Dropout Predictors Research Articles

Identification of Psychological Treatment Dropout Predictors Using Machine Learning Models on Italian Patients Living with Overweight and Obesity Ineligible for Bariatric Surgery.

According to the main international guidelines, patients with obesity and psychiatric/psychological disorders who cannot be addressed to surgery are recommended to follow a nutritional approach and a psychological treatment. A total of 94 patients (T0) completed a battery of self-report measures: Symptom Checklist-90-Revised (SCL-90-R), Barratt Impulsiveness Scale-11 (BIS-11), Binge-Eating Scale (BES), Obesity-Related Well-Being Questionnaire-97 (ORWELL-97), and Minnesota Multiphasic Personality Inventory-2 (MMPI-2). Then, twelve sessions of a brief psychodynamic psychotherapy were delivered, which was followed by the participants completing the follow-up evaluation (T1). Two groups of patients were identified: Group 1 (n = 65), who fully completed the assessment in both T0 and T1; and Group 2-dropout (n = 29), who fulfilled the assessment only at T0 and not at T1. Machine learning models were implemented to investigate which variables were most associated with treatment failure. The classification tree model identified patients who were dropping out of treatment with an accuracy of about 80% by considering two variables: the MMPI-2 Correction (K) scale and the SCL-90-R Phobic Anxiety (PHOB) scale. Given the limited number of studies on this topic, the present results highlight the importance of considering the patient's level of adaptation and the social context in which they are integrated in treatment planning. Cautionary notes, implications, and future directions are discussed.

Predictors of dropout, time spent on the program and client satisfaction in an internet-based, telephone-assisted CBT anxiety program among elementary school children in a population-based sample.

Increasing evidence has shown that childhood anxiety can be effectively treated by Internet-based cognitive behavioral therapy (ICBT). Being able to predict why participants decide to drop out of such programs enables scarce resources to be used appropriately. The aim of this study was to report dropout predictors for a population-based ICBT intervention aimed at children with anxiety, together with the time they and their parents spent on the program and client satisfaction rates. The study focused on 234 Finnish children aged 10-13 who received an ICBT intervention, with telephone support, for anxiety symptoms, as a part of a randomized control trial. Their parents also had access to Internet-based material and participated in the weekly telephone calls with the coach. Possible drop out factors were explored and these included various family demographics, child and parent psychopathology and therapeutic alliance. Just under a fourth (23.9%) of the children dropped out of the intervention. The risk was higher if the child did not fulfill the criteria for any anxiety diagnosis or reported a poorer therapeutic alliance. Family demographics and the COVID-19 pandemic did not increase the risk. The families spent an average of 127min on the webpage each week and an average of 32min on the phone calls. The overall satisfaction with the program was 87% for the children and 95% for the parents. Both the children and the parents found the telephone calls helpful. These findings are important in clinical practice when assessing a family's eligibility for ICBT.

Clusters Based on Within-Treatment Symptom Trajectories as Predictors of Dropout in Treatment for Posttraumatic Stress Disorder and Substance Use Disorder

Objective: Dropout rates are high in treatments for co-occurring posttraumatic stress disorder (PTSD) and substance use disorders (SUDs). We examined dropout predictors in PTSD-SUD treatment. Methods: Participants were 183 veterans receiving integrated or phased motivational enhancement therapy and prolonged exposure. Using survival models, we examined demographics and symptom trajectories as dropout predictors. Using latent trajectory analysis, we incorporated clusters based on symptom trajectories to improve dropout prediction. Results: Hispanic ethnicity (integrated arm), Black or African American race (phased arm), and younger age (phased arm) predicted dropout. Clusters based on PTSD and substance use trajectories improved dropout prediction. In integrated treatment, participants with consistently-high use and low-and-improving use had the highest dropout. In phased treatment, participants with the highest and lowest PTSD symptoms had lower dropout; participants with the lowest substance use had higher dropout. Conclusions: Identifying within-treatment symptom trajectories associated with dropout can help clinicians intervene to maximize outcomes. ClinicalTrials.gov Identifier: NCT01211106.

