Van Marrewijk et al raise a number of useful questions about our study in their eletter.1 There are no conflicts of interest among the GPs studied which would impact on reduction in self-monitoring of blood glucose that was demonstrated. No payment was made to them for taking part in this evaluation and no payment was used to incentivise reduction.
I think it highly unlikely that socioeconomic factors had any substantive impact on the greater reduction in the two intervention CCGs. All three CCGs share very similar socioeconomic profiles.
I agree that the assumption that the same reduction can be achieved nationally is just an assumption and we gave indicative costs were this to be achieved. Making the reduction may be a little more difficult because in these CCGs we have a ‘home advantage’ and strong track record of successful improvement programmes supported by IT and a University improvement group, which are not necessarily simply applicable across all CCGs.
Lastly the early decrease is in part due to extensive discussions that took place with stakeholders in the 2 years before formal implementation, which influenced prescribing behaviour. This included diabetes specialist nurses, hospital specialists, prescribing advisors, patients, pharmacists, CCG and practice staff, and GPs. Getting an initial local consensus is a key component of change and requires quite a lot of time, usually at least a year, to achieve a major change like this.
We do agree that a randomised trial would have been a better study but there are some practical difficulties in randomising CCGs that might need to be overcome in order to achieve this. Some imaginative thinking on design might be needed along with far greater resources than we had at our command. We had no special funding for this study.
- © British Journal of General Practice 2015
REFERENCE
- 1.↵