Screening for anxiety and depression in community mental health: the beck anxiety and depression inventories

Identifying cases of common mental disorders in the context of HIV care in low and middle income countries

One of the key behavioural factors inhibiting adherence to antiretroviral therapy (ART) is the presence of common mental disorders (CMDs). Correct identification of CMDs can facilitate referral for treatment, the amelioration of symptoms, and consequently improved adherence to ART. To save time and resources, screening is an alternative to conducting diagnostic interviews in case identification. However, an elevated score on a screening instrument does not indicate caseness for a mental health condition, given poor sensitivity and low positive predictive values of many screeners. A large number of false positives means that many people would be incorrectly identified as having a mental health condition and inappropriately referred for treatment. A large number of false negatives means that people who actually require treatment will not be identified as such and may thus go untreated. Thus it is recommended that public health services in low resource countries consider a two-stage approach to screening. When implementing routine screening, only those persons who screen above a commonly used cut-point would undergo a diagnostic interview to determine the presence of a common mental disorder. True cases may then be referred for treatment where these are available, such as anti-depressive medication or psychological treatment.

Distress among cancer patients: some considerations in identification and treatment

This article discusses the merits of screening for distress in the context of a chronic illness, such as cancer, with a particular focus on resource-constrained health care systems such as those in low-and-middle income countries. Despite calls for distress to be considered a vital sign, like pain it is not objectively verifiable as it relies solely on the person’s subjective appraisal. Accordingly, the Distress Thermometer has limited validity considering its concordance with the Hospital Anxiety and Depression Scale, which itself has limitations in terms of its psychometric properties. Indeed, an elevated score on a self-report measure does not indicate caseness for a mental health condition. Distress is often self-limiting and transient, whereas common mental disorders require evidence-informed treatment. In the context of scarce resources as is the case in low-and-middle income countries, efforts should instead be directed at identifying common mental disorders among persons living with cancer and others who have serious health threats. Such an approach will increase the likelihood of resources being directed at those who are most likely to benefit from psychological interventions. Where persons living with cancer indicate the need for psychosocial services, ways to manage distress include problem-solving therapy, motivational interviewing, and mindfulness-based stress reduction.

Web-Based Skin Cancer Assessment and Classification Using Machine Learning and Mobile Computerized Adaptive Testing in a Rasch Model: Development Study

Background

Web-based computerized adaptive testing (CAT) implementation of the skin cancer (SC) risk scale could substantially reduce participant burden without compromising measurement precision. However, the CAT of SC classification has not been reported in academics thus far.

Objective

We aim to build a CAT-based model using machine learning to develop an app for automatic classification of SC to help patients assess the risk at an early stage.

Methods

We extracted data from a population-based Australian cohort study of SC risk (N=43,794) using the Rasch simulation scheme. All 30 feature items were calibrated using the Rasch partial credit model. A total of 1000 cases following a normal distribution (mean 0, SD 1) based on the item and threshold difficulties were simulated using three techniques of machine learning—naïve Bayes, k-nearest neighbors, and logistic regression—to compare the model accuracy in training and testing data sets with a proportion of 70:30, where the former was used to predict the latter. We calculated the sensitivity, specificity, receiver operating characteristic curve (area under the curve [AUC]), and CIs along with the accuracy and precision across the proposed models for comparison. An app that classifies the SC risk of the respondent was developed.

Results

We observed that the 30-item k-nearest neighbors model yielded higher AUC values of 99% and 91% for the 700 training and 300 testing cases, respectively, than its 2 counterparts using the hold-out validation but had lower AUC values of 85% (95% CI 83%-87%) in the k-fold cross-validation and that an app that predicts SC classification for patients was successfully developed and demonstrated in this study.

Conclusions

The 30-item SC prediction model, combined with the Rasch web-based CAT, is recommended for classifying SC in patients. An app we developed to help patients self-assess SC risk at an early stage is required for application in the future.

Predicting Posttraumatic Stress Disorder Caseness Using the PTSD Checklist for DSM-5 Among Patients Receiving Care for HIV

This study assessed the ability of the Posttraumatic Stress Disorder (PTSD) Checklist for the DSM-5 (PCL-5) to distinguish between caseness and noncaseness for PTSD among South Africans receiving care for HIV. The PCL-5 and the Structured Clinical Interview for DSM-5-Research Version (SCID-RV) module for PTSD were administered to 688 patients receiving antiretroviral therapy (ART) at two HIV care clinics in the greater Cape Town (South Africa) area. In total, nearly half of the sample (n = 324, 47.1%) reported experiencing an index traumatic event, and 101 participants (14.74%, 95% CI [12.17%, 17.62%]) met the diagnostic criteria for PTSD as measured using the SCID-RV. An ROC curve analysis suggested that a PCL-5 cutoff score of 32 yielded optimal sensitivity (0.88) and specificity (0.88), indicating that the measure was successful in determining caseness for PTSD 88% of the time and noncaseness 88% of the time. The AUC was 94.3%, 95% CI [92.6% to 96.1%], indicating high accuracy. The positive predictive value and negative predictive values were 56.3% and 97.7%, respectively, which suggests that the PCL-5 is an effective screening instrument to determine the presence of PTSD among South African ART users. Undetected and, thus, untreated PTSD may reduce quality of life, impede optimal adherence to ART, and increase the likelihood of risk behaviors among individuals living with HIV, contributing to further infections. The PCL-5 may be used for detection, referral, and treatment of PTSD as a way to enhance its management among individuals receiving HIV care.

Predicting Depression in Community Dwellers Using a Machine Learning Algorithm

Depression is one of the leading causes of disability worldwide. Given the socioeconomic burden of depression, appropriate depression screening for community dwellers is necessary. We used data from the 2014 and 2016 Korea National Health and Nutrition Examination Surveys. The 2014 dataset was used as a training set, whereas the 2016 dataset was used as the hold-out test set. The synthetic minority oversampling technique (SMOTE) was used to control for class imbalances between the depression and non-depression groups in the 2014 dataset. The least absolute shrinkage and selection operator (LASSO) was used for feature reduction and classifiers in the final model. Data obtained from 9488 participants were used for the machine learning process. The depression group had poorer socioeconomic, health, functional, and biological measures than the non-depression group. From the initial 37 variables, 13 were selected using LASSO. All performance measures were calculated based on the raw 2016 dataset without the SMOTE. The area under the receiver operating characteristic curve and overall accuracy in the hold-out test set were 0.903 and 0.828, respectively. Perceived stress had the strongest influence on the classifying model for depression. LASSO can be practically applied for depression screening of community dwellers with a few variables. Future studies are needed to develop a more efficient and accurate classification model for depression.

Optimizing screening for depression, anxiety disorders, and post-traumatic stress disorder in inpatient addiction treatment: A preliminary investigation

OBJECTIVE

Substance use disorders (SUD) are frequently comorbid with other psychiatric conditions, but a comprehensive diagnostic assessment is often not feasible clinically. Efficient psychometrically-validated screening tools exist for commonly comorbid conditions, but cutoff accuracies have typically not been evaluated in addiction treatment settings. This study examined the performance of several widely-used screening measures in relation to diagnostic status from a clinical interview to identify and validate cutoff scores in an inpatient SUD treatment setting.

METHOD

Participants were 99 patients in a large residential SUD treatment program in Ontario, Canada. Participants completed a screening battery, including the Patient Health Questionnaire - 9 (PHQ-9), Generalized Anxiety Disorder - 7 (GAD-7), and Post-Traumatic Stress Disorder Checklist-5 (PCL-5), and underwent a semi-structured diagnostic clinical interview. Receiver operating characteristic curves were used to determine optimal cutoff scores on the screening tool against the interview-based diagnosis.

RESULTS

Area under the curve (AUC) was statistically significant for all screens and were as follows: PHQ-9 = 0.70 (95% CI = 0.59-0.80), GAD-7 = 0.74 (95% CI = 0.63-0.84), and PCL-5 = 0.79 (95% CI = 0.66-0.91). The optimal accuracy cutoff scores based on sensitivity and specificity were: PHQ-9 ≥ 16, GAD-7 ≥ 9, the PCL-5 ≥ 42.

CONCLUSIONS

In general, the candidate screeners performed acceptably in this population. However, the optimal cutoff scores were notably higher than existing guidelines for depression and PTSD, potentially due to the general elevations in negative affectivity among individuals initiating SUD treatment. Further validation of these cutoff values is warranted.

PUBLIC HEALTH SIGNIFICANCE

This study provides modified screening cutoff scores for major depression, anxiety disorders, and post-traumatic stress disorder in addiction treatment settings.

Predicting caseness of major depressive disorder using the Center for Epidemiological Studies Depression Scale (CESD-R) among patients receiving HIV care

Patients receiving HIV care at two public health facilities in South Africa were assessed using the Centres for Epidemiological Studies Depression Scale Revised (CESD-R) and the Structured Clinical Interview for DSM5 to assess major depressive disorder. Of the 688 study participants, 229 (33.3%; CI = 29.8%; 36.9%) scored in the elevated range on the CESD-R and 170 (24.7%; CI = 21.5% to 28.1%) met the diagnostic criteria for major depressive disorder (MDD). ROC curve analysis indicated that a CESD-R cut-point of 26.5 (rounded to 27) yielded optimal sensitivity (0.81) and specificity (0.82) in determining caseness for MDD. Positive and negative predictive values of the CESD-R were 60.26% and 92.98%, respectively. The AUC was 0.89 (89%) (95% CI = 0.86-0.92) indicating moderate to high accuracy. For the purpose of routine screening in the context of HIV care, the CESD-R appears to hold promise in identifying cases of MDD. Those who screen positive may then undergo a formal diagnostic interview to determine whether they are true cases for MDD, and referred for treatment.

Gender Differences in the Utility of the Alcohol Use Disorder Identification Test in Screening for Alcohol Use Disorder Among HIV Test Seekers in South Africa

We evaluated the effectiveness of the Alcohol Use Disorder Identification Test (AUDIT) in screening for alcohol use disorder (AUD) among 500 men and women seeking HIV testing. Receiver operating characteristic (ROC) curve analysis was used to determine the utility of the AUDIT in discriminating between AUD caseness and non-caseness. For men, a cut-off score of 10 on the AUDIT predicted AUD with 81% sensitivity and 77% specificity. For women, a cut-off score of 7 yielded optimal sensitivity (82%) and specificity (82%). For men, the AUDIT yielded a positive predictive value (PPV) of 49% and a negative predictive value (NPV) of 94%; for women the PPV and NPV were 49 and 96%, respectively. While the AUDIT can be used to rapidly screen large numbers of men and women seeking HIV testing, the instrument's low PPV indicates that individuals who screen positive may need to undergo further evaluation to detect cases of AUD.

Evaluation of the Beck Anxiety Inventory in predicting generalised anxiety disorder among individuals seeking HIV testing in the Western Cape province, South Africa

Background

Routine anxiety screening is needed among HIV test seekers, given the lack of health-care professionals with the ability to identify individuals with generalised anxiety.

Aim

The aim of this study was to determine the effectiveness of the Beck anxiety inventory (BAI) in predicting caseness for generalised anxiety disorder (GAD) among persons seeking HIV testing, using the structured clinical interview for the DSM-5 (SCID-5) as the gold standard.

Setting

Five HIV testing sites in the Western Cape region of South Africa.

Method

We recruited 500 persons seeking HIV testing from five non-medical testing sites in the Western Cape, South Africa. We used receiver operating curve analysis to determine the optimal cut-off point on the BAI to discriminate between GAD caseness and non-caseness.

Results

3.4% of the sample met the DSM-5 criteria for a diagnosis of GAD. Using an optimal cut-off point of 21.5, the sensitivity and specificity of the BAI were 82% and 80%, respectively. The positive predictive value was 13%, while the negative predictive value was 99%.

Conclusion

Our data suggest that while the BAI may be used to screen for GAD, it is likely to yield a high number of false positives. A two-tiered method may be useful to mitigate against case over-identification. Thus, in a public health setting, persons screening positive on the BAI should receive a diagnostic interview to determine whether they are true cases for GAD. Within resource-constrained communities in South Africa, referral trajectories should be integrated with routine screening and HIV testing.

Utility of the Beck Depression Inventory in measuring major depression among individuals seeking HIV testing in the Western Cape, South Africa

The Beck Depression Inventory (BDI) is often used to screen individuals for symptoms of major depressive disorder (MDD). Yet, its effectiveness in correctly discriminating between MDD cases and non-cases among individuals seeking HIV testing has not been investigated. We report on the effectiveness of the BDI-I in predicting caseness for MDD with the Structured Clinical Interview for the DSM (SCID) as a gold standard. A total of 500 HIV test-seekers were recruited at five non-medical testing sites in the Western Cape, South Africa. Receiver operating characteristic curve analysis was used to determine the extent to which the screening instrument was able to discriminate between MDD caseness or non-caseness. The SCID-based prevalence of MDD was 14.4%. The BDI-I predicted MDD with 67% sensitivity and 67% specificity, with an area under the curve (AUC) of 77%. The positive and negative predictive values were 0.25 and 0.92, respectively. Even though the BDI-I is often used to screen large numbers of people for depression, especially in psychiatric and medical settings, its ability to predict MDD is limited. Persons screening positive for MDD may still require evaluation with a clinical interview by a trained professional to be diagnosed with depression.

Correspondence Between Self-Report Measures and Clinician Assessments of Psychopathology in Female Intimate Partner Violence Survivors

Intimate partner violence (IPV) has potentially severe and long-lasting mental health consequences for survivors, including elevated symptoms and diagnoses of posttraumatic stress disorder (PTSD), depression, and generalized anxiety disorder (GAD). The current study examined the relationship between three self-report measures of psychological distress and ratings obtained from the corresponding clinician-administered measures in women seeking assessment for mental health problems following IPV ( N = 185). PTSD symptoms were assessed using the self-report Impact of Event Scale-Revised (IES-R) and the interview-based Clinician-Administered PTSD Scale (CAPS). Depression symptoms were assessed using the self-report Beck Depression Inventory-II (BDI-II) and the depressive disorders sections from the clinician-administered Anxiety Disorders Interview Schedule-IV (ADIS-IV). Anxiety symptoms were assessed using the self-report Beck Anxiety Inventory (BAI) and the clinician-administered GAD section from the ADIS-IV. Results indicated that psychological distress was prevalent in the sample, with 27% receiving a PTSD diagnosis, 40% diagnosed with a depressive disorder, and 55% meeting criteria for GAD. Although each self-report measure was significantly and positively correlated with its corresponding clinician-administered measure, rates of diagnostic concordance were mixed. The BDI-II showed a high degree of agreement with the ADIS-IV depression section, but the IES-R and the CAPS were discordant at classifying PTSD. The BAI had acceptable sensitivity but poor specificity in relation to the ADIS-IV GAD section. These findings suggest that multiple assessment modalities should be considered when rating symptoms and estimating the prevalence of diagnoses among survivors of IPV.

Development of a Microsoft Excel tool for one-parameter Rasch model of continuous items: an application to a safety attitude survey

Background

Many continuous item responses (CIRs) are encountered in healthcare settings, but no one uses item response theory’s (IRT) probabilistic modeling to present graphical presentations for interpreting CIR results. A computer module that is programmed to deal with CIRs is required.

To present a computer module, validate it, and verify its usefulness in dealing with CIR data, and then to apply the model to real healthcare data in order to show how the CIR that can be applied to healthcare settings with an example regarding a safety attitude survey.

Methods

Using Microsoft Excel VBA (Visual Basic for Applications), we designed a computer module that minimizes the residuals and calculates model’s expected scores according to person responses across items. Rasch models based on a Wright map and on KIDMAP were demonstrated to interpret results of the safety attitude survey.

Results

The author-made CIR module yielded OUTFIT mean square (MNSQ) and person measures equivalent to those yielded by professional Rasch Winsteps software. The probabilistic modeling of the CIR module provides messages that are much more valuable to users and show the CIR advantage over classic test theory.

Conclusions

Because of advances in computer technology, healthcare users who are familiar to MS Excel can easily apply the study CIR module to deal with continuous variables to benefit comparisons of data with a logistic distribution and model fit statistics.

Electronic supplementary material

The online version of this article (doi:10.1186/s12874-016-0276-2) contains supplementary material, which is available to authorized users.

Simulation study of activities of daily living functions using online computerized adaptive testing

Background

Computer adaptive testing (CAT) of the activities of daily living (ADL) functions is required (i) to reveal the advantages of using an efficient and accurate estimation method, (ii) to determine the cutpoint for classifying ADL strata in patients with stroke, and (iii) to evaluate the feasibility of online CAT used in clinical settings for smartphones.

Methods

Normally standardized distributions of ADL measurements were simulated using item parameters from published papers. We retrieved item parameters of the combined Barthel Index and Frenchay Activities Index from the literature (the 23-item comprehensive ADL [CADL] and 34-item ADL scales) and simulated three 1000-person measures from a normal standard CAT distribution: [i] CADL (CADL-CAT), [ii] ADL (ADL-CAT), and [iii] NAT (Non-Adaptive Testing). The cutpoints of ADL person strata were determined using a norm-reference method. Maximum a posteriori estimation, expected a posteriori estimation, and maximum likelihood estimation (MAP) were used to compare the Pearson correlation coefficients and different number ratios of paired measures yielded by CAT and NAT. The number of items and the cutpoints for the scale were separately determined.

Results

We found that (i) correlation coefficients for the three CAT-estimated measures were 0.77 (CADL), 0.93 (Male ADL), and 0.93 (Female ADL) compared with their NAT counterparts. Different number ratios of person-paired measures between CAT and NAT for the three scales were all less than 5 %, indicating no difference exists between CAT and NAT. However, CAT might be 66 % more efficient than NAT. (ii) The estimated cutpoints of T scores (i.e., with a mean of 50 and a standard deviation of 10) were 45, 55, and 65 (e.g., separating person ADL function to four strata with not active, fairly active, active, and very active). (iii) An available-for-download online ADL-CAT APP for clinical practice was demonstrated.

Conclusions

An online ADL-CAT APP using the MAP method was created and used on smartphones to classify ADL strata in patients with stroke.

Electronic supplementary material

The online version of this article (doi:10.1186/s12911-016-0370-8) contains supplementary material, which is available to authorized users.

Improving Inpatient Surveys: Web-Based Computer Adaptive Testing Accessed via Mobile Phone QR Codes

Background

The National Health Service (NHS) 70-item inpatient questionnaire surveys inpatients on their perceptions of their hospitalization experience. However, it imposes more burden on the patient than other similar surveys. The literature shows that computerized adaptive testing (CAT) based on item response theory can help shorten the item length of a questionnaire without compromising its precision.

Objective

Our aim was to investigate whether CAT can be (1) efficient with item reduction and (2) used with quick response (QR) codes scanned by mobile phones.

Methods

After downloading the 2008 inpatient survey data from the Picker Institute Europe website and analyzing the difficulties of this 70-item questionnaire, we used an author-made Excel program using the Rasch partial credit model to simulate 1000 patients’ true scores followed by a standard normal distribution. The CAT was compared to two other scenarios of answering all items (AAI) and the randomized selection method (RSM), as we investigated item length (efficiency) and measurement accuracy. The author-made Web-based CAT program for gathering patient feedback was effectively accessed from mobile phones by scanning the QR code.

Results

We found that the CAT can be more efficient for patients answering questions (ie, fewer items to respond to) than either AAI or RSM without compromising its measurement accuracy. A Web-based CAT inpatient survey accessed by scanning a QR code on a mobile phone was viable for gathering inpatient satisfaction responses.

Conclusions

With advances in technology, patients can now be offered alternatives for providing feedback about hospitalization satisfaction. This Web-based CAT is a possible option in health care settings for reducing the number of survey items, as well as offering an innovative QR code access.

Estimating Skin Cancer Risk: Evaluating Mobile Computer-Adaptive Testing

Background

Response burden is a major detriment to questionnaire completion rates. Computer adaptive testing may offer advantages over non-adaptive testing, including reduction of numbers of items required for precise measurement.

Objective

Our aim was to compare the efficiency of non-adaptive (NAT) and computer adaptive testing (CAT) facilitated by Partial Credit Model (PCM)-derived calibration to estimate skin cancer risk.

Methods

We used a random sample from a population-based Australian cohort study of skin cancer risk (N=43,794). All 30 items of the skin cancer risk scale were calibrated with the Rasch PCM. A total of 1000 cases generated following a normal distribution (mean [SD] 0 [1]) were simulated using three Rasch models with three fixed-item (dichotomous, rating scale, and partial credit) scenarios, respectively. We calculated the comparative efficiency and precision of CAT and NAT (shortening of questionnaire length and the count difference number ratio less than 5% using independent t tests).

Results

We found that use of CAT led to smaller person standard error of the estimated measure than NAT, with substantially higher efficiency but no loss of precision, reducing response burden by 48%, 66%, and 66% for dichotomous, Rating Scale Model, and PCM models, respectively.

Conclusions

CAT-based administrations of the skin cancer risk scale could substantially reduce participant burden without compromising measurement precision. A mobile computer adaptive test was developed to help people efficiently assess their skin cancer risk.

Neurocognitive functioning in adolescents with eating disorders: a population-based study

INTRODUCTION

Neurocognitive deficits have been identified in eating disorders, including anorexia nervosa and bulimia nervosa. However, current data do not allow for firm conclusions regarding the nature or extent of these deficits. The current study aimed to evaluate neurocognitive functioning in a population-based sample of adolescents with and without eating disorders.

METHODS

Participants (N=669) were drawn from the Western Australian Pregnancy Cohort (Raine) Study. Cognitive testing was conducted using the computerised CogState assessment battery. Eating disorder symptoms were assessed using questions adapted from the Child Eating Disorder Examination and Eating Disorder Examination-Questionnaire. Adolescents who met full or partial criteria for a DSM-IV eating disorder (n=58) were compared to adolescents with no significant eating pathology (n=592).

RESULTS

The eating disorder sample showed impaired performance on measures of executive functioning, including global processing and set shifting, but performed better than control participants on measures of visual attention and vigilance.

CONCLUSIONS

This is the first study to evaluate neurocognitive functioning in a population-based sample of adolescents with eating disorders. Support is provided for weak central coherence and set-shifting difficulties early in the course of eating disorders. Research is needed to determine if these deficits precede and predict eating disorder onset.

Applying computerized adaptive testing to the Negative Acts Questionnaire-Revised: Rasch analysis of workplace bullying

Background

Workplace bullying is a prevalent problem in contemporary work places that has adverse effects on both the victims of bullying and organizations. With the rapid development of computer technology in recent years, there is an urgent need to prove whether item response theory–based computerized adaptive testing (CAT) can be applied to measure exposure to workplace bullying.

Objective

The purpose of this study was to evaluate the relative efficiency and measurement precision of a CAT-based test for hospital nurses compared to traditional nonadaptive testing (NAT). Under the preliminary conditions of a single domain derived from the scale, a CAT module bullying scale model with polytomously scored items is provided as an example for evaluation purposes.

Methods

A total of 300 nurses were recruited and responded to the 22-item Negative Acts Questionnaire-Revised (NAQ-R). All NAT (or CAT-selected) items were calibrated with the Rasch rating scale model and all respondents were randomly selected for a comparison of the advantages of CAT and NAT in efficiency and precision by paired t tests and the area under the receiver operating characteristic curve (AUROC).

Results

The NAQ-R is a unidimensional construct that can be applied to measure exposure to workplace bullying through CAT-based administration. Nursing measures derived from both tests (CAT and NAT) were highly correlated (r=.97) and their measurement precisions were not statistically different (P=.49) as expected. CAT required fewer items than NAT (an efficiency gain of 32%), suggesting a reduced burden for respondents. There were significant differences in work tenure between the 2 groups (bullied and nonbullied) at a cutoff point of 6 years at 1 worksite. An AUROC of 0.75 (95% CI 0.68-0.79) with logits greater than –4.2 (or >30 in summation) was defined as being highly likely bullied in a workplace.

Conclusions

With CAT-based administration of the NAQ-R for nurses, their burden was substantially reduced without compromising measurement precision.

Eating disorder symptom trajectories in adolescence: effects of time, participant sex, and early adolescent depressive symptoms

BACKGROUND

Adolescence is a period of developmental risk for eating disorders and eating disorder symptoms. This study aimed to describe the prevalence and trajectory of five core eating disorder behaviours (binge eating, purging, fasting, following strict dietary rules, and hard exercise for weight control) and a continuous index of dietary restraint and eating, weight and shape concerns, in a cohort of male and female adolescents followed from 14 to 20 years. It also aimed to determine the effect of early adolescent depressive symptoms on the prevalence and trajectory of these different eating disorder symptoms. Participants (N = 1,383; 49% male) were drawn from the Western Australian Pregnancy Cohort (Raine) Study, a prospective cohort study that has followed participants from pre-birth to age 20 years. An adapted version of the Eating Disorder Examination-Questionnaire was used to assess eating disorder symptoms at ages 14, 17 and 20 years. The Beck Depression Inventory for Youth was used to assess depressive symptoms at age 14. Longitudinal changes in the prevalence of eating disorder symptoms were tested using generalised estimating equations and linear mixed models.

RESULTS

Symptom trajectories varied according to the eating disorder symptom studied, participant sex, and the presence of depressive symptoms in early adolescence. For males, eating disorder symptoms tended to be stable (for purging, fasting and hard exercise) or decreasing (for binge eating and global symptom scores) from 14 to 17 years, and then stable to 20 years. For females, fasting and global symptom scores increased from age 14 to peak in prevalence at age 17. Rates of binge eating in females were stable from age 14 to age 17 and increased significantly thereafter, whilst rates of purging and hard exercise increased from age 14 to age 17, and then remained elevated through to age 20. Depressive symptoms at age 14 impacted on eating disorder symptom trajectories in females, but not in males.

CONCLUSIONS

Prevention, screening and intervention initiatives for adolescent eating disorders need to be tailored to gender and age. Purging behaviour appears to be an important target for work with early to middle adolescent females.

A web-based computerized adaptive testing (CAT) to assess patient perception in hospitalization

BACKGROUND

Many hospitals have adopted mobile nursing carts that can be easily rolled up to a patient's bedside to access charts and help nurses perform their rounds. However, few papers have reported data regarding the use of wireless computers on wheels (COW) at patients' bedsides to collect questionnaire-based information of their perception of hospitalization on discharge from the hospital.

OBJECTIVE

The purpose of this study was to evaluate the relative efficiency of computerized adaptive testing (CAT) and the precision of CAT-based measures of perceptions of hospitalized patients, as compared with those of nonadaptive testing (NAT). An Excel module of our CAT multicategory assessment is provided as an example.

METHOD

A total of 200 patients who were discharged from the hospital responded to the CAT-based 18-item inpatient perception questionnaire on COW. The numbers of question administrated were recorded and the responses were calibrated using the Rasch model. They were compared with those from NAT to show the advantage of CAT over NAT.

RESULTS

Patient measures derived from CAT and NAT were highly correlated (r = 0.98) and their measurement precisions were not statistically different (P = .14). CAT required fewer questions than NAT (an efficiency gain of 42%), suggesting a reduced burden for patients. There were no significant differences between groups in terms of gender and other demographic characteristics.

CONCLUSIONS

CAT-based administration of surveys of patient perception substantially reduced patient burden without compromising the precision of measuring patients' perceptions of hospitalization. The Excel module of animation-CAT on the wireless COW that we developed is recommended for use in hospitals.

Web-based computer adaptive assessment of individual perceptions of job satisfaction for hospital workplace employees

Background

To develop a web-based computer adaptive testing (CAT) application for efficiently collecting data regarding workers' perceptions of job satisfaction, we examined whether a 37-item Job Content Questionnaire (JCQ-37) could evaluate the job satisfaction of individual employees as a single construct.

Methods

The JCQ-37 makes data collection via CAT on the internet easy, viable and fast. A Rasch rating scale model was applied to analyze data from 300 randomly selected hospital employees who participated in job-satisfaction surveys in 2008 and 2009 via non-adaptive and computer-adaptive testing, respectively.

Results

Of the 37 items on the questionnaire, 24 items fit the model fairly well. Person-separation reliability for the 2008 surveys was 0.88. Measures from both years and item-8 job satisfaction for groups were successfully evaluated through item-by-item analyses by using t-test. Workers aged 26 - 35 felt that job satisfaction was significantly worse in 2009 than in 2008.

Conclusions

A Web-CAT developed in the present paper was shown to be more efficient than traditional computer-based or pen-and-paper assessments at collecting data regarding workers' perceptions of job content.

Detecