Applying Machine Learning Techniques to Implementation Science.

Machine learning (ML) approaches could expand the usefulness and application of implementation science methods in clinical medicine and public health settings. The aim of this viewpoint is to introduce a roadmap for applying ML techniques to address implementation science questions, such as predicting what will work best, for whom, under what circumstances, and with what predicted level of support, and what and when adaptation or deimplementation are needed. We describe how ML approaches could be used and discuss challenges that implementation scientists and methodologists will need to consider when using ML throughout the stages of implementation.

Storylines of family medicine V: ways of thinking-honing the therapeutic self.

Storylines of Family Medicine is a 12-part series of thematically linked essays with accompanying illustrations that explore the many dimensions of family medicine, as interpreted by individual family physicians and medical educators in the USA and elsewhere around the world. In 'V: ways of thinking-honing the therapeutic self', authors present the following sections: 'Reflective practice in action', 'The doctor as drug-Balint groups', 'Cultivating compassion', 'Towards a humanistic approach to doctoring', 'Intimacy in family medicine', 'The many faces of suffering', 'Transcending suffering' and 'The power of listening to stories.' May readers feel a deeper sense of their own therapeutic agency by reflecting on these essays.

Impact of Healthcare Location Concordance on Receipt of Preventive Care Among Children Whose Parents have a Substance Use and/or Mental Health Diagnosis.

AIMS: Children of parents with substance use and/or other mental health (SU/MH) diagnoses are at increased risk for health problems. It is unknown whether these children benefit from receiving primary care at the same clinic as their parents. Thus, among children of parents with >1 SU/MH diagnosis, we examined the association of parent-child clinic concordance with rates of well-child checks (WCCs) and childhood vaccinations. DESIGN: Retrospective cohort study using electronic health record (EHR) data from the OCHIN network of community health organizations (CHOs), 2010-2018. Setting: 280 CHOs across 17 states. PARTICIPANTS/CASES: 41,413 parents with >1 SU/MH diagnosis, linked to 65,417 children aged 0 to 17 years, each with >1 visit to an OCHIN clinic during the study period. MEASUREMENTS: Dependent variables: rates of WCCs during (1) the first 15 months of life, and (2) ages 3 to 17 years; vaccine completeness (3) by the age of 2, and (4) before the age of 18. Estimates were attained using generalized estimating equations Poisson or logistic regression. FINDINGS: Among children utilizing the same clinic as their parent versus children using a different clinic (reference group), we observed greater WCC rates in the first 15 months of life [adjusted rate ratio (aRR) = 1.06; 95% confidence interval (CI) = 1.02-1.10]; no difference in WCC rates in ages 3 to 17; higher odds for vaccine completion before age 2 [adjusted odds ratio (aOR) = 1.12; 95% CI = 1.03-1.21]; and lower odds for vaccine completion before age 18 (aOR = 0.88; 95% CI = 0.81-0.95). CONCLUSION: Among children whose parents have at least one SU/MH diagnosis, parent-child clinic concordance was associated with greater rates of WCCs and higher odds of completed vaccinations for children in the youngest age groups, but not the older children. This suggests the need for greater emphasis on family-oriented healthcare for young children of parents with SU/MH diagnoses; this may be less important for older children.

A Scoping Literature Review on Evidence-Based Strategies to Increase Cervical Cancer Screening.

Previous reviews of strategies to increase cervical cancer screening are more than 10 years old, the U.S. continues to fall short of the Healthy People 2030 cervical cancer screening goal, and guidelines were revised in 2018, therefore an updated review of the existing literature is needed. We conducted a scoping review using electronic databases PubMed, Scopus, and Ovid Medline that included publication dates between 2012 and 2021 to answer the question, "Which strategies implemented in U.S. primary care settings have been most successful in increasing rates of cervical cancer screening since the 2012 US Preventative Services Task Force cervical cancer screening guidelines were published?" We mapped findings to pre-specified implementation strategy categories. After initially identifying 399 articles, we excluded 350 due to duplicates or not meeting review criteria, leaving 49 articles for full review. We excluded 37 of these during full-text review and identified 2 additional articles from the manual search of reference lists for a total of 14 studies for abstraction. Eleven articles reported on strategies resulting in increased cervical cancer screening, and 3 did not. Clinic workflow re-design strategies showed the greatest promise in improving cervical cancer screening rates, education strategies for patients had mixed results, and quality management strategies were not effective. These findings suggest clinical workflow re-structures and patient education strategies can increase cervical cancer screening in primary care settings. Results are particularly important in settings that care for underserved populations, as these settings may need additional implementation strategies to decrease cervical cancer screening disparities.