Gamified antimicrobial decision support app (GADSA) changes antibiotics prescription behaviour in surgeons in Nigeria: a hospital-based pilot study

AIMS

Surgical Antibiotic Prophylaxis (SAP) in Nigeria is often not evidence based. The aim of this study is to test if the GADSA application can change prescription behaviour of surgeons in Nigeria. In addition, the study aims to identify AMS strategies and policies for the future.

METHODS

The GADSA gamified decision support app uses WHO and Sanford prescribing guidelines to deliver real-time persuasive technology feedback to surgeons through an interactive mentor. The app can advise on whether clinician's decisions align with SAP recommendations and provides the opportunity for clinicians to make adjustments. Twenty surgeons actively participated in a 6-month pilot study in three hospitals in Nigeria. The surgeons determined the risk of infection of a surgical procedure, and the need, type and duration of SAP. The study used a longitudinal approach to test whether the GADSA app significantly changed prescribing behaviour of participating surgeons by analysing the reported prescription decisions within the app.

RESULTS

321 SAP prescriptions were recorded. Concerning the surgical risk decision, 12% of surgeons changed their decision to be in line with guidelines after app feedback (p < 0.001) and 10% of surgeons changed their decision about the need for SAP (p = 0.0035) to align with guidelines. The change in decision making for SAP use in terms of "type" and "duration" to align with guidelines was similar with 6% and 5% respectively (both p-values < 0.001).

CONCLUSION

This study suggests that the GADSA app, with its game based and feedback feature, could significantly change prescribing behaviour at the point of care in an African setting, which could help tackle the global challenge of antibiotic resistance.

OUP accepted manuscript

Background

In Nigeria, the prescription of surgical antibiotic prophylaxis for prevention of surgical site infection tends to be driven by local policy rather than by published guidelines (e.g. WHO and Sanford).

Objectives

To triangulate three datasets and understand key barriers to implementation using a behavioural science framework.

Methods

Surgeons (N = 94) from three teaching hospitals in Nigeria participated in an online survey and in focus group discussions about barriers to implementation. The theoretical domains framework (TDF) was used to structure question items and interview schedules. A subgroup (N = 20) piloted a gamified decision support app over the course of 6 months and reported barriers at the point of care.

Results

Knowledge of guidelines and intention to implement them in practice was high. Key barriers to implementation were related to environmental context and resources and concern over potential consequences of implementing recommendations within the Nigerian context applicable for similar settings in low-to-middle-income countries.

Conclusions

The environmental context and limited resource setting of Nigerian hospitals currently presents a significant barrier to implementation of WHO and Sanford guidelines. Research and data collected from the local context must directly inform the writing of future international guidelines to increase rates of implementation.