Statistical identification of predictors of dropout in secondary education: evidence from Malaysia

The issue of school dropout has been long rooted in Malaysia. To address this issue effectively, further insight into the predictors of the school dropout and the early statistical identification of dropouts are crucial. Nevertheless, uncertainty has been persistent on these two aspects. This paper aims to identify the predictors and assess the feasibility of early identification of school dropouts. Using a school administrative database that comprised data on more than 100,000 students, the school dropout predictors, such as gender, academic achievement, and household income were identified. The estimated model correctly identified early on the 20.83% of the school dropouts. This is substantially higher than the overall sample mean of school dropouts (5%). Thus, it is feasible to identify early on the school dropout statistically. The findings have provided insightful inputs that could strengthen our intervention strategies and policies to help alleviate the problem of school dropout in Malaysia.

Predictors of Dropout Among Psychosomatic Rehabilitation Patients During the COVID-19 Pandemic: Secondary Analysis of a Longitudinal Study of Digital Training.

High dropout rates are a common problem reported in web-based studies. Understanding which risk factors interrelate with dropping out from the studies provides the option to prevent dropout by tailoring effective strategies. This study aims to contribute an understanding of the predictors of web-based study dropout among psychosomatic rehabilitation patients. We investigated whether sociodemographics, voluntary interventions, physical and mental health, digital use for health and rehabilitation, and COVID-19 pandemic-related variables determine study dropout. Patients (N=2155) recruited from 4 psychosomatic rehabilitation clinics in Germany filled in a web-based questionnaire at T1, which was before their rehabilitation stay. Approximately half of the patients (1082/2155, 50.21%) dropped out at T2, which was after the rehabilitation stay, before and during which 3 voluntary digital trainings were provided to them. According to the number of trainings that the patients participated in, they were categorized into a comparison group or 1 of 3 intervention groups. Chi-square tests were performed to examine the differences between dropout patients and retained patients in terms of sociodemographic variables and to compare the dropout rate differences between the comparison and intervention groups. Logistic regression analyses were used to assess what factors were related to study dropout. The comparison group had the highest dropout rate of 68.4% (173/253) compared with the intervention groups' dropout rates of 47.98% (749/1561), 50% (96/192), and 42.9% (64/149). Patients with a diagnosis of combined anxiety and depressive disorder had the highest dropout rate of 64% (47/74). Younger patients (those aged <50 y) and patients who were less educated were more likely to drop out of the study. Patients who used health-related apps and the internet less were more likely to drop out of the study. Patients who remained in their jobs and patients who were infected by COVID-19 were more likely to drop out of the study. This study investigated the predictors of dropout in web-based studies. Different factors such as patient sociodemographics, physical and mental health, digital use, COVID-19 pandemic correlates, and study design can correlate with the dropout rate. For web-based studies with a focus on mental health, it is suggested to consider these possible dropout predictors and take appropriate steps to help patients with a high risk of dropping out overcome difficulties in completing the study.

Early dropout predictors in social sciences and management degree students

ABSTRACT Student dropout is a major concern in studies investigating retention strategies in higher education. This study identifies which variables are important to predict student dropout, using academic data from 3583 first-year students on the Business Administration (BA) degree at the University of Barcelona (Spain). The results indicate that two variables, the percentage of subjects failed and not attended in the first semester, demonstrate significant predictive power. This has been corroborated with an additional sample of 10,784 students from three-degree programs (Law, BA, and Economics) at the Complutense University of Madrid (Spain), to assess the robustness of the results. Three different algorithms have also been utilized: neural networks, random forest, and logit. In the specific case of neural networks, the NeuralSens methodology has been employed, which is based on the use of sensitivities, allowing for its interpretation. The outcomes are highly consistent in all cases: both a simple model (logit) and more sophisticated ones (neural networks and random forest) exhibit high accuracy (correctly predicted values) and sensitivity (correctly predicted dropouts). In test set average values of 77% and 69% have been respectively achieved. In this regard, a noteworthy point is that only academic data from the university itself was used to develop the models. This ensures that there’s no dependence on other personal or organizational variables, which can often be difficult to access.

