Lexicon-Based Sentiment Analysis in Behavioral Research.

A complete science of human behavior requires a comprehensive account of the verbal behavior those humans exhibit. Existing behavioral theories of such verbal behavior have produced compelling insight into language's underlying function, but the expansive program of research those theories deserve has unfortunately been slow to develop. We argue that the status quo's manually implemented and study-specific coding systems are too resource intensive to be worthwhile for most behavior analysts. These high input costs in turn discourage research on verbal behavior overall. We propose lexicon-based sentiment analysis as a more modern and efficient approach to the study of human verbal products, especially naturally occurring ones (e.g., psychotherapy transcripts, social media posts). In the present discussion, we introduce the reader to principles of sentiment analysis, highlighting its usefulness as a behavior analytic tool for the study of verbal behavior. We conclude with an outline of approaches for handling some of the more complex forms of speech, like negation, sarcasm, and speculation. The appendix also provides a worked example of how sentiment analysis could be applied to existing questions in behavior analysis, complete with code that readers can incorporate into their own work.

Social network structure as a suicide prevention target.

PURPOSE: The structure of relationships in a social network affects the suicide risk of the people embedded within it. Although current interventions often modify the social perceptions (e.g., perceived support and sense of belonging) for people at elevated risk, few seek to directly modify the structure of their surrounding social networks. We show social network structure is a worthwhile intervention target in its own right. METHODS: A simple model illustrates the potential of interventions to modify social structure. The effect of these basic structural interventions on suicide risk is simulated and evaluated. Its results are briefly compared to emerging empirical findings for real network interventions. RESULTS: Even an intentionally simplified intervention on social network structure (i.e., random addition of social connections) is likely to be both effective and safe. Specifically, this illustrative intervention had a high probability of reducing the overall suicide risk, without increasing the risk of those who were healthy at baseline. It also frequently resolved stable, high-risk clusters of people at elevated risk. These illustrative results are generally consistent with emerging evidence from real social network interventions for suicide. CONCLUSION: Social network structure is a neglected, but valuable intervention target for suicide prevention.

Interpersonal stress and proinflammatory activity in emerging adults with a history of suicide risk: A pilot study.

Interpersonal stress during adolescence and young adulthood can threaten healthy developmental trajectories. A "primed" proinflammatory response to acute stress may serve as an underlying process that results in negative outcomes for youth. The present pilot study examined the relation between interpersonal stress and two proinflammatory cytokines in a sample of 42 university-recruited emerging adults with recent suicidal thoughts and behaviors. Participants completed self-report measures of mood, suicidal thoughts and behaviors, recent peer-related stressors, and interpersonal sensitivity. They also participated in an acute laboratory social stress task and provided three saliva samples to measure their proinflammatory responses (IL-6 and TNF-α) to the stressor. Participants reported significant increases in sadness and exclusion, and significant decreases in inclusion, following task participation. Importantly, no participants reported an increase in or onset of suicidal thoughts. No significant associations between interpersonal stress and proinflammatory cytokines were found. Changes in affect during the task coupled with lack of increased suicidal thoughts indicate it is acceptable to use this exclusion and rejection paradigm with this population, with proper debriefing and positive mood induction procedures. Given all other nonsignificant associations, future research considerations are discussed, including impact of COVID-19 on task potency and incorporation of multiple stress response systems.

Examination of proinflammatory activity as a moderator of the relation between momentary interpersonal stress and suicidal ideation.

INTRODUCTION: Peer-related interpersonal stress can increase risk for suicidal thoughts among adolescents and young adults. However, not all individuals who undergo peer-related interpersonal stressors experience suicidal thoughts. Heightened proinflammatory activity is one factor that may amplify the relation between interpersonal stress and suicidal thinking. METHODS: This pilot study examined the relation between interpersonal stress and suicidal ideation in real time, as well as whether proinflammatory cytokine (IL-6 and TNF-α) activity across a laboratory social stressor moderated this association in a sample of 42 emerging adults with recent suicidal ideation. Participants completed 28 days of 6×/daily ecological momentary assessment that assessed for suicidal ideation (presence vs. absence, ideation intensity), occurrence of negative peer events, and feelings of exclusion. RESULTS: There was a trend for within-person increases in feelings of exclusion to be associated with increases in concurrent suicidal ideation intensity. Additionally, within-person increases in negative peer events were associated with increased odds of subsequent suicidal ideation among individuals with very low IL-6 activity. However, this finding is considered preliminary. CONCLUSION: Interventions targeting perceptions of exclusion and increasing social support may be of benefit. However, findings require replication in larger samples, and thus must be interpreted with caution.

