Machine Learning Approaches to the Prediction of Osteoarthritis Phenotypes and Outcomes.

PURPOSE OF REVIEW: Osteoarthritis (OA) is a complex heterogeneous disease with no effective treatments. Artificial intelligence (AI) and its subfield machine learning (ML) can be applied to data from different sources to (1) assist clinicians and patients in decision making, based on machine-learned evidence, and (2) improve our understanding of pathophysiology and mechanisms underlying OA, providing new insights into disease management and prevention. The purpose of this review is to improve the ability of clinicians and OA researchers to understand the strengths and limitations of AI/ML methods in applications to OA research. RECENT FINDINGS: AI/ML can assist clinicians by prediction of OA incidence and progression and by providing tailored personalized treatment. These methods allow using multidimensional multi-source data to understand the nature of OA, to identify different OA phenotypes, and for biomarker discovery. We described the recent implementations of AI/ML in OA research and highlighted potential future directions and associated challenges.

Attitudes toward Tobacco, Alcohol, and Non-Alcoholic Beverage Advertisement Themes among Adolescent Boys.

BACKGROUND: Previous studies have examined what adolescents find appealing in tobacco and alcohol advertisements and how different themes in advertisements are used to manipulate consumer behaviors. Yet, we know little about the relationship between the themes portrayed in advertisements and youth attitudes towards those themes. OBJECTIVES: This study compared attitudes towards advertisements for different consumer products in a sample of urban and rural adolescent boys in order to examine how key marketing themes impact adolescent attitudes towards those advertisements. METHODS: Participants were 11- to 16-year-old boys (N = 1220) residing in either urban or rural Ohio Appalachian counties. Each participant viewed five print advertisements (one each for cigarettes, electronic cigarettes (e-cigarettes), smokeless tobacco (SLT), non-alcoholic beverages, and alcohol), presented in a random order, for eight seconds each. All advertisements had appeared in magazines that adolescent males commonly read. Attitudes towards each of the five advertisements were assessed. The advertisements were then coded for the presence of various themes, including social acceptance and masculinity. Analyses were conducted to determine associations between advertisement type and the attitude measure, and between the presence of a theme and the attitude measure. RESULTS: Overall, participants preferred non-tobacco advertisements to tobacco advertisements, rural participants had less positive attitudes and participants who had peers who used tobacco had more positive attitudes. Social acceptance and entertainment themes increased the appeal of SLT advertisements, and sex appeal increased the appeal of e-cigarette advertisements. Conclusions/Importance: Findings suggest that advertisements that promote the social nature of use in SLT advertisements may be of particular concern for their influence on adolescent boys.

Classifying Pseudogout Using Machine Learning Approaches With Electronic Health Record Data.

OBJECTIVE: Identifying pseudogout in large data sets is difficult due to its episodic nature and a lack of billing codes specific to this acute subtype of calcium pyrophosphate (CPP) deposition disease. The objective of this study was to evaluate a novel machine learning approach for classifying pseudogout using electronic health record (EHR) data. METHODS: We created an EHR data mart of patients with ≥1 relevant billing code or ≥2 natural language processing (NLP) mentions of pseudogout or chondrocalcinosis, 1991-2017. We selected 900 subjects for gold standard chart review for definite pseudogout (synovitis + synovial fluid CPP crystals), probable pseudogout (synovitis + chondrocalcinosis), or not pseudogout. We applied a topic modeling approach to identify definite/probable pseudogout. A combined algorithm included topic modeling plus manually reviewed CPP crystal results. We compared algorithm performance and cohorts identified by billing codes, the presence of CPP crystals, topic modeling, and a combined algorithm. RESULTS: Among 900 subjects, 123 (13.7%) had pseudogout by chart review (68 definite, 55 probable). Billing codes had a sensitivity of 65% and a positive predictive value (PPV) of 22% for pseudogout. The presence of CPP crystals had a sensitivity of 29% and a PPV of 92%. Without using CPP crystal results, topic modeling had a sensitivity of 29% and a PPV of 79%. The combined algorithm yielded a sensitivity of 42% and a PPV of 81%. The combined algorithm identified 50% more patients than the presence of CPP crystals; the latter captured a portion of definite pseudogout and missed probable pseudogout. CONCLUSION: For pseudogout, an episodic disease with no specific billing code, combining NLP, machine learning methods, and synovial fluid laboratory results yielded an algorithm that significantly boosted the PPV compared to billing codes.

Acute Calcium Pyrophosphate Crystal Arthritis Flare Rate and Risk Factors for Recurrence.

OBJECTIVE: Little is known about acute calcium pyrophosphate (CPP) crystal arthritis flare rates and risk factors for recurrence. We characterized flares and determined the rate and predictors of acute CPP crystal arthritis flares in an academic medical center cohort. METHODS: We performed a retrospective cohort study among a random sample of patients with acute CPP crystal arthritis identified in the Partners HealthCare electronic medical record, 1991-2017. Flare was defined as self-limited, acute-onset synovitis with synovial fluid CPP crystals and/or chondrocalcinosis, not better explained by another cause. We calculated incidence rates (IR) for acute CPP crystal arthritis flare among all subjects and by sex. We estimated HR for recurrent flare using univariate Cox models that accounted for within-person correlated data. RESULTS: We identified 70 patients with acute CPP crystal arthritis with a total of 111 flares. Recurrent flares occurred in 24% of patients; half of flares occurred in a previously unaffected joint. The acute CPP crystal arthritis flare rate was 11.4 per 100 person-years overall (95% CI 8.2-15.4), 14.2 in women (95% CI 9.6-0.1), and 7.1 in men (95% CI 3.4-13.0). Cancer (HR 2.98, 95% CI 1.33-6.68) and chronic kidney disease (HR 2.92, 95% CI 1.10-7.76) were associated with a higher rate for recurrent flare. CONCLUSION: Recurrent flares occurred in about one-fourth of patients with acute CPP crystal arthritis and often occurred in previously unaffected joints. The acute CPP crystal arthritis flare rate was twice as high in women as in men.

Voting by People with Mental Illness.

While voting laws trend toward universal suffrage, there are still some who encounter barriers in exercising the right to vote. Citizens with mental illness or cognitive and emotional impairments are especially vulnerable to exclusion from the political process, contributing to disenfranchisement. Facilitating the process for hospitalized patients to vote can increase their agency and amplify their voices and concerns. Through exercising their civic responsibility, psychiatric patients can have a hand in shaping a community in which they feel valued. In this article we will review the literature about voting, the current voting laws, and our lessons learned facilitating voting by proxy at Cambridge Hospital in the 2016 U.S. Presidential election, as well as the obstacles encountered. We will also propose methods to improve implementation of voting by hospitalized psychiatric patients for upcoming elections.