Treatment dropout among veterans and their families: Quantitative and qualitative findings.

Psychotherapy noncompletion rates for veterans and their families are high. This study sought to (a) measure noncompletion rates of such patients at a university-based treatment center, (b) compare veteran and family member attrition rates, (c) identify dropout predictors, and (d) explore clinicians' perspectives on treatment noncompletion. Using quantitative and qualitative approaches, we analyzed demographic and clinical characteristics of 141 patients (90 military veterans; 51 family members) in a university treatment center. We defined dropout as not completing the time-limited therapy contract. Reviewing semistructured interview data assessing clinicians' perspectives on their patients' dropout, three independent raters agreed on key themes, with interrater coefficient kappa range .74 to 1. Patient attrition was 24%, not differing significantly between veterans and family members. Diagnosis of major depression (MDD) and exposure-based therapies predicted noncompletion, as did higher baseline Hamilton Depression Rating Scale (HDRS) total scores, severe depression (HDRS > 20), lack of Beck Depression Inventory weekly improvement, and history of military sexual trauma. Clinicians mostly attributed noncompletion to patient difficulties coping with intense emotions, especially in exposure-based therapies. Noncompletion rate at this study appeared relatively low compared to other veteran-based treatment centers, if still unfortunately substantial. Patients with comorbid MDD/PTSD and exposure-based therapies carried greater noncompletion risk due to the MDD component, and this should be considered in treatment planning. Ongoing discussion of dissatisfaction and patient discontinuation, in the context of a strong therapeutic alliance, might reduce noncompletion in this at-risk population. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

A Path Model of University Dropout Predictors: The Role of Satisfaction, the Use of Self-Regulation Learning Strategies and Students’ Engagement

University dropout is a phenomenon that is a concern in many countries all over the world. However, although there are studies in which the direct relationship of the personal and contextual variables is observed individually to predict dropout, there is little research to know whether any of these variables mediate each other in a more dynamic and complex model. Thus, the objective of this study was to analyze the extent to which the intention to drop out of university courses is predicted by (i) satisfaction and expectations with the course, (ii) engagement with the course, and (iii) by the use of Self-Regulated Learning (SRL) strategies. Eight hundred and seventy-seven students from two Spanish universities completed the CARE questionnaire. Path analyses were performed using Mplus 8.3. The data obtained indicate that the intention to drop out is directly and significantly explained by students´ engagement (in 17.8%) and indirectly explained by the use of SRL strategies through engagement. Changes in engagement and in the use of SRL strategies were seen to be associated with satisfaction. Finally, the effect of satisfaction and the use of SRL strategies explained a proportion of students’ engagement (53.6%). It is important for research or interventions focused on students’ intention to drop out to understand that there are multiple variables that both directly and indirectly influence those intentions.

The Academic Dropout Wheel Analyzing the Antecedents of Higher Education Dropout in Education Studies

Higher education dropout has been considered a major concern for several researchers in the field of education around the World. Although different studies were carried out to deal with the topic, they all faced common limitations. This paper explores the twofold research conducted to investigate higher education dropout in Education studies at two of the main Catalan public universities. First, a review of the different theoretical perspectives that tackled dropout was carried out to combine the direct and indirect dropout predictors. Second, some research was conducted in the faculties of Education in the selected two Catalan public universities to investigate the antecedents of higher education dropout. The utilized instrument discussed in this paper is the survey. In the first part, the predictors of higher education dropout were combined in a 5-layer model called the ¨Dimensions of Academic Dropout” wheel (Naaman, 2018). In the second part, the results of the survey analyzed quantitively revealed that dissatisfaction from the program and academic failure are the most important antecedents of higher education dropout in education studies in Catalonia.

Investigating Effectiveness and Predicting Outcome and Dropout From Systems Training for Emotional Predictability and Problem Solving for Emotional Intensity Difficulties (STEPPS EI) in UK Primary Care.

