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.

Impact of the COVID-19 pandemic on the research activities of UK ophthalmologists.

BACKGROUND: The COVID-19 pandemic has impacted negatively on many areas of biomedical research and there is concern that academic recovery will take several years. This survey aimed to define the impact of the COVID-19 pandemic on UK ophthalmologists' research activities and understand the implications for recovery. METHODS: An online survey comprising multiple choice and free-text questions was designed, piloted and then distributed to Royal College of Ophthalmologists (RCOphth) members in January 2021. Respondent characteristics, research expectations and experiences through the pandemic were captured. Descriptive and comparative statistics were applied to quantitative data alongside content analysis of qualitative data. RESULTS: In total, 148 respondents (3.7% of RCOphth membership) comprised 46 trainees (31.1%), 97 consultants (65.5%) and 5 SAS doctors (3.4%); 54 had clinical-academic roles (36.5%) and 65/94 (69.1%) ophthalmologists with fully clinical posts identified as research-active. Of 114 research-active respondents, 104 (91.2%) reported an impact on their research from COVID-19; negative impacts included loss of research time (n = 69), research delays (n = 96) and funding shortfalls (n = 63). Content analysis identified five common themes; type of research activity, clinical demands, institutional challenges, COVID-19 alignment and work-life balance. CONCLUSIONS: UK ophthalmology research has been adversely impacted by the pandemic. A substantial proportion of UK ophthalmologists are research active, but 20.4% of those surveyed felt that the pandemic had made research less attractive. Strategic steps must be taken to nurture UK ophthalmologists' engagement with research, especially for those who currently do no research, if the profession is to align itself with the Government vision of 'Research for All'.

Microbial keratitis in corneal grafts: predisposing factors and outcomes.

PurposeTo identify the nature of microbial keratitis in corneal grafts and the clinical outcomes at a tertiary hospital in the United Kingdom.Patients and methodsA retrospective case series of microbial keratitis in corneal grafts at the Royal Victoria Infirmary, Newcastle upon Tyne over a 17-year period (1997-2014).ResultsA total of 759 consecutive corneal grafts were identified from the Cornea Transplantation database. Of these, 59 episodes of microbial keratitis occurred in 41 eyes of 41 patients (5.4%; 19 male, 46.3%). Median patient age was 73 years (SD=19.4 years). The most common indication for corneal transplantation was bullous keratopathy (11/41, 26.8%). There were 34/59 (57.6%) episodes of culture-positive graft keratitis; Streptococcus pneumoniae and Staphylococcus aureus were each isolated in 5/34 (14.7%) culture-positive episodes. In all, 35/59 (59.3%) episodes of microbial keratitis occurred in 22 previously failed grafts and 3 de novo graft failures. Gram-negative keratitis was more likely to cause reduced BCVA after (χ2-test, P=0.02). Median graft duration was 49.5 months (SD=43.7 months). Failed grafts were significantly older (median 69 vs 27 months, P=0.009).ConclusionThis represents the longest published follow-up data on microbial keratitis and is the only of its kind in the United Kingdom. The incidence of 5.4% is comparable to that within the developed world. Graft age was significantly associated with graft failure in microbial keratitis; the ongoing risk of microbial keratitis warrants providing patients with long-term open access to hospital eye services.

An 8-year longitudinal analysis of UK ophthalmic publication rates.

Purpose44.5% of abstracts presented at biomedical conferences are published. 26.5% of abstracts presented are basic science. The 2005 Walport Report reformed clinical academic training in the United Kingdom (UK) to promote trainee research. This study aims to analyse UK Ophthalmology research output following the reconstruction of clinical academic training.Patients and Methods1862 abstracts presented at The Royal College of Ophthalmologists' (RCOphth) Annual Congress from May 2005-May 2012 were examined using PubMed. Publication trends were analysed using SPSS v22 (IBM), using Spearman's rank coefficient and Mann-Whitney U test.Results44 (2.4%) abstracts were randomized controlled trials (RCTs), 88 (4.7%) basic science, and 231 (12.4%) oral presentations. 486 (26.6%) abstracts were published to a mean impact factor (IF) of 2.39 (95% CI 2.21-2.57). Mean time to publication from presentation was 15.17 (13.88-16.48) months, negatively correlating with IF (r=-0.149, P<0.003). Oral presentation (P<0.0001), RCTs (P=0.002), and basic science (P<0.0001) abstracts all made publication significantly more likely, with hazard ratios of 2.63 (2.13-2.24), 2.07 (1.3-3.2), and 1.92 (1.41-2.59), respectively. Higher IF was associated with oral presentation (3.4 vs 2.16, P<0.0001), basic science (3.57 vs 2.35, P<0.0001), and RCTs (4.78 vs 2.38, P=0.002). No significant change in publication rate was seen across the 8 years (P=0.61).ConclusionThe proportion of basic science and total abstracts published that are presented at the RCOphth is lower than that in other biomedical conferences. RCTs, basic science abstracts, and oral presentations are more likely to be published. There was no improvement in publication rates following the 2005 Walport Report.

Surgical outcome and patterns of recurrence for retroperitoneal sarcoma at a single centre.

INTRODUCTION: Retroperitoneal sarcoma is a surgically managed condition that can recur locally following macroscopically complete resection. Owing to the low incidence of the condition, advances in treatment are reported infrequently but complete compartmental resection and adjuvant or neoadjuvant radiotherapy are areas under investigation. Given the practical difficulty of randomised trials, observational data can highlight advantages from progressive treatment approaches. METHODS: A retrospective database of consecutive retroperitoneal sarcoma resections performed at a single referral centre between March 1997 and March 2013 was interrogated. Histological, radiological and clinical data were collected. Univariate and multivariate analyses for disease free and overall survival were performed to establish independent predictors of disease recurrence and patient survival. RESULTS: A total of 79 patients underwent 90 resections (63 primary). The mean five-year overall and disease free survival rates were 55.3% and 24.8% respectively. Higher patient age, high tumour grade, presence of extraretroperitoneal disease and invasive tumour phenotype were found to significantly predict survival following multivariate analysis. Half (50%) of the tumours displayed invasive behaviour on histopathology and 42% of locoregional recurrence was intraperitoneal. CONCLUSIONS: Retroperitoneal sarcoma is commonly an infiltrative tumour and often recurs outside of the retroperitoneum. These features limit the therapeutic impact of interventions that focus on gaining local control such as complete compartmental resection and radiotherapy. It seems likely that future advances in the management of this cancer will involve new systemic agents to treat this frequently systemic disease.