Impact of Sources of Strength on adolescent suicide deaths across three randomized trials.

Universal interventions are key to reducing youth suicide rates, yet no universal intervention has demonstrated reduction in suicide mortality through an RCT. This study pooled three cluster-RCTs of Sources of Strength (n=78 high schools), a universal social network-informed intervention. In each trial, matched pairs of schools were assigned to immediate intervention or wait-list. Six schools were assigned without a pair due to logistical constraints. During the study period, no suicides occurred in intervention schools vs four in control schools, that is, suicide rates of 0 vs. 20.86/100,000, respectively. Results varied across statistical tests of impact. A state-level exact test pooling all available schools showed fewer suicides in intervention vs. control schools (p=0.047); whereas a stricter test involving only schools with a randomised pair found no difference (p=0.150). Results suggest that identifying mortality-reducing interventions will require commitment to new public-health designs optimised for population-level interventions, including adaptive roll-out trials.

The feasibility and acceptability of using smartphones to assess suicide risk among Spanish-speaking adult outpatients.

OBJECTIVE: Hispanic/Latino adults, particularly Spanish speakers, are underserved in mental healthcare, and little is known about the day-to-day variation in their suicide risk. Smartphones have the potential to overcome geographical and linguistic barriers to mental health assessment and intervention. The purpose of the current study was to examine (a) the feasibility/acceptability of smartphone-based ecological momentary assessment (EMA) to assess suicide ideation and suicide risk factors among high-risk Spanish-speaking adults and (b) the degree of within-person variation of suicide ideation and suicide risk in this population. METHOD: Sixteen primary Spanish-speaking psychiatry outpatients completed EMA measures of suicide ideation and suicide risk factors four times a day for 14 days. RESULTS: A majority of participants consented to active and passive remote assessments and reported the acceptability of study procedures. Adherence to EMA was high and not associated with symptom severity. EMA instances completed were not associated with symptom severity at follow-up. Average point-to-point variability in suicide ideation and risk factors were moderate to high, respectively. EMA captured more dramatic changes than standard baseline and follow-up assessments. CONCLUSIONS: Results provide preliminary support for the feasibility and acceptability of using smartphones to assess suicide risk in a real-time and real-world setting among high-risk Spanish-speaking adults.

Wingman-Connect Program increases social integration for Air Force personnel at elevated suicide risk: Social network analysis of a cluster RCT.

U.S. military suicides are increasing and disrupted relationships frequently precede them. Group-level interventions are needed that reduce future suicide vulnerability among healthy members and also ameliorate risk among those already suicidal. We examined whether our Wingman-Connect Program (W-CP) strengthened Air Force relationship networks and socially integrated at-risk members. Air Force personnel classes in training were randomized to W-CP or active control (cluster RCT), followed up at 1 and 6 months (94% and 84% retention). Data were collected in 2017-2019 and analyzed in 2020-2021. Participants were 1485 male and female Airmen in 215 technical training classes. W-CP training involved strengthening group bonds, skills for managing career and personal stressors, and diffusion of healthy norms. Active control was stress management training. Primary outcomes were social network metrics based on Airmen nominations of valued classmates after 1 month. Baseline CAT-SS >34 defined elevated suicide risk. W-CP increased social network integration, with largest impact for Airmen already at elevated suicide risk (n = 114, 7.7%). For elevated risk Airmen, W-CP improved all network integration metrics, including 53% average gain in valued connection nominations received from other Airmen (RR = 1.53, 95% CI = 1.12, 2.08) and eliminated isolation. No elevated risk Airmen in W-CP were isolates with no valued connections after 1-month vs. 10% among controls (P < .035). In contrast to at-risk controls, at-risk W-CP Airmen increased connections after intervention. W-CP's effect on a key indicator, ≥2 connections, was still greater 2-4 months after classes disbanded (6-months). Wingman-Connect Program built enhanced suicide protection into unit relationship networks and counteracted standard drift towards disconnection for at-risk Airmen, despite no explicit content targeting connections specifically to at-risk Airmen. Findings support a growing case for the unique contribution of group-level interventions to improve social health of broader military populations while also ameliorating risk among individuals already at elevated suicide risk.