Systems Training for Emotional Predictability and Problem Solving (STEPPS) for borderline personality disorder (BPD) has been adapted for a primary care setting into a 13-week group treatment for emotional intensity difficulties (STEPPS EI). This is the first study to examine the effectiveness and potential outcome and dropout predictors of STEPPS EI in a primary care setting. Severity of BPD, depression, and anxiety symptoms were measured pre- and postintervention for 148 participants. Treatment completers showed improvements in depression, anxiety, and BPD symptoms with medium to large effect sizes. A predictor of symptom improvement was higher baseline severity for each of the symptom measures. Attending a psychoeducational group prior to STEPPS EI was associated with lower odds of dropout. The findings support the use of STEPPS EI in a primary care setting, with the potential to alleviate the burden from other local services.

Dropout predictors at gyms: a retrospective study

ABSTRACT This study aimed to verify the risk factors associated with gym regulars’ dropout over 12 months. This is an observational, retrospective, cohort study. The sample consisted of two gyms (Gym 1: n=2027, age=39.1 ±13.1 years; Gym 2: n=1775, age=36.8 ±11.6 years). Each participant was followed up for 12 months, or until dropout, by controlling the biometric entry system using a turnstile. The adherence rate was of 11% at Gym 1 and 19% at Gym 2. At both gyms, training frequency ≤3x/month and 4-6x/month was associated with dropout. At gym 1, being a female, aged ≥43 years, and enrollment between October and December were dropout predictors. At Gym 2, dropout was associated with contract length ≤31 days and 32-186 days and enrollment between April and December.

Low 6-Minute Walk Distance and Muscle Mass Predict Drop out in Cardiac Rehabilitation.

Cardiac rehabilitation after percutaneous coronary intervention decreases recurrence and mortality but has a high dropout rate. The aim of this study is to identify dropout predictors by comparing the characteristics of complete and dropout patients in cardiac rehabilitation. The study included 593 patients (455 men and 138 women) who received percutaneous coronary intervention and were enrolled in a 1-year cardiac rehabilitation program consisting of home-based cardiac rehabilitation with three center visits. Dropout was defined as participation in the first center visit but not the second or third center visits. Blood lipids, quality of life, socioeconomic status, and 6-minute walk distance measurements at the first visit were compared between participants who completed and dropped out of cardiac rehabilitation. For both men and women, the dropout rate significantly correlated with a low 6-minute walk distance and low muscle mass ratio. The dropout rate was significantly higher for men, but not women, with low education and low income. However, the dropout rate was decreased for women, but not men, with low blood pressure and triglycerides. An improved understanding of the characteristics of participants and the cardiac rehabilitation dropout rate are expected to contribute to the development of cardiac rehabilitation strategies that decrease patient dropout.

Psychopathological Features and Drop-Out Predictors in a Sample of Individuals with Substance Use Disorder Under Residential Community Treatment.

ObjectiveSeveral studies have been conducted to investigate the relationship between addiction and crimes, but little is known about the treatment of individuals with substance use disorder (SUD) with criminal records. This study aimed to assess the treatment progress of a group of individuals with SUD who underwent treatment within a residential community, and to analyze their personality profiles to identify drop-out predictors.MethodWe evaluated 49 subjects using the Psychopathic Personality Inventory-Revised (PPI-R), the Psychological Treatment Inventory (PTI), the Barratt Impulsiveness Scale-11 (BIS-11), and the Toronto Alexithymia Scale (TAS-20) and carrying out various statistical analyses, including the t-test, Cohen’s d, analysis of variance (ANOVA), and discriminant analysis.ResultsResults are discussed within the context of previous studies on this topic. Our results showed that variables such impulsiveness, cold-heartedness, alexithymia, and psychopathic traits influenced the premature treatment abandonment of individuals with SUD and criminal records.ConclusionsThis study provides a further piece for the understanding of subjects with SUD and criminal records, suggesting the importance of a psychodynamic integrated approach, and showing the impact of some psychopathological features on treatment drop-out.

