Alice Guilleux

RespOnse Shift ALgorithm in Item response theory (ROSALI) for response shift detection with missing data in longitudinal patient-reported outcome studies

AbstractPurposeSome IRT models have the advantage of being robust to missing data and thus can be used with complete data as well as different patterns of missing data (informative or not). The purpose of this paper was to develop an algorithm for response shift (RS) detection using IRT models allowing for non-uniform and uniform recalibration, reprioritization RS recognition and true change estimation with these forms of RS taken into consideration if appropriate. MethodsThe algorithm is described, and its implementation is shown and compared to Oort’s structural equation modeling (SEM) procedure using data from a clinical study assessing health-related quality of life in 669 hospitalized patients with chronic conditions.ResultsThe results were quite different for the two methods. Both showed that some items of the SF-36 General Health subscale were affected by response shift, but those items usually differed between IRT and SEM. The IRT algorithm found evidence of small recalibration and reprioritization effects, whereas SEM mostly found evidence of small recalibration effects.ConclusionAn algorithm has been developed for response shift analyses using IRT models and allows the investigation of non-uniform and uniform recalibration as well as reprioritization. Differences in RS detection between IRT and SEM may be due to differences between the two methods in handling missing data. However, one cannot conclude on the differences between IRT and SEM based on a single application on a dataset since the underlying truth is unknown. A next step would be to implement a simulation study to investigate those differences.

Towards a Validation of the Three Pathways Model of Pathological Gambling

With the aim of validating the three pathways hypothesis of pathological gambling (Blaszczynski and Nower in Addiction 97:487–499, 2002) 372 pathological gamblers meeting DSM IV (2000) criteria were assessed via a structured clinical interview as well as being subjected to personality tests and evaluation of their gambling practices. Our results show that it is possible to identify three subgroups corresponding to the three pathways: behaviourally conditioned problem gamblers, emotionally vulnerable problem gamblers and antisocial impulsivist problem gamblers. Our results particularly demonstrate that impulsivist gamblers preferentially choose semi-skilful gambling (horse racing and sports gambling) whereas emotionally vulnerable gamblers are significantly more attracted to games of chance (one-armed bandits, scratch cards, etc.) This led us to propose a functional presentation of the three pathways model which differs somewhat from the Blaszczynski and Nower presentation.

Power and Sample Size Determination in the Rasch Model: Evaluation of the Robustness of a Numerical Method to Non-Normality of the Latent Trait

Patient-reported outcomes (PRO) have gained importance in clinical and epidemiological research and aim at assessing quality of life, anxiety or fatigue for instance. Item Response Theory (IRT) models are increasingly used to validate and analyse PRO. Such models relate observed variables to a latent variable (unobservable variable) which is commonly assumed to be normally distributed. A priori sample size determination is important to obtain adequately powered studies to determine clinically important changes in PRO. In previous developments, the Raschpower method has been proposed for the determination of the power of the test of group effect for the comparison of PRO in cross-sectional studies with an IRT model, the Rasch model. The objective of this work was to evaluate the robustness of this method (which assumes a normal distribution for the latent variable) to violations of distributional assumption. The statistical power of the test of group effect was estimated by the empirical rejection rate in data sets simulated using a non-normally distributed latent variable. It was compared to the power obtained with the Raschpower method. In both cases, the data were analyzed using a latent regression Rasch model including a binary covariate for group effect. For all situations, both methods gave comparable results whatever the deviations from the model assumptions. Given the results, the Raschpower method seems to be robust to the non-normality of the latent trait for determining the power of the test of group effect.

Statistical challenges of quality of life and cancer: new avenues for future research

Statistical modeling conference on the quality of life measurements of the French National Platform of Quality of Life and Cancer Faculty of Science in Luminy, Marseille, France, 12–13 September 2013 The French National Platform of Quality of Life and Cancer and the statistical team of the Mathematical Institute of Luminy undertook a successful first conference addressing the statistical challenges of measuring the quality of life in the field of oncology. More than 15 presentations were made over a 2-day period by the Faculty of Sciences in Luminy. The conference managed to assemble participants from different disciplines, such as mathematics and statistics, public health, epidemiology and psychology, to debate the key statistical and methodological issues of quality of life measurement and analysis. Three main topics were covered in this conference: the treatment of missing data, the development of item banking and computerised adaptive testing and the detection/understanding of response shift.