The #MeToo Movement and Perceptions of Sexual Assault: College Students' Recognition of Sexual Assault Experiences Over Time.

OBJECTIVE: On October 15, 2017, Alyssa Milano encouraged anyone who had been sexually harassed or assaulted to respond on Twitter with the phrase, #MeToo. Millions responded and a cultural reckoning ensued. Anecdotally, the #MeToo movement appears to have affected survivors' acceptance and acknowledgment of their own sexual assault experiences, but empirical evidence is lacking. To address this gap, the aim of this study was to examine associations between behavioral and labeled reports of sexual assault and time since the #MeToo movement began. METHODS: Participants were 2,566 college students who completed a sexual assault survey over the course of three years, overlapping with the onset of the #MeToo movement. RESULTS: Regarding our hypothesis that the prevalence of sexual assault - indicated by standardized behaviorally specific questions - would be relatively constant over time after controlling for demographics, a Bayesian logistic regression model yielded inconclusive results. However, among the 596 students who endorsed behaviorally specific screeners for sexual assault, a Bayesian linear regression model revealed that, after controlling for demographics and characteristics of the assault, participants were increasingly likely to label the experience a "sexual assault" with more time post #MeToo. CONCLUSIONS: Overall, findings revealed no evidence for or against changes in prevalence of sexual assault, but suggested there were associations between the #MeToo movement and greater recognition of past unwanted sexual experiences as "sexual assault" over time. These findings highlight the importance of considering social context in research examining sexual assault survivors' cognitions.

Temporal sequences of suicidal and nonsuicidal self-injurious thoughts and behaviors among inpatient and community-residing military veterans.

BACKGROUND: Suicidal and nonsuicidal self-injurious thoughts and behaviors (SITBs) are major health concerns among military veterans yet little is known about the temporal relations among these outcomes. This study examined the temporal relations between suicidal and nonsuicidal SITBs among higher-risk veterans. Specifically, we identified when SITBs emerged and evaluated the role of nonsuicidal self-injury (NSSI) in the medical lethality of suicide attempts (SA), relative risk, and survival time of suicidal SITBs (i.e., suicide ideation [SI], suicide plan, SA). METHOD: Cross-sectional data were collected from two samples examining suicide risk among veterans receiving inpatient psychiatric care (n = 157) and community-residing veterans with current depression and/or past month SI (n = 200). Participants completed an interview to assess SITBs. RESULTS: SITBs emerged between ages 14-28 years with behaviors emerging, on average, earlier among inpatient veterans. The time lag between SITBs was not significantly different between groups. Inpatient veterans had a significantly shorter time lag from SI to SA. NSSI history predicted an increase in relative risk for all suicidal SITBs and shorter survival time. There was no association between NSSI history and medical lethality of the most serious SA for both groups. LIMITATIONS: Limitations included use of cross-sectional, retrospective self-report with age-of-onset endorsed in years and not all SITBs were assessed (e.g., passive SI). CONCLUSIONS: Veterans with a NSSI history are at high risk for suicidal SITBs and have a shorter survival time. Results showed thoughts (i.e., NSSI thoughts, SI) emerged before behavior (i.e., NSSI, SA) and NSSI emerged before SA.

A gentle tutorial on power simulations in suicide research: Worked examples in R.

INTRODUCTION: Power analysis is critical for both planning future research samples and evaluating the reasonability of answers produced by pre-existing and fixed samples. Unfortunately, the irregularity of suicide-related data and the need for increasingly complex models in suicide research can make traditional power formulas inaccurate or even unusable. Ignoring these common problems risks both over- and under-recruiting, as well as obscuring the true quality of the results (up and down) to future reviewers and readers. METHOD: A better option is to use Monte Carlo power simulations. RESULTS: These techniques produce answers that are equivalent to traditional power formulas when traditional assumptions are met, but produce more accurate results in the common case when those assumptions are violated. CONCLUSION: What follows is a tutorial on how suicide researchers can conduct such simulations. It begins by building the reader's intuition for why simulations work, followed by two worked examples in R. Discussion also includes guidelines for conducting and reporting simulations, along with answers to frequently asked questions. Appendices provide code examples researchers can model and adapt to their own simulations as needed.

Using categorical data analyses in suicide research: Considering clinical utility and practicality.

