What predicts citation counts and translational impact in headache research? A machine learning analysis.

BACKGROUND: We aimed to develop the first machine learning models to predict citation counts and the translational impact, defined as inclusion in guidelines or policy documents, of headache research, and assess which factors are most predictive. METHODS: Bibliometric data and the titles, abstracts, and keywords from 8600 publications in three headache-oriented journals from their inception to 31 December 2017 were used. A series of machine learning models were implemented to predict three classes of 5-year citation count intervals (0-5, 6-14 and, >14 citations); and the translational impact of a publication. Models were evaluated out-of-sample with area under the receiver operating characteristics curve (AUC). RESULTS: The top performing gradient boosting model predicted correct citation count class with an out-of-sample AUC of 0.81. Bibliometric data such as page count, number of references, first and last author citation counts and h-index were among the most important predictors. Prediction of translational impact worked optimally when including both bibliometric data and information from the title, abstract and keywords, reaching an out-of-sample AUC of 0.71 for the top performing random forest model. CONCLUSION: Citation counts are best predicted by bibliometric data, while models incorporating both bibliometric data and publication content identifies the translational impact of headache research.

High-Intensity vs Low-Intensity Knowledge Translation Interventions for Surgeons and Their Association With Process and Outcome Measures Among Patients Undergoing Rectal Cancer Surgery.

Importance: Surgeon-directed knowledge translation (KT) interventions for rectal cancer surgery are designed to improve patient measures, such as rates of permanent colostomy and in-hospital mortality, and to improve survival. Objective: To evaluate the association of sustained, iterative, integrated KT rectal cancer surgery interventions directed at all surgeons with process and outcome measures among patients undergoing rectal cancer surgery in a geographic region. Design, Setting, and Participants: This quality improvement study used administrative data from patients who underwent rectal cancer surgery from April 1, 2004, to March 31, 2015, in 14 health regions in Ontario, Canada. Follow-up was completed on March 31, 2020. Exposures: Surgeons in 2 regions were offered intensive KT interventions, including annual workshops, audit and feedback sessions, and, in 1 of the 2 regions, operative demonstrations, from 2006 to 2012 (high-intensity KT group). Surgeons in the remaining 12 regions did not receive these interventions (low-intensity KT group). Main Outcomes and Measures: Among patients undergoing rectal cancer surgery, proportions of preoperative pelvic magnetic resonance imaging (MRI), preoperative radiotherapy, and type of surgery were evaluated, as were in-hospital mortality and overall survival. Logistic regression models with an interaction term between group and year were used to assess whether process measures and in-hospital mortality differed between groups over time. Results: A total of 15 683 patients were included in the analysis (10 052 [64.1%] male; mean [SD] age, 65.9 [12.1] years), of whom 3762 (24.0%) were in the high-intensity group (2459 [65.4%] male; mean [SD] age, 66.4 [12.0] years) and 11 921 (76.0%) were in the low-intensity KT group (7593 [63.7%] male; mean [SD] age, 65.7 [12.1] years). A total of 1624 patients (43.2%) in the high-intensity group and 4774 (40.0%) in the low-intensity KT group underwent preoperative MRI (P < .001); 1321 (35.1%) and 4424 (37.1%), respectively, received preoperative radiotherapy (P = .03); and 967 (25.7%) and 2365 (19.8%), respectively, received permanent stoma (P < .001). In-hospital mortality was 1.6% (59 deaths) in the high-intensity KT group and 2.2% (258 deaths) in the low-intensity KT group (P = .02). Differences remained significant in multivariable models only for permanent stoma (odds ratio [OR], 1.67; 95% CI, 1.24-2.24; P < .001) and in-hospital mortality (OR, 0.67; 95% CI, 0.51-0.87; P = .003). In both groups over time, significant increases in the proportion of patients undergoing preoperative MRI (from 6.3% to 67.1%) and preoperative radiotherapy (from 16.5% to 44.7%) occurred, but there were no significant changes for permanent stoma (25.4% to 25.3% in the high-intensity group and 20.0% to 18.3% in the low-intensity group) and in-hospital mortality (0.8% to 0.8% in the high-intensity group and 2.2% to 1.8% in the low-intensity group). Time trends were similar between groups for measures that did or did not change over time. Patient overall survival was similar between groups (hazard ratio, 1.00; 95% CI, 0.90-1.11; P = .99). Conclusions and Relevance: In this quality improvement study, between-group differences were found in only 2 measures (permanent stoma and in-hospital mortality), but these differences were stable over time. High-intensity KT group interventions were not associated with improved patient measures and outcomes. Proper evaluation of KT or quality improvement interventions may help avoid opportunity costs associated with ineffective strategies.

Patient engagement in preclinical laboratory research: A scoping review.

BACKGROUND: 'Patient engagement' involves meaningful collaboration between researchers and 'patient partners' to co-create research. It helps ensure that research being conducted is relevant to its ultimate end-users. Although patient engagement within clinical research has been well documented, the prevalence and effects of patient engagement in translational preclinical laboratory research remain unclear. The aim of this scoping review is to present current patient engagement activities reported in preclinical laboratory research. METHODS: MEDLINE, Embase, and grey literature were systematically searched from inception to April 2021. Studies that described or investigated patient engagement in preclinical laboratory research were included. Patient engagement activities where patients (i.e. patients, family members, caregivers or community members) provided input, or consultation on at least one element of the research process were eligible for inclusion. Study characteristics and outcomes were extracted and organized thematically. FINDINGS: 32 reports were included (30 primary studies, 1 narrative review, and 1 researcher guide). Most studies engaged patients at the education or priority setting stages (n=26). The most frequently reported benefit of patient engagement was 'providing a mutual learning opportunity'. Reported barriers to patient engagement reflected concerns around 'differences in knowledge and research experience' and how this may challenge communication and limit meaningful collaboration. INTERPRETATION: Patient engagement is feasible and beneficial for preclinical laboratory research. Future work should focus on assessing the impacts of patient engagement in this area of research. FUNDING: None.