Gambling behaviors and psychopathology related to Attention-Deficit/Hyperactivity Disorder (ADHD) in problem and non-problem adult gamblers

Previous studies showed that Pathological Gambling and Attention Deficit/Hyperactivity Disorder (ADHD) often co-occur. The aim of this study was to examine whether ADHD is associated with specific severity patterns in terms of gambling behavior, psychopathology and personality traits. 599 problem and non-problem-gamblers were recruited in addiction clinics and gambling places in France. Subjects were assessed with the Wender-Utah Rating Scale-Child, the Adult ADHD Self-Report Scale, the Mini International Neuropsychiatric Interview, the Temperament and Character Inventory, the South Oaks Gambling Screen and questionnaires assessing gambling related cognitive distortions and gambling habits. 20.7% (n=124) of gamblers were screened positive for lifetime or current ADHD. Results from the multivariate analysis showed that ADHD was associated with a higher severity of gambling-related problems and with more psychiatric comorbidity. Among problem gamblers, subjects with history of ADHD were also at higher risk for unemployment, psychiatric comorbidity and specific dysfunctional personality traits. This study supports the link between gambling related problems and ADHD in a large sample of problem and non-problem gamblers, including problem-gamblers not seeking treatment. This points out the necessity to consider this disorder in the prevention and in the treatment of pathological gambling.

Cognitive distortions and ADHD in pathological gambling: A national longitudinal case-control cohort study

Introduction The primary outcome of our study was to assess the links between the level of cognitive distortions and the severity of gambling disorder. We also aimed at assessing the links between patient gambling trajectories and attention deficit and hyperactivity disorder (ADHD). Materials and methods The study population (n = 628) was comprised of problem and non-problem gamblers of both sexes between 18 and 65 years of age, who reported gambling on at least one occasion during the previous year. Data encompassed socio-demographic characteristics, gambling habits, the South Oaks Gambling Screen, the Gambling Attitudes and Beliefs Survey – 23, the Wender Utah Rating Scale – Child, and the Adult ADHD Self-report Scale. Results The cognitive distortions with the greatest correlation to the severity of gambling disorder were the “Chasing” and “Emotions.” These two dimensions were able to distinguish between problem gamblers seeking treatment or not. While age of onset of gambling and length of gambling practice were not associated with the level of distorted cognitions, a period of abstinence of at least 1 month was associated with a lower level of distorted cognitions. The presence of ADHD resulted in a higher level of distorted cognitions. Conclusion Cognitive work is essential to the prevention, and the treatment, of pathological gambling, especially with respect to emotional biases and chasing behavior. The instauration of an abstinence period of at least 1 month under medical supervision could be a promising therapeutic lead for reducing gambling-related erroneous thoughts and for improving care strategies of pathological gamblers.

Gambling disorder-related illegal acts: Regression model of associated factors

Background and aims Gambling disorder-related illegal acts (GDRIA) are often crucial events for gamblers and/or their entourage. This study was designed to determine the predictive factors of GDRIA. Methods Participants were 372 gamblers reporting at least three DSM-IV-TR (American Psychiatric Association, 2000) criteria. They were assessed on the basis of sociodemographic characteristics, gambling-related characteristics, their personality profile, and psychiatric comorbidities. A multiple logistic regression was performed to identify the relevant predictors of GDRIA and their relative contribution to the prediction of the presence of GDRIA. Results Multivariate analysis revealed a higher South Oaks Gambling Scale score, comorbid addictive disorders, and a lower level of income as GDRIA predictors. Discussion and conclusion An original finding of this study was that the comorbid addictive disorder effect might be mediated by a disinhibiting effect of stimulant substances on GDRIA. Further studies are necessary to replicate these results, especially in a longitudinal design, and to explore specific therapeutic interventions.