OBJECTIVE: Categorical data analysis is relevant to suicide risk and prevention research that focuses on discrete outcomes (e.g., suicide attempt status). Unfortunately, results from these analyses are often misinterpreted and not presented in a clinically tangible manner. We aimed to address these issues and highlight the relevance and utility of categorical methods in suicide research and clinical assessment. Additionally, we introduce relevant basic machine learning methods concepts and address the distinct utility of the current methods. METHOD: We review relevant background concepts and pertinent issues with references to helpful resources. We also provide non-technical descriptions and tutorials of how to convey categorical statistical results (logistic regression, receiver operating characteristic [ROC] curves, area under the curve [AUC] statistics, clinical cutoff scores) for clinical context and more intuitive use. RESULTS: We provide comprehensive examples, using simulated data, and interpret results. We also note important considerations for conducting and interpreting these analyses. We provide a walk-through demonstrating how to convert logistic regression estimates into predicted probability values, which is accompanied by Appendices demonstrating how to produce publication-ready figures in R and Microsoft Excel. CONCLUSION: Improving the translation of statistical estimates to practical, clinically tangible information may narrow the divide between research and clinical practice.

Causal inference in suicide research: When you should (and should not!) control for extraneous variables.

OBJECTIVE: Although causal inference is often straightforward in experimental contexts, few research questions in suicide are amenable to experimental manipulation and randomized control. Instead, suicide prevention specialists must rely on observational data and statistical control of confounding variables to make effective causal inferences. We provide a brief summary of recent covariate practice and a tutorial on casual inference tools for covariate selection in suicide research. METHOD: We provide an introduction to modern causal inference tools, suggestions for statistical control selection, and demonstrations using simulated data. RESULTS: Statistical controls are often mistakenly selected due to their significant correlation with other study variables, their consistency with previous research, or no explicit reason at all. We clarify what it means to control for a variable and when controlling for the wrong covariates systematically distorts results. We describe directed acyclic graphs (DAGs) and tools for identifying the right choice of covariates. Finally, we provide four best practices for integrating causal inference tools in future studies. CONCLUSION: The use of causal model tools, such as DAGs, allows researchers to carefully and thoughtfully select statistical controls and avoid presenting distorted findings; however, limitations of this approach are discussed.

Further application of the generalized matching law to multialternative sports contexts.

Cox et al. (2017) successfully applied the multialternative version of the generalized matching law (GML) to pitch selection among a sample of Major League Baseball (MLB) pitchers. The purpose of the present study was to replicate and extend these findings by fitting the multialternative GML to pitch data among a sample of MLB pitchers with varying levels of success in the major leagues. We also examined how matching parameters changed as a function of novel antecedent game contexts such as the infield shift, game location, and number of times the pitcher faced the batters in the batting order. These results replicate the findings from Cox et al. and suggest the multialternative GML is a robust descriptor of pitch selection among MLB pitchers. Together, these findings further extend the generality of the multialternative GML to naturalistic, nonlaboratory environments.

Posttraumatic Stress Disorder and Suicidal Ideation: The Moderating Effect of Posttraumatic Cognitions.

Although individuals with posttraumatic stress disorder (PTSD) are at an increased risk for suicidal ideation (SI), it is unclear what factors might influence this association. Investigators have hypothesized that posttraumatic cognitions (PTCs), such as self-blame (SB) or negative cognitions about the self (NCAS) or world (NCAW), would play a role, but this has not been investigated empirically. Accordingly, we evaluated a model in which the association between PTSD symptoms and SI was moderated by PTCs in a sample of trauma-exposed undergraduate students (N = 410). To identify the specific source of this hypothesized moderation effect, we ran the moderation model separately for PTSD total severity, PTSD total severity without the cognition-related items, and each of four DSM PTSD symptom clusters in combination with each of three types of PTCs (i.e., NCAS, NCAW, SB), accounting for quadratic effects. The results revealed that NCAW moderated the positive association between all six of the PTSD variables and SI, f2 s < .01 to .04. Analyses of simple slopes generally revealed strong positive associations between PTSD symptoms with SI at high levels of NCAW, no associations at moderate levels, and negative associations at low levels. We also found one statistically significant quadratic effect when examining avoidance and NCAW. In contrast, neither NCAS nor SB emerged as a significant moderator in any of our regression models. These findings highlight the importance of addressing PTCs-particularly NCAW-in trauma survivors.

