Evaluating the translation of implementation science to clinical artificial intelligence: a bibliometric study of qualitative research

Introduction

Whilst a theoretical basis for implementation research is seen as advantageous, there is little clarity over if and how the application of theories, models or frameworks (TMF) impact implementation outcomes. Clinical artificial intelligence (AI) continues to receive multi-stakeholder interest and investment, yet a significant implementation gap remains. This bibliometric study aims to measure and characterize TMF application in qualitative clinical AI research to identify opportunities to improve research practice and its impact on clinical AI implementation.

Methods

Qualitative research of stakeholder perspectives on clinical AI published between January 2014 and October 2022 was systematically identified. Eligible studies were characterized by their publication type, clinical and geographical context, type of clinical AI studied, data collection method, participants and application of any TMF. Each TMF applied by eligible studies, its justification and mode of application was characterized.

Results

Of 202 eligible studies, 70 (34.7%) applied a TMF. There was an 8-fold increase in the number of publications between 2014 and 2022 but no significant increase in the proportion applying TMFs. Of the 50 TMFs applied, 40 (80%) were only applied once, with the Technology Acceptance Model applied most frequently (n = 9). Seven TMFs were novel contributions embedded within an eligible study. A minority of studies justified TMF application (n = 51,58.6%) and it was uncommon to discuss an alternative TMF or the limitations of the one selected (n = 11,12.6%). The most common way in which a TMF was applied in eligible studies was data analysis (n = 44,50.6%). Implementation guidelines or tools were explicitly referenced by 2 reports (1.0%).

Conclusion

TMFs have not been commonly applied in qualitative research of clinical AI. When TMFs have been applied there has been (i) little consensus on TMF selection (ii) limited description of selection rationale and (iii) lack of clarity over how TMFs inform research. We consider this to represent an opportunity to improve implementation science's translation to clinical AI research and clinical AI into practice by promoting the rigor and frequency of TMF application. We recommend that the finite resources of the implementation science community are diverted toward increasing accessibility and engagement with theory informed practices. The considered application of theories, models and frameworks (TMF) are thought to contribute to the impact of implementation science on the translation of innovations into real-world care. The frequency and nature of TMF use are yet to be described within digital health innovations, including the prominent field of clinical AI. A well-known implementation gap, coined as the "AI chasm" continues to limit the impact of clinical AI on real-world care. From this bibliometric study of the frequency and quality of TMF use within qualitative clinical AI research, we found that TMFs are usually not applied, their selection is highly varied between studies and there is not often a convincing rationale for their selection. Promoting the rigor and frequency of TMF use appears to present an opportunity to improve the translation of clinical AI into practice.

149 Reducing Complications and Improving Outcomes Following Total Hip Arthroplasty

Aim

Neck of femur (NOF) fractures are increasing, and the need to improve mobility and reduce complications whilst improving discharge time is an ongoing challenge. Hip fracture fixation and arthroplasty are essential in improving pain and mobility. NOF patients often undergo a fascia iliaca block (FIB) to improve outcomes. Guay J’s Cochrane systematic review confirms that FIBs reduce pain pre and postoperatively, decrease hospital-acquired pneumonia risk and improve mobility. Patients on anticoagulants are typically refused bedside blocks, however we are unsure if they truly impact adverse outcomes in FIBs for NOF patients.

Method

A retrospective observational study was undertaken to assess complication rates in 53 NOF patients who had a FIB at St Mary’s Hospital between 07/01/2020 and 26/06/2020. Patients whose NOF was the result of an inpatient fall were excluded. Pre-existing anticoagulant and antiplatelet use were noted, as were the A&E admission blood coagulation results. Local complications which may have arisen post-FIB were closely screened for in documentation until the patient’s discharge date.

Results

There were no local complications in all 53 patients. Interestingly, one patient had a documented thigh haematoma pre-nerve block, but did not have any evolution of the haematoma and achieved good pain relief from the block.

Conclusions

FIBs appear to be a safe and effective analgesic tool in the perioperative NOF patient and use has been reported to reduce early complications in those who undergo hip fracture treatment. We aim to implement these findings and increase the rate of nerve block procedures in NOF fractures.

Capture probability of released males of two Bactrocera species (Diptera: Tephritidae) in detection traps in California

The genus Bactrocera (Diptera: Tephritidae) includes approximately 70 polyphagous species that are major pests of fruit and vegetable crops. Most Bactrocera species have limited geographic distributions, but several species are invasive, and many countries operate continuous trapping programs to detect infestations. In the United States, California maintains approximately 25,000 traps (baited with male lures) specifically for Bactrocera detection distributed over an area of approximately 6,400 km2 (2,500 miles2) in the Los Angeles area. Although prior studies have used male lures to describe movement of Bactrocera males, they do not explicitly relate capture probability with fly distance from lure-baited traps; consequently, they do not address the relative effectiveness of male lures in detecting incipient populations of Bactrocera species. The objective of this study was to measure the distance-dependent capture probability of marked, released males of Bactrocera dorsalis (Hendel) and Bactrocera cucurbitae (Coquillett) (methyl eugenol- and cue lure-responding species, respectively) within the detection trapping grid operating in southern California. These data were then used to compute simple probability estimates for detecting populations of different sizes of the two species. Methyl eugenol was the more powerful attractant, and based on the mark-recapture data, we estimated that B. dorsalis populations with as few as approximately 50 males would always (>99.9%) be detected using the current trap density of five methyl eugenol-baited traps per 2.6 km2 (1 mile2). By contrast, we estimated that certain detection of B. cucurbitae populations would not occur until these contained approximately 350 males. The implications of the results for the California trapping system are discussed, and the findings are compared with mark-release-recapture data obtained for the same two species in Hawaii.