Method variation in the impact of missing data on response shift detection

PurposeMissing data due to attrition or item non-response can result in biased estimates and loss of power in longitudinal quality-of-life (QOL) research. The impact of missing data on response shift (RS) detection is relatively unknown. This overview article synthesizes the findings of three methods tested in this special section regarding the impact of missing data patterns on RS detection in incomplete longitudinal data.MethodsThe RS detection methods investigated include: (1) Relative importance analysis to detect reprioritization RS in stroke caregivers; (2) Oort’s structural equation modeling (SEM) to detect recalibration, reprioritization, and reconceptualization RS in cancer patients; and (3) Rasch-based item-response theory-based (IRT) models as compared to SEM models to detect recalibration and reprioritization RS in hospitalized chronic disease patients. Each method dealt with missing data differently, either with imputation (1), attrition-based multi-group analysis (2), or probabilistic analysis that is robust to missingness due to the specific objectivity property (3).ResultsRelative importance analyses were sensitive to the type and amount of missing data and imputation method, with multiple imputation showing the largest RS effects. The attrition-based multi-group SEM revealed differential effects of both the changes in health-related QOL and the occurrence of response shift by attrition stratum, and enabled a more complete interpretation of findings. The IRT RS algorithm found evidence of small recalibration and reprioritization effects in General Health, whereas SEM mostly evidenced small recalibration effects. These differences may be due to differences between the two methods in handling of missing data.ConclusionsMissing data imputation techniques result in different conclusions about the presence of reprioritization RS using the relative importance method, while the attrition-based SEM approach highlighted different recalibration and reprioritization RS effects by attrition group. The IRT analyses detected more recalibration and reprioritization RS effects than SEM, presumably due to IRT’s robustness to missing data. Future research should apply simulation techniques in order to make conclusive statements about the impacts of missing data according to the type and amount of RS.

Marked Body Shape Concerns in Female Patients Suffering from Eating Disorders: Relevance of a Clinical Sub-Group

Concerns about body shape and weight are core diagnostic criteria for eating disorders although intensity varies between patients. Few studies have focused on the clinical differences relative to the intensity of these concerns. Nonetheless, they might have a prognostic value. This study was aimed at identifying the characteristics associated with marked body shape concerns in patients with an eating disorder. Data was collected from a systematic and standardized clinical assessment of outpatients seeking treatment in our department for eating disorders. Only female patients, suffering from anorexia nervosa or bulimia nervosa, and with “no / mild” or “marked” body shape concerns according to the Body Shape Questionnaire, were included for the present study. We focused on sociodemographic characteristics, eating disorder characteristics, axis 1 disorders, types of attachment, self-esteem and dissociation. A multiple logistic regression was performed to identify factors related to “marked” body shape concerns. In our sample (123 participants, with a mean age of 24.3 years [range 16–61]), 56.9% had marked concerns with body shape. Marked body shape concerns were associated with a major depressive episode (OR = 100.3), the use of laxatives (OR = 49.8), a high score on the item “body dissatisfaction” of the Eating Disorders Inventory scale (OR = 1.7), a higher minimum body mass index (OR = 1.73), and a high score on the item “loss of control over behavior, thoughts and emotions” from the dissociation questionnaire (OR = 10.74). These results are consistent with previous studies, and highlight the importance of denial.

Gambling among European professional athletes. Prevalence and associated factors.

ABSTRACT In Europe, the prevalence of gambling disorders in the general population ranges from 0.15 to 6.6%. Professional athletes are known for having risk factors for addictive behaviors, such as young age or sensation seeking, though no study has yet tried to evaluate the prevalence of gambling and gambling disorders among this specific population. The goals of this study were to estimate the prevalence of gambling, problematic or not, among European professional athletes and to explore the factors that are associated with gambling practice and gambling problems in professional athletes. A self-completion questionnaire was specifically designed for this study. The questionnaires were distributed by European Union athletes to professional ice hockey, rugby, handball, basketball, football, indoor football, volleyball, and cricket teams in Spain, France, Greece, Ireland, Italy, Sweden, and the United Kingdom. Socio-demographic variables (age, sex, education, marital and parental status, sport, country of birth, and country of practice), variables linked to gambling (gambling habits, screening of gambling problems with the Lie/Bet questionnaire, and gambling related cognitions), and impulsive behavior data (urgency, premeditation, perseverance, and sensation seeking [UPPS]-Short Form questionnaire) were gathered. There were 1,236 questionnaires filled out. The percentage of professional athletes that had gambled at least once during the previous year was 56.6%. The prevalence of problem gambling, current or past, was 8.2%. A certain number of variables were associated with the gambling status. In particular, betting on ones own team (OR = 4.1, CI95% [1.5–11.5]), betting online (OR = 2.9, CI95% [1.6–5.4]), gambling regularly (OR = 4.0, CI95% [2.1–7.6]), and having a high positive urgency score (OR = 1.5, CI95% [1.3–1.7]) were associated with gambling problems, current or past, among professional athletes. Professional athletes are particularly exposed to both gambling and problem gambling.