Machine Learning for Drug Overdose Surveillance

ABSTRACT

We describe two recently proposed machine learning approaches for discovering emerging trends in fatal accidental drug overdoses. The Gaussian Process Subset Scan (Herlands, McFowland, Wilson, & Neill, 2017 Neill, D. B. (2017). Multidimensional tensor scan for drug overdose surveillance. Journal of Public Health Informatics, 9(1), e20. doi:10.5210/ojphi.v9i1.7598[Crossref] , [Google Scholar]) enables early detection of emerging patterns in spatio-temporal data, accounting for both the complex, correlated nature of the data and the fact that detecting subtle patterns requires integration of information across multiple spatial areas and multiple time steps. We apply this approach to 17 years of county-aggregated data for monthly opioid overdose deaths in the New York City metropolitan area, showing clear advantages in the utility of discovered patterns as compared to typical anomaly detection approaches. To detect and characterize emerging overdose patterns that differentially affect a subpopulation of the data, including geographic, demographic, and behavioral patterns (e.g., which combinations of drugs are involved), we apply the Multidimensional Tensor Scan (Neill, 2017 Neill, D. B. (2017). Multidimensional tensor scan for drug overdose surveillance. Journal of Public Health Informatics, 9(1), e20. doi:10.5210/ojphi.v9i1.7598[Crossref] , [Google Scholar]) to 8 years of case-level overdose data from Allegheny County, Pennsylvania. We discover previously unidentified overdose patterns which reveal unusual demographic clusters, show impacts of drug legislation, and demonstrate potential for early detection and targeted intervention. These approaches to early detection of overdose patterns can inform prevention and response efforts, as well as understanding the effects of policy changes.     

When Father Steps Forward and Mother Steps Back: The Moderating Role of Simultaneity in Parents' Coparenting Behaviors in the Development of Anxiety in 4‐ to 30‐Month‐Olds

Infant negative affectivity predicts child anxiety. Coparenting might influence the development of anxiety by weakening this association in the case of supportive coparenting, or by strengthening this association in the case of undermining coparenting. Parents can display coparenting behaviors simultaneously (both parents being supportive or undermining), or divergently (only one parent being supportive or undermining). In our longitudinal study, we investigated whether coparenting moderated the relation between infant negative affectivity at 4 months and child anxiety symptoms 2 years later. Hundred‐sixteen couples dressed up their firstborn infants in a clothes‐changing task. We coded cooperative, mutual, neutral, and competitive coparenting behaviors. Both parents rated infant negative affectivity and child anxiety symptoms. Infant negative affectivity significantly predicted child anxiety. This association was moderated by parents' divergent cooperative coparenting: It was stronger when mothers were cooperative while fathers were neutral, and weaker when fathers were cooperative while mothers were neutral. When fathers step forward (i.e., being cooperative) and mothers step back (i.e., leaving space), they may protect their at‐risk child from developing anxiety.     

When the Lord builds the house: An autoethnographic account of working in an evangelical for-profit organizational culture

Religion is often perceived as a private topic, thus belonging outside the traditional conceptualization of a work environment. However, religion is a pervasive organizing framework and an understanding of its role in the workplace is needed from a communicative perspective. In the following autoethnography, I describe my experience working for a Christian for-profit homebuilder – starting with new employee orientation and moving through my final day at Local Homes. I focus on instances where tension emerges between expectations and reality as I learn about and interact within the overtly evangelical Christian organizational culture. The tension is illustrated through four vignettes that correspond to the traditional phases of socialization. Through writing this autoethnographic account, I attempt to move beyond my personal experience and situate my story within larger disciplinary and societal conversations about the socialization process in a religious organization. With this goal in mind, I conclude with methodological and theoretical implications.     

Self-Directed Gambling Changes: Trajectory of Problem Gambling Severity in Absence of Treatment

Most problem gamblers do not seek formal treatment, recovering on their own through cognitive re-appraisal or self-help strategies. Although barriers to treatment have been extensively studied, there is a paucity of research on self-directed changes in problem gambling and very few studies have examined these changes prospectively. The aim of this study was to examine the trajectory of gambling severity and behavior change over an 18-month period, among a sample of non-treatment seeking/attending problem gamblers recruited from the community (N?=?204) interested in quitting or reducing gambling. Separate mixed effects models revealed that in absence of formal treatment, significant reductions in gambling severity, frequency, and amount gambled could be observed over the course of a 6 to 9-month period and that changes experienced within the first 12 months were maintained for an extended 6 months. Problem gambling severity at baseline was significantly associated with changes in severity over time, such that participants with more severe gambling problems demonstrated greater reductions in their gambling severity over time. A total of 11.1% of participants gambled within a low-risk threshold at 18 months, although 28.7% of the sample reported consecutive gambling severity scores below problem levels for the duration of 1 year or longer. The findings suggest that among problem gamblers motivated to quit or reduce their gambling, significant self-directed changes in gambling severity can occur over a relatively short time. Additional prospective studies are needed to document the role of specific self-help tools or thought processes in exacting gambling changes.     

