Practitioner and patient-targeted interventions to address excessive antibiotic use

By Dr Olga KostopoulouReader in Medical Decision Making and Professor Brendan DelaneyChair in Medical Informatics and Decision Making at Imperial College London 

Antibiotics Combatting antimicrobial resistance (AMR) is high on policy agendas internationally. One of the key means advocated is judicious antibiotic prescribing. Over 80% of all NHS antibiotic prescriptions are issued in primary care, where despite numerous campaigns, mandates and financial incentives, rates have fallen only slightly in the past year. Acute respiratory infections and associated complications, such as pneumonia, are the commonest justification for primary care antibiotic use, despite strong evidence of small to modest symptomatic benefits. GPs admit to prescribing because of diagnostic uncertainty, defensive practice, and real or perceived patient demand. Patients request antibiotics because of their perceived effectiveness, poor understanding of AMR, and lack of awareness of their susceptibility to complications of infections caused by their own resistant bacteria.

Digital interventions in the form of decision support systems (DSS) have been proposed as a solution to the problem of integrating complex evidence into practice. However, evidence for the effectiveness of digital interventions to support antimicrobial prescribing is limited. A recent meta-analysis found a lack of high-quality evaluation studies, all of which were in hospital settings, and no evidence of effect on patient mortality or length of stay (1). The limited effectiveness of this type of intervention was due to limited use by physicians but also the design and timing of the support. In primary care too, decision support for prescribing comes too late in the process of formulating a decision, after the GP has considered, negotiated, and reached agreement with the patient (2). Therefore, one requirement for the design of decision support is that it comes at an appropriate point in time, so that it can influence decision making rather than attempt to change the decision after it has been made. In our previous work on diagnosis, funded as part of an EU FP7 programme, we found that presenting GPs with a list of diagnoses to consider at the start of the consultation, before they started asking the patient questions and testing their hypotheses, improved diagnostic accuracy over control (3, 4). Similar approaches could be taken with antibiotic prescribing, by addressing the uncertainty that drives it from the start of the clinical encounter.

From the patient perspective, simple ‘nudge’ approaches are increasingly favoured by policy makers to tackle persistent behaviours that are seen as detrimental to the individual and ultimately the common good (5). Nudges consist of subtle changes in the environment, and co-opt people’s systematic cognitive biases, so that they intuitively gravitate towards the option considered to be more beneficial. They are non-coercive and non-incentivising, i.e. do not involve things like monetary benefits or mandates. A successful nudge in the area of antibiotic prescribing was a public pledge of US primary care physicians to avoid inappropriate antibiotic prescribing that took the form of a poster displayed on the wall of their clinic (6). Public pledges are also employed with patients: on the Public Health England website, people can pledge how they will act responsibly in a situation where they have to decide about antibiotics. The number of pledges made is displayed to communicate the social norm (what others do) and the injunctive norm (what one should do). However, the long-term effects of nudges have not been studied, impact over time may reduce, repeated presentations may become less effective or even backfire, while behaviour may return to pre-nudge levels once they are removed. It has been argued that enduring behaviour change involves ‘an identity-change process whereby people proactively choose to engage in behaviour that is perceived as identity-consistent and therefore seen as the right thing to do’ (7). It remains to be seen what patient-targeted interventions can achieve this.

References:
(1) Baysari MT, Lehnbom EC, Li L, et al. The effectiveness of information technology to improve antimicrobial prescribing in hospitals: A systematic review and meta-analysis. Int J Med Inform. 2016;92:15-34.
(2) Hayward J, Thomson F, Milne H, et al. “Too much, too late”: mixed methods multi-channel video recording study of computerized decision support systems and GP prescribing. JAMIA. 2013;20(e1):e76-e84.
(3) Kostopoulou O, Rosen A, Round T, Wright E, Douiri A, Delaney BC. Early diagnostic suggestions improve accuracy of GPs: a randomised controlled trial using computer-simulated patients. BJGP 2015; 65(630): e49-e54. http://dx.doi.org/10.3399/bjgp15X683161
(4) Kostopoulou O, Porat T, Corrigan D, Mahmoud S, Delaney BC. Supporting first impressions reduces diagnostic error: evidence from a high-fidelity simulation. BJGP. In Press.
(5) Sunstein C, Thaler R. Nudge: Improving Decisions About Health, Wealth and Happiness. Yale University Press; 2008
(6) Meeker D, Knight TK, Friedberg MW, et al. Nudging Guideline-Concordant Antibiotic Prescribing. JAMA Intern Med. 2014;174(3):425.
(7) Mols F, Haslam SA, Jetten J, Steffens NK. Why a nudge is not enough: A social identity critique of governance by stealth. Eur J Polit Res. 2015;54(1):81-98.

Advertisements
This entry was posted in Antimicrobial resistance and tagged , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s