The impact of direct healthcare professional communication on prescribing practice in the UK hospital setting: an interrupted time series analysis.

BACKGROUND

Direct Healthcare Professional Communications (DHPCs) aim to quickly disseminate information to key healthcare professionals to inform practice and minimize patient harm. The Medicines and Healthcare products Regulatory Agency (MHRA) issues warnings and alerts to communicate safety information effectively in the UK.

OBJECTIVE

To investigate the impact of MHRA DHPCs on prescribing practice in the secondary-care setting, looking specifically at a drug-drug interaction-the concomitant use of clopidogrel and proton pump inhibitors (PPIs) [as omeprazole]-and a drug-disease contraindication-the use of conventional (typical) antipsychotics in dementia.

METHODS

The effects of the MHRA DHPCs were analysed using segmented binary logistic regression of interrupted time series. This allowed for the detection of any significant changes in prescribing practice occurring after the MHRA warnings were issued, whilst controlling for the baseline period.

RESULTS

Of the patients concomitantly prescribed clopidogrel and omeprazole on admission, the rate at which omeprazole was substituted for either another PPI (with the exception of esomeprazole), or for a histamine H2-antagonist showed a significant step-change increase after the DHPC was issued. The modelled rate increased from 5.1 % in the month directly before the intervention to 25.1 % in the following month (odds ratio [OR] 6.18; p < 0.001). However, the action taken in the switching of therapy was not always consistent with the advice from the current MHRA warning. The rate of typical antipsychotic prescribing in patients with dementia was declining significantly by 3.9 % per quarter prior to the DHPC being issued (OR 0.970; p = 0.035). No significant step-change was detected immediately after the DHPC (p = 0.962). However, the rate of decline increased significantly in the post-warning period to 12.3 % per quarter (OR 0.938; p = 0.006).

CONCLUSION

This study has shown that DHPCs issued by the MHRA as warnings are associated with changes in prescribing practices in secondary care. However, their impact is variable depending on the intervention described by the warning. A national initiative to ensure patient safety information is effectively translated into practice and the effect of the warning continues beyond the period of the issue would be beneficial.