Reply to Comment on ‘Impact of NICE guidance on tamoxifen prescribing in England 2011–2017: an interrupted time series analysis”

We thank Dr Hopcroft for taking the time to read and comment on our findings.¹ His letter suggests that our paper² was framed as a criticism of GPs. We do not believe that most people reading our paper would get that impression. This observational analysis of prescribing behaviour was not intended to be critical of GPs, nor blame them for inaction. We found that uptake of new guidance was slow, and discussed various possible contributing factors, including the influence of patient preference and referral to specialist care. We do not consider the impact of GPs initiating prescribing anywhere in the Discussion; we only added mention of GPs to the Introduction in response to a peer review comment requesting it.

We do, however, think it is somewhat relevant. We cite a previous paper by primary care researchers which reported variable knowledge among GPs of the NICE guideline.³ Indeed, that finding triggered our research: we wanted to see if there was variation in practice, after others reported variation in knowledge. We set that paper on variable GP knowledge in context, saying: “discussion will often be initiated in primary care, thus leaving room for variation in care driven by variation in primary care physicians’ knowledge”. This is indeed a possibility. GPs can act as an important facilitator in referring patients, and widening the use of chemoprevention. This seems uncontroversial to us.

Dr Hopcroft suggests that CCGs should use data to monitor GP prescribing and ensure that GPs are following NICE guidance: this is exactly what we recommend. Indeed, as discussed in the paper, we produce an openly accessible tool—OpenPrescribing.net—that does exactly this, with 50,000 unique users last year. Improvements in data monitoring, audit, and knowledge dissemination can only support GPs in staying current with guidelines across a wide range of clinical areas.

We agree that the guidance to only offer chemoprevention to presenting patients, rather than actively seeking out patients, may be another contributing factor to low rates of adoption. However, variation among practices’ prescribing data could still be driven by clinicians’ variable actions in response to patients presenting, as well as variation in spontaneous presentations. We are separately concerned by the prospect that variation in use of primary prevention will be driven by variation in patients’ choices to spontaneously present, which may in turn be a driver of inequality in health outcomes; in our view this is a worthy topic for research.

Competing interests

B.G. has received funding from the Health Foundation, the National Institute for Health Research School of Primary Care Research, the NIHR Biomedical Research Centre Oxford, the West of England Academic Health Sciences Network and NHS England for work on UK prescribing the data. H.C. is employed on these grants. B.G. has additionally received funding from the Laura and John Arnold Foundation, the Wellcome Trust, and the World Health Organisation to work on better use of the data in medicine; and receives personal income from speaking and writing for lay audiences on the misuse of science.