Unclaimed Prize Information Biases Perceptions of Winning in Scratch Card Gambling

Unclaimed prize information (i.e., the number of prizes still available to be won) is information commonly provided to scratch card gamblers. However, unless the number of tickets remaining to be purchased is also provided, this information is uninformative. Despite its lack of utility in assisting gamblers in choosing the most favourable type of scratch card to play, we hypothesized that unclaimed prize information would bias participants’ judgments within a scratch card gambling context. In Experiment 1 (N?=?201), we showed that participants are influenced by this information such that they felt more likely to win, were more excited to play, and preferred to hypothetically purchase more of the scratch card with the greatest number of unclaimed prizes. In Experiment 2 (N?=?201), we attempted to ameliorate this bias by providing participants with the number of tickets remaining to be purchased and equating the payback percentages of all three games. The bias, although attenuated, still persisted in these conditions. Finally, in Experiment 3 (N?=?200), we manipulated the hypothetical scratch cards such that games with the highest number of unclaimed prizes were the least favourable, and vice versa. As in Experiment 2, participants still favoured cards with greater numbers of unclaimed prizes. Possible mechanisms underlying this bias are discussed. In conclusion, across three experiments, we demonstrate that salient unclaimed prize information is capable of exerting a strong effect over judgments related to scratch card games.     

Implicit Valuation of the Near-Miss is Dependent on Outcome Context

Gambling studies have described a “near-miss effect” wherein the experience of almost winning increases gambling persistence. The near-miss has been proposed to inflate the value of preceding actions through its perceptual similarity to wins. We demonstrate here, however, that it acts as a conditioned stimulus to positively or negatively influence valuation, dependent on reward expectation and cognitive engagement. When subjects are asked to choose between two simulated slot machines, near-misses increase valuation of machines with a low payout rate, whereas they decrease valuation of high payout machines. This contextual effect impairs decisions and persists regardless of manipulations to outcome feedback or financial incentive provided for good performance. It is consistent with proposals that near-misses cause frustration when wins are expected, and we propose that it increases choice stochasticity and overrides avoidance of low-valued options. Intriguingly, the near-miss effect disappears when subjects are required to explicitly value machines by placing bets, rather than choosing between them. We propose that this task increases cognitive engagement and recruits participation of brain regions involved in cognitive processing, causing inhibition of otherwise dominant systems of decision-making. Our results reveal that only implicit, rather than explicit strategies of decision-making are affected by near-misses, and that the brain can fluidly shift between these strategies according to task demands.     

Barriers and Facilitators of Responding to Problem Gambling: Perspectives from Australian Mental Health Services

Despite high rates of comorbidity between problem gambling and mental health disorders, few studies have examined barriers or facilitators to the implementation of screening for problem gambling in mental health services. This exploratory qualitative study identified key themes associated with screening in mental health services. Semi-structured interviews were undertaken with 30 clinicians and managers from 11 mental health services in Victoria, Australia. Major themes and subthemes were identified using qualitative content analysis. Six themes emerged including competing priorities, importance of routine screening, access to appropriate screening tools, resources, patient responsiveness and workforce development. Barriers to screening included a focus on immediate risk as well as gambling being often considered as a longer-term concern. Clinicians perceived problem gambling as a relatively rare condition, but did acknowledge the need for brief screening. Facilitators to screening were changes to system processes, such as identification of an appropriate brief screening instrument, mandating its use as part of routine screening, as well as funded workforce development activities in the identification and management of problem gambling.     

Shyness,Preference for Solitude,and Adolescent Internalizing: The Roles of Maternal,Paternal, and Best‐Friend Support

The researchers examined differential outcomes related to two distinct motivations for withdrawal (preference for solitude and shyness) as well as the possibility that support from important others (mothers, fathers, and best friends) attenuate any such links. Adolescents (159 males, 171 females) reported on their motivations to withdraw, internalizing symptoms, and relationship quality in eighth grade, as well as their anxiety and depression in ninth grade. Using structural equation modeling, the authors found that maternal support weakened the association between shyness and internalizing problems; friend support weakened the association between preference for solitude and depression; and friend support strengthened the association between shyness and depression. Results suggest that shy adolescents may not derive the same benefits from supportive friendships as their typical peers.