The When and Who of Graduation and Dropout Predictors: A Moderated Hazard Analysis

This study examined graduation and persistence among social and behavioral science students at a regional comprehensive university. Hazard analyses identified predictors of student trajectories, times at which predictors were more or less impactful, and interactions between predictors such that particular risk factors were more detrimental for certain groups. Results revealed that first-time freshmen and first-time transfer students operate under very different risk models. Total units enrolled and cumulative grade point average emerged as the most salient predictors across all models; however, numerous predictors varied in their salience across time and student subgroup. Differential predictors functioned as risks early versus late in student trajectories. Underrepresented minority status emerged as a risk factor in interaction with other predictors, such that it amplified other risks. These findings suggest that examinations of graduation and persistence must highlight the complex ways in which traditional risk factors interact with time and context to impact student success.

Predictors of Dropout in Exercise Trials in Older Adults: The Generation 100 Study.

Dropout from exercise programs, both in the real world and in research, is a challenge, and more information on dropout predictors is needed for establishing strategies to increase the likelihood of maintaining participants in a prescribed exercise program. The aim of the present study was to determine the dropout rate and its predictors during a 3-yr exercise program in older adults. In total, 1514 men and women (mean ± SD age = 72.4 ± 1.9 yr) were included in the present study. Participants were randomized to either a supervised exercise intervention or to follow national guidelines for physical activity (PA). Self-reported demographics (e.g., education), general health, morbidity (e.g., heart disease, memory loss, and psychological distress), smoking, and PA were examined at baseline. Cardiorespiratory fitness (CRF) and grip strength were directly measured at baseline. Dropout rate was evaluated after 1 and 3 yr. Multivariate logistic regression analysis was used to identify dropout predictors. The total dropout rate was 11.0% (n = 166) after 1 yr and 14.9% (n = 225) after 3 yr. Significant predictors of dropout after 1 yr were low education, low grip strength, lower cardiorespiratory fitness, low PA level, and randomization to supervised exercise. The same predictors of dropout were significant after 3 yr, with reduced memory status as an additional predictor. This is the largest study to identify dropout predictors in a long-term exercise program in older adults. Our findings provide new and important knowledge about potential risk factors of dropout in long-term exercise programs in older adults.

Identification of dropout predictors to a community-based physical activity programme that uses motivational interviewing.

Participant dropout reduces intervention effectiveness. Predicting dropout has been investigated for Exercise Referral Schemes, but not physical activity (PA) interventions with Motivational Interviewing (MI). Data from attendees (n = 619) to a community-based PA programme utilizing MI techniques were analysed using a chi-squared test to determine dropout and attendance group differences. Binary logistic regression investigated the likelihood of dropout before 12 weeks. A total of 44.7% of participants dropped out, with statistical (P < 0.05) differences between groups for age, PA and disability. Regression for each variable showed participants aged 61-70 years (OR = 0.28, CI = 0.09-0.79; P = 0.018), >70 years (OR = 0.30, CI = 0.09-0.90; P = 0.036), and high PA (OR = 0.40, CI = 0.20-0.75; P = 0.006) reduced dropout likelihood. Endocrine system disorders (OR = 4.24, CI = 1.19-19.43; P = 0.036) and musculoskeletal disorders (OR = 3.14, CI = 1.84-5.45; P < 0.001) increased dropout likelihood. Significant variables were combined in a single regression model. Dropout significantly reduced for 61-70 years old (OR = 0.31, CI = 0.10-0.90; P = 0.035), and high PA (OR = 0.39, CI = 0.19-0.76; P = 0.008). Musculoskeletal disorders increased dropout (OR = 2.67, CI = 1.53-4.75; P < 0.001). Age, PA and disability type significantly influence dropout at 12 weeks. These are the first results specific to MI based programmes indicating the inclusion of MI and highlighting the need for further research.

Impulsivity and problem awareness predict therapy compliance and dropout from treatment for gambling disorder.

