest and with weight bearing, as rated on a five-point scale, compared with the placebo group. Rates of adverse events were similar – 27% of those taking naproxen, and 28% of those on placebo.

The authors concluded that non-prescription doses of naproxen ‘effectively relieve pain in patients with mild to moderate OA of the hip or knee’.

RCGP annual conference 2016, poster

Domestic violence course for GPs

A new evidence-based training intervention could help GPs manage patients suffering from domestic abuse.

The RESPONDS intervention encourages GPs to engage ‘more extensively’ with such patients and helps them prepare to safeguard children and vulnerable adults.

Lead researcher Dr Eszter Szilassy, from the University of Bristol’s Centre for Academic Primary Care, said: ‘While GPs are fully aware of their child safeguarding responsibilities… the lack of relevant training contributes to failures to translate this knowledge into safe and effective domestic violence-related practice strategies.’

Health Soc Care Community 2016, online 14 Oct

CPD Tip of the month

Subungual haematoma or malignant melanoma?

Lesions under the nail can be difficult to assess, but there are a number of pointers to look for:

  • Check for any history of trauma.
  • Is there any evidence of extension of colour change outside of the subungual area and into the nail bed and surrounding finger or toe pulp (a positive Hutchinson’s sign)? This would be highly indicative of a melanocytic lesion and quite probably a malignant melanoma.
  • How long ago did the patient notice the pigmented colour change under the nail? A subungual haematoma should grow out over six to eight weeks.

How not to miss melanoma (1.5 CPD hours)