The detection of psychiatric illness by questionnaire

25. Multiple Imputation for Nonresponse in Surveys. By D. B. Rubin. ISBN 0 471 08705 X. Wiley, Chichester, 1987. 258 pp. £30.25.

Multiple Imputation for Nonresponse in Surveys

This sales letter may not influence you to be smarter, but the book that we offer will evoke you to be smarter. Yeah, at least you'll know more than others who don't. This is what called as the quality life improvisation. Why should this research methods in social relations? It's because this is your favourite theme to read. If you like this theme about, why don't you read the book to enrich your discussion?

Research methods in social relations

The goal of multiple imputation is to provide valid inferences for statistical estimates from incomplete data. To achieve that goal, imputed values should preserve the structure in the data, as well as the uncertainty about this structure, and include any knowledge about the process that generated the missing data. Two approaches for imputing multivariate data exist: joint modeling (JM) and fully conditional specification (FCS). JM is based on parametric statistical theory, and leads to imputation procedures whose statistical properties are known. JM is theoretically sound, but the joint model may lack flexibility needed to represent typical data features, potentially leading to bias. FCS is a semi-parametric and flexible alternative that specifies the multivariate model by a series of conditional models, one for each incomplete variable. FCS provides tremendous flexibility and is easy to apply, but its statistical properties are difficult to establish. Simulation work shows that FCS behaves very well in ...

Multiple imputation of discrete and continuous data by fully conditional specification

Rob Eisinga, Manfred te Grotenhuis, Ben Pelzer Department of Social Science Research Methods and Department of Sociology, Radboud University Nijmegen, PO Box 9104, 6500 HE Nijmegen, The Netherlands October 8 2012 To obtain reliable measures researchers prefer multiple-item questionnaires rather than single-item tests. Multiple-item questionnaires may be costly however and time-consuming for participants to complete. They therefore frequently administer two-item measures, the reliability of which is commonly assessed by computing a reliability coefficient. There is some disagreement, however, what the most appropriate indicator of scale reliability is when a measure is composed of two items. The most frequently reported reliability statistic for multiple-item scales is Cronbach’s coefficient alpha and many researchers report this coefficient for their two-item measure

/pdf/the-reliability-of-a-two-item-scale-pearson-cronbach-or-18modapbyn.pdf

The reliability of a two-item scale: Pearson, Cronbach, or Spearman-Brown?

Background. Subthreshold depression is a highly prevalent condition and a risk factor for developing a major depressive episode. Internet-based cognitive behaviour therapy may be a promising approach for the treatment of subthreshold depression. The current study had two aims : (1) to determine whether an internet-based cognitive behaviour therapy intervention and a group cognitive behaviour therapy intervention are more effective than a waiting-list control group ; and (2) to determine whether the effect of the internet-based cognitive behaviour therapy differs from the group cognitive behaviour therapy intervention. Method. A total of 191 women and 110 men with subthreshold depression were randomized into internet-based treatment, group cognitive behaviour therapy (Lewinsohn's Coping With Depression course), or a waiting-list control condition. The main outcome measure was treatment response after 10 weeks, defined as the difference in pre- and post-treatment scores on the Beck Depression Inventory (BDI). Missing data, a major limitation of this study, were imputed using the Multiple Imputation (MI) procedure Data Augmentation. Results. In the waiting-list control group, we found a pre- to post-improvement effect size of 0 . 45, which was 0 . 65 in the group cognitive behaviour therapy condition and 1 . 00 within the internet- based treatment condition. Helmert contrasts showed a significant difference between the waiting- list condition and the two treatment conditions (p=0 . 04) and no significant difference between both treatment conditions (p=0 . 62). Conclusions. An internet-based intervention may be at least as effective as a commonly used group cognitive behaviour therapy intervention for subthreshold depression in people over 50 years of age.