This study investigates the predictive value of impulsivity traits (as measured by the UPPS-P impulsive behaviour scale) and relevant covariates (sociodemographics, gambling severity, dysphoric mood, other potentially addictive behaviours, and non-verbal intelligence) with regard to treatment dropout and level of adherence to therapy guidelines and instructions in patients with gambling disorder. Sixty-six patients seeking treatment for gambling disorder, and recruited to participate in a larger protocol (G-Brain), were initially assessed in impulsivity traits and relevant covariates in the first six months after admission. Of these, 24 patients dropped out (DO) and 42 patients remained in therapy (NDO) during the subsequent 6-month follow-up period. A multivariate analysis of impulsivity subscales suggested prospective differences between DO and NDO, with affect-driven dimensions (positive and negative urgency) seemingly driving these differences. Among these, only positive urgency independently predicted a slight increase in the drop-out probability. In the NDO group, a higher degree of adherence to therapy was independently predicted by lower sensation-seeking scores and stronger awareness of gambling-related problems. Results suggest the presence of affect-driven impulsivity traits as dropout predictors in patients with gambling disorder. Awareness of gambling-related problems and lower sensation-seeking enhanced compliance with therapeutic guidelines and instructions.

Longitudinal drop-out and weighting against its bias

BackgroundThe bias caused by drop-out is an important factor in large population-based epidemiological studies. Many studies account for it by weighting their longitudinal data, but to date there is no detailed final approach for how to conduct these weights.MethodsIn this study we describe the observed longitudinal bias and a three-step longitudinal weighting approach used for the longitudinal data in the MoMo baseline (N = 4528, 4–17 years) and wave 1 study with 2807 (62%) participants between 2003 and 2012.ResultsThe most meaningful drop-out predictors were socioeconomic status of the household, socioeconomic characteristics of the mother and daily TV usage. Weighting reduced the bias between the longitudinal participants and the baseline sample, and also increased variance by 5% to 35% with a final weighting efficiency of 41.67%.ConclusionsWe conclude that a weighting procedure is important to reduce longitudinal bias in health-oriented epidemiological studies and suggest identifying the most influencing variables in the first step, then use logistic regression modeling to calculate the inverse of the probability of participation in the second step, and finally trim and standardize the weights in the third step.

A review of mental health treatment dropout by ethnic minority youth

A large proportion of treatments in youth mental health care are prematurely terminated by the patient. Treatment dropout can have severe consequences. Since ethnic minority youth are treated less often for mental disorders than other youth, it is important to analyse their risk for dropout and to determine if there are ethnicity-specific determinants. This review aimed to provide an overview of the findings from empirical studies on child and adolescent therapy dropout by ethnic minority and to determine if there were ethnicity-specific dropout determinants. An extensive literature search was performed to locate relevant journal articles. Identified articles were inspected for relevant references and these articles were then included in the meta-analysis. A total of 27 studies were accepted for analysis. The results showed that ethnic minority patients have a higher risk of treatment dropout than ethnic majority patients and that dropout rates are ethnically specific. Several differences in dropout predictors among the ethnic groups were found. In spite of diverse results, review limitations, and the lack of several key variables in the available research, some clinical recommendations are made. The review indicates that to prevent dropout, therapists should pay attention to variables such as ethnic background, therapist-patient ethnic match, and the quality of the therapeutic relationship.

Gambling Disorder: Exploring Pre-treatment and In-treatment Dropout Predictors. A UK Study.

The aim of this study was to identify predictors of treatment dropout in a sample of gamblers attending a specialist clinic for gambling disorder. We analysed data on 846 treatment-seeking pathological gamblers. Firstly, we investigated differences in socio-demographic and clinical variables between treatment completers and pre-treatment dropouts, as well as between treatment completers and during-treatment dropouts. Subsequently, variables were entered into a multinomial logistic regression model to identify significant predictors of pre-treatment and in-treatment dropout. Overall, 44.8% of clients did not complete the treatment: 27.4% dropped out before starting it, while 17.4% dropped out during the treatment. Younger age and use of drugs were associated with pre-treatment dropout, while family history of gambling disorder, a lower PGSI score, and being a smoker were related with in-treatment dropout. Our findings suggest that pre-treatment dropouts differ from in-treatment dropouts, and, thus, further research will benefit from considering these groups separately. In addition, this newly gained knowledge will also be helpful in increasing treatment retention in specific subgroups of problem gamblers.