Network Analysis of Posttraumatic Stress Disorder Symptoms in a National Sample of U.S. Adults: Implications for the Phenotype and the ICD-11 Model of PTSD.

The phenotype for posttraumatic stress disorder (PTSD) in the fifth edition of the Diagnostic and Statistical Manual of Mental Diseases (DSM-5) includes 20 symptoms in four clusters. In contrast, the PTSD model in the 11th revision of the International Classification of Diseases (ICD-11) includes six symptoms in three clusters. Whether those six symptoms are, in fact, the most central symptoms of the PTSD phenotype remains an open question. In a previous network analysis of DSM-5 PTSD symptoms, Mitchell and colleagues (2017) reported limited overlap between central PTSD symptoms and those in the ICD-11 model in a national sample of U.S. veterans. The present study sought to replicate and extend upon these findings in a large national sample of U.S. adults (N = 2,953). Centrality statistics from both a replication sample (i.e., participants with DSM-5 PTSD, n = 173) and an extension sample (i.e., participants who had been exposed to potentially traumatic events, n = 2,468) were moderately strongly convergent with the findings reported by Mitchell et al., rs = .54-.73. Additionally, only three of the six most central symptoms in both the replication and extension samples overlapped with the ICD-11 model, indicating that the ICD-11 model (a) failed to include network-central symptoms of the PTSD phenotype and (b) included extra symptoms that were not network-central. Several symptoms from the DSM-5 Criterion D cluster (negative alterations in cognition and mood) that were excluded in ICD-11 were found to be among the most central PTSD symptoms.

Application of the generalized matching law to chess openings: A gambit analysis.

During the opening moves of a chess game, a player (typically White) may offer a number of gambits, which involve sacrificing a chess piece for an opponent for capture to achieve long-term positional advantages. One of the most popular gambits is called the Queen's Gambit and involves White offering a pawn to Black, which will open a lane for White's Queen if accepted by Black. In the present study, the generalized matching law (GML) was applied to chess openings involving the Queen's Gambit using over 71,000 archived chess games. Overall, chess players' opening moves involving the Queen's Gambit exhibited orderly matching as predicted by the GML, and the GML accounted for more variance in players' chess decision making as their relative playing experience increased. This study provides support for the generality of the GML and its application to complex operant behavior outside of laboratory contexts.

Assortativity of suicide-related posting on social media.

Since the turn of the century, interdisciplinary research on networks-their formation, structure, and influence-has advanced so rapidly, it is now a science unto itself, offering new and powerful quantitative tools for studying human behavior, whose potential psychologists are just beginning to glimpse. Among these tools is a formula for quantifying assortativity, the propensity of similar people to be socially connected with one another more often than their dissimilar counterparts. With this formula, this investigation establishes a foundation for examining assortative patterns in suicidal behavior and highlights how they can be exploited for improved prevention. Specifically, the established clustering of suicide fatalities in time and space implies such fatalities have assortative features. This suggests other forms of suicide-related behavior may as well. Thus, the assortativity of suicide-related verbalizations (SRVs) was examined by machine coding 64 million posts from 17 million users of a large social media platform-Twitter-over 2 distinct 28-day periods. Users were defined as socially linked in the network if they mutually replied to each other at least once. Results show SRVs were significantly more assortative than chance, through 6 degrees of separation. This implies that if a person posts SRVs, their friends' friends' friends' friends' friends' friends are more likely than chance to do the same even though they have never met. SRVs also remained significantly assortative through 2 degrees, even when mood was controlled. Discussion illustrates how these assortative patterns can be exploited to improve the true-positive rate of suicide risk screenings. (PsycInfo Database Record (c) 2020 APA, all rights reserved).

Traumatic Event Exposure, Posttraumatic Stress Disorder, and Sleep Disturbances in a National Sample of U.S. Adults.

Posttraumatic stress disorder (PTSD) is a highly prevalent, debilitating disorder found to develop after exposure to a potentially traumatic event (PTE). Individuals with PTSD often report sleep disturbances, specifically nightmares and insomnia, which are listed within the criteria for PTSD. This research examined prevalence of insomnia and nightmares within a national sample of 2,647 adults (data weighted by age and sex to correct for differences in sample distribution) who had been exposed to one or more PTEs. Prevalence of self-reported sleep disturbance, sleep disturbances by PTE type, and gender differences were examined. All participants completed a self-administered, structured online interview that assessed exposure to stressful events and PTSD symptoms. Among individuals who met DSM-5 criteria for PTSD, a large majority (more than 92%) reported at least one sleep disturbance. Insomnia was relatively more prevalent than PTE-related nightmares among individuals with PTSD and among all PTE-exposed individuals. A higher number of PTEs experienced significantly increased the likelihood of both trauma-related nightmares and insomnia, McFadden's pseudo R2 = .07, p < .001. Women exposed to PTEs were more likely to endorse experience of insomnia, χ2 (1, N = 2,647) = 99.13, p < .001, φ = .194, and nightmares compared to men, χ2 (1, N = 2,648) = 82.98, p < .001, φ = .177, but this gender difference was not significant among individuals with PTSD, ps = .130 and .050, respectively. Differences in sleep disturbance prevalence by PTE type were also examined. Implications for treatment and intervention and future directions are discussed.

Spanish Abstracts by Asociación Chilena de Estrés Traumático (ACET) Exposición a eventos traumáticos, trastorno de estrés postraumático y trastornos del sueño en una muestra nacional de adultos EXPOSICION AL TRAUMA, TEPT Y SUEÑO El trastorno de estrés postraumático (TEPT) es un trastorno altamente prevalente y debilitante que se desarrolla después de la exposición a un evento potencialmente traumático (EPT). Los individuos con TEPT a menudo reportan alteraciones del sueño, específicamente pesadillas e insomnio, que se enumeran dentro de los criterios para el TEPT. Esta investigación examinó la prevalencia de insomnio y pesadillas en una muestra nacional de 2647 adultos (datos ponderados por edad y sexo para corregir las diferencias en la distribución de la muestra) quienes habían sido expuestos a uno o más EPTs. Se examinó la prevalencia de trastornos del sueño autoinformados, los trastornos del sueño por tipo de EPT y las diferencias de género. Todos los participantes completaron una entrevista estructurada auto-administrada online, que evaluó la exposición a eventos estresantes y los síntomas de TEPT. Entre las personas que cumplieron con los criterios del DSM-5 para el TEPT, una gran mayoría (más de 92%) reportó al menos una alteración del sueño. El insomnio fue relativamente más prevalente que las pesadillas relacionadas con EPT entre individuos con TEPT y entre todos los individuos expuestos a EPT. Un mayor número de los que experimentaron EPT aumentó significativamente la probabilidad de ambos, pesadillas relacionadas a trauma e insomnio, pseudo R2 de McFadden = .07, p <.001. Las mujeres expuestas a los EPT fueron más propensas a respaldar la experiencia de insomnio, χ2 (1, N = 2647) = 99.13, p <.001, φ = .194, y pesadillas en comparación con los hombres, χ2 (1, N = 2648) = 82.98, p <.001, φ = .177, pero esta diferencia de género no fue significativa entre los individuos con TEPT, ps = .130 y .050, respectivamente. También se examinaron las diferencias en la prevalencia de trastornos del sueño por tipo de EPT. Se discuten las implicaciones para el tratamiento y la intervención y las direcciones futuras.

Patients' reasons for weight loss and their relations to clinical weight loss outcomes in a comprehensive lifestyle intervention.

OBJECTIVE: Research suggests that individuals seeking weight loss treatment do so for a variety of reasons. Limited work has explored relations of reasons for weight loss to patient characteristics or to weight loss outcomes. The current study examined these relations. METHODS: The sample consisted of 588 patients in a 15-week fee-for-service weight loss programme. Prior to the intervention, patients completed questionnaires including items on reasons for weight loss, demographic characteristics, and a variety of weight-based characteristics. Patients' weight change outcomes were expressed as percent weight loss and also categorized into one of three previously described weight loss trajectories. RESULTS: The results of chi-squared and t-test analyses suggested that endorsement of health concerns, mobility concerns, or another person's recommendation was associated with higher body mass index (BMI) and older age. These reasons were more likely to be endorsed by White patients than Black patients and by male patients than female patients. Endorsement of doctor recommendation was more likely to be seen among Black patients than White patients. There was no significant relation of any weight loss reason with weight loss outcome. CONCLUSIONS: While certain reasons for weight loss were more often cited by certain patient groups, no specific reason predicted a better or worse outcome.