Corticosteroids for sore throat: a clinical practice guideline

We thank Dr. Linder for taking the time to consider and respond to our recommendation. A focus on hypothesis testing (and p-values) within individual studies can be misleading for a variety of reasons. Individual studies might be underpowered or overpowered, and p-values tell us nothing about the effect size or the importance of the outcomes. Instead of vote counting significant p-values, we and the GRADE approach suggest a focus confidence intervals around the absolute treatment effects for each patient-important outcome. We perform this assessment across the whole body of evidence, including the study by Hayward and colleagues, which provided the impetus for this Rapid Recommendation. The Hayward study's results are for the most part consistent with the results of the other nine randomised trials. As summarized in our infographic and in the systematic review supporting this recommendation (http://www.bmj.com/content/358/bmj.j3887), we also considered other factors contributing to the certainty of effects, namely the risk of bias, indirectness, inconsistency, and publication bias. Our panel considered all these factors across the entirety of the evidence base to issue the Rapid Recommendation.

We appreciate the discussion triggered by our trustworthy recommendation where we suggest to offer steroids for people with a sore throat attending their primary care physician. Most of the comments spoke from a societal point of view. We acknowledge the potential challenges such a change in practice may raise for health care systems, including a potential increase in consultations and opportunity costs that may follow. Careful readers would hopefully have noted that we take an individual patient perspective, rather than a system perspective, in creating the recommendations. Indeed, we encourage scrutiny and adaptation of the recommendations to contextualize to particular settings such as the NHS if needed. We hope those in need of adapting recommendations will follow standards for trustworthy guidelines and be explicit and transparent about judgments made when potentially arriving at modified recommendations against steroids for acute sore throat.

 Responders to our Rapid Recommendations state that there is a small extent of benefit, and the magnitude of benefit is insufficient to warrant a visit to the physician. This judgment is not, however, appropriately in the domain of the clinician, but rather in that of the patient.  An individual whose every swallow is an agony may well disagree that the magnitude of benefit is unimportant.  Indeed, our panel, in particular the patients on the panel, considered that the majority of patients – like those included in the trials – whose distress is sufficient to motivate them to make a visit to a health care facility, would judge the reduction in sore throat duration well worth a single dose of dexamethasone.  We aim to be transparent about these judgments as reflected in the statement “the panel believes that there is a great variability on how much reduction in pain severity or time to complete pain resolution each patient would consider important. Shared decision making may help establish what matters most to each patient.” We issued a weak recommendation to reflect the fine balance between desirable and undesirable consequences and need for shared decision-making.

 
That is what these Rapid Recommendations are all about. Informing clinicians and patients about current best evidence and bringing this evidence into practice with understandable data. This approach is perhaps to some extent disruptive to current practices in health care but nevertheless well aligned with accepted standards for evidence-based health care and shared decision-making. Patients should - if they would like to - be able to decide together with their clinicians which approach they choose, based upon trustworthy evidence. We believe this approach takes us one step further towards the goal of shared decision making, taking individual patients values and preferences and other factors into account.

Although we suspect that the vision of a tsunami of sore throat sufferers appearing at physicians' doors is unlikely, the system should – in one way or another – accommodate patients who judge the benefit important, and knowing that it is available would now (though not previously) have made a visit to a health care provider. Perhaps the day when a single dose of dexamethasone for sore throat is available at the local pharmacy may not be distant.

'The treatment is inexpensive and likely to be offered in the context of a consultation that would have happened anyway'.

I'm afraid this misses the point.
We have spent many years educating our patients to understand that for the vast majority of sore throats there is no useful therapy and that they do not need to attend the doctor each time they develop one.
If we suggest that they should start attending each time -to save 11.1 hours of illness- then we will immediately increase attendances with these very minor problems. This will divert time and energy from more important work and thus add significantly to the burdens our profession is suffering, making it harder for the significantly unwell to access our services -whilst achieving almost nothing.
This looks like research which should have been suppressed for the greater good!
Stephen Haywood
GP Northumberland

This article (1) recommends a major change to the management of sore throat in the UK; quite apart from the issue of the increased workload which this would create, I am not convinced that the possible small benefit justifies the risks. By my calculation, the most recent trial included in the review (2) had a NNT of 12 for patients with sore throat who were given dexamethasone being pain-free after 48 hours. That is not impressive.

The authors state that "A single dose of corticosteroids is unlikely to cause serious adverse events". However a recent US cohort study (3) showed that short courses of oral steroids are associated with a fivefold increase in the risk of sepsis in the following 30 days, as well as a threefold risk of VTE and double the risk of fractures. The courses used in this article were for a median dose of 20mg prednisolone and a median duration of 6 days (120mg in total). This new guideline recommends using 10mg of dexamethasone: a high dose of oral steroid, equivalent to 65 mg prednisolone (130mg in total if given for two days).

Certainly oral steroids should be considered as an option for the patient in severe pain, but I do not believe that the existing evidence justifies their routine prescription.

References
1. Aertgeerts Bert, Agoritsas Thomas, Siemieniuk Reed A C, Burgers Jako, Bekkering Geertruida E, Merglen Arnaud et al. Corticosteroids for sore throat: a clinical practice guideline BMJ 2017; 358 :j4090
2. Hayward GN, Hay AD, Moore MV, et al. Effect of oral dexamethasone without immediate antibiotics vs placebo on acute sore throat in adults: a randomized clinical trial. JAMA2017;358:1535-43
3. Waljee Akbar K, Rogers Mary A M, Lin Paul, Singal Amit G, Stein Joshua D, Marks Rory M et al. Short term use of oral corticosteroids and related harms among adults in the United States: population based cohort study BMJ 2017; 357 :j1415