/pdf/internet-based-cognitive-behavioural-therapy-for-15z5pffmzr.pdf

Internet-based cognitive behavioural therapy for subthreshold depression in people over 50 years old: a randomized controlled clinical trial

Background: Self-help therapies are often effective in reducing mental health problems. We developed a new Web-based self-help intervention based on problem-solving therapy, which may be used for people with different types of comorbid problems: depression, anxiety, and work-related stress.
Objective: The aim was to study whether a Web-based self-help intervention is effective in reducing depression, anxiety, and work-related stress (burnout).
Methods: A total of 213 participants were recruited through mass media and randomized to the intervention (n = 107) or a waiting list control group (n = 106). The Web-based course took 4 weeks. Every week an automated email was sent to the participants to explain the contents and exercises for the coming week. In addition, participants were supported by trained psychology students who offered feedback by email on the completed exercises. The core element of the intervention is a procedure in which the participants learn to approach solvable problems in a structured way. At pre-test and post-test, we measured the following primary outcomes: depression (CES-D and MDI), anxiety (SCL-A and HADS), and work-related stress (MBI). Quality of life (EQ-5D) was measured as a secondary outcome. Intention-to-treat analyses were performed.
Results: Of the 213 participants, 177 (83.1%) completed the baseline and follow-up questionnaires; missing data were statistically imputed. Of all 107 participants in the intervention group, 9% (n = 10) dropped out before the course started and 55% (n = 59) completed the whole course. Among all participants, the intervention was effective in reducing symptoms of depression (CES-D: Cohen’s d = 0.50, 95% confidence interval (CI) 0.22-0.79; MDI: d = 0.33, 95% CI 0.03-0.63) and anxiety (SCL-A: d = 0.42, 95% CI 0.14-0.70; HADS: d = 0.33, 95% CI 0.04-0.61) as well as in enhancing quality of life (d = 0.31, 95% CI 0.03-0.60). Moreover, a higher percentage of patients in the intervention group experienced a significant improvement in symptoms (CES-D: odds ratio [OR] = 3.5, 95% CI 1.9-6.7; MDI: OR = 3.7, 95% CI 1.4-10.0; SCL-A: OR = 2.1, 95% CI 1.0-4.6; HADS: OR = 3.1, 95% CI 1.6-6.0). Patients in the intervention group also recovered more often (MDI: OR = 2.2; SCL-A: OR = 2.0; HADS < 8), although these results were not statistically significant. The course was less effective for work-related stress, but participants in the intervention group recovered more often from burnout than those in the control group (OR = 4.0, 95% CI 1.2-13.5).
Conclusions: We demonstrated statistically and clinically significant effects on symptoms of depression and anxiety. These effects were even more pronounced among participants with more severe baseline problems and for participants who fully completed the course. The effects on work-related stress and quality of life were less clear. To our knowledge, this is the first trial of a Web-based, problem-solving intervention for people with different types of (comorbid) emotional problems. The results are promising, especially for symptoms of depression and anxiety. Further research is needed to enhance the effectiveness for work-related stress.
Trial Registration: International Standard Randomized Controlled Trial Number (ISRCTN) 14881571 [J Med Internet Res 2008;10(1):e7]

/pdf/effectiveness-of-a-web-based-self-help-intervention-for-rsrkkgr87f.pdf

Effectiveness of a Web-Based Self-Help Intervention for Symptoms of Depression, Anxiety, and Stress: Randomized Controlled Trial

The Mental Health Inventory (MHI)-5 is an attractive, brief screening questionnaire for depression and anxiety disorders. It has been suggested that the three questions on depression (MHI-d) may be as good as the full MHI-5 in assessing depressive disorders. We examined the validity of the MHI-d and the MHI-a (the remaining two items on anxiety) in a large population-based sample of 7076 adults in the Netherlands. We also examined the validity of the MHI in assessing specific anxiety disorders. The presence of depressive and anxiety disorders in the past month was assessed with the Composite International Diagnostic Interview (CIDI), computerized version 1.1. ROC analyses indicated no significant difference between the MHI-5 (area under the curve of 0.93) and the MHI-d (area under the curve of 0.91) in detecting major depression and dysthymia. There was no difference either between the MHI-5 (area under the curve of 0.73) and the MHI-a (area under the curve 0.73) in detecting anxiety disorders. Both the MHI-5 and the MHI-a also seem to be adequate as a screener for some anxiety disorders (generalized anxiety disorder; panic disorder; obsessive-compulsive disorder), but not others, especially phobias (agoraphobia; social phobia; simple phobia).

Screening for mood and anxiety disorders with the five-item, the three-item, and the two-item mental health inventory

Background. Internet-based cognitive behaviour therapy (CBT) is a promising new approach for the treatment of depressive symptoms. The current study had two aims: (1) to determine whether, after 1 year, an internet-based CBT intervention was more effective than a waiting-list control group; and (2) to determine whether the effect of the internet-based CBT differed from the group CBT intervention, 1 year after the start of treatment. Method. A total of 191 women and 110 men (mean age=55 years, s.d.=4.6) with subthreshold depression were randomized into internet-based treatment, group CBT (Lewinsohn's Coping with Depression Course), or a waiting-list control condition. The main outcome measure was treatment response after 1 year, defined as the difference in pretreatment and follow-up scores on the Beck Depression Inventory (BDI). Missing data were imputed using the multiple imputation procedure of data augmentation. Analyses were performed using multiple imputation inference. Results. In the waiting-list control group, we found a pretreatment to follow-up improvement effect size of 0.69, which was 0.62 in the group CBT condition and 1.22 with the internet-based treatment condition. Simple contrasts showed a significant difference between the waiting-list condition and internet-based treatment (p=0.03) and no difference between both treatment conditions (p=0.08). Conclusions. People aged over 50 years with subthreshold depression can still benefit from internet-based CBT 1 year after the start of treatment. © 2008 Cambridge University Press.

/pdf/one-year-follow-up-results-of-a-randomized-controlled-1lfonocjxa.pdf

One-year follow-up results of a randomized controlled clinical trial on internet-based cognitive behavioural therapy for subthreshold depression in people over 50 years

Identification of subgroups of patients for whom treatment A is more effective than treatment B, and vice versa, is of key importance to the development of personalized medicine. Tree-based algorithms are helpful tools for the detection of such interactions, but none of the available algorithms allow for taking into account clustered or nested dataset structures, which are particularly common in psychological research. Therefore, we propose the generalized linear mixed-effects model tree (GLMM tree) algorithm, which allows for the detection of treatment-subgroup interactions, while accounting for the clustered structure of a dataset. The algorithm uses model-based recursive partitioning to detect treatment-subgroup interactions, and a GLMM to estimate the random-effects parameters. In a simulation study, GLMM trees show higher accuracy in recovering treatment-subgroup interactions, higher predictive accuracy, and lower type II error rates than linear-model-based recursive partitioning and mixed-effects regression trees. Also, GLMM trees show somewhat higher predictive accuracy than linear mixed-effects models with pre-specified interaction effects, on average. We illustrate the application of GLMM trees on an individual patient-level data meta-analysis on treatments for depression. We conclude that GLMM trees are a promising exploratory tool for the detection of treatment-subgroup interactions in clustered datasets.

/pdf/detecting-treatment-subgroup-interactions-in-clustered-data-3icmnqw1ft.pdf

Niels Smits

Papers

Internet-based cognitive behavioural therapy for subthreshold depression in people over 50 years old: a randomized controlled clinical trial

Effectiveness of a Web-Based Self-Help Intervention for Symptoms of Depression, Anxiety, and Stress: Randomized Controlled Trial

Screening for mood and anxiety disorders with the five-item, the three-item, and the two-item mental health inventory

One-year follow-up results of a randomized controlled clinical trial on internet-based cognitive behavioural therapy for subthreshold depression in people over 50 years

Detecting treatment-subgroup interactions in clustered data with generalized linear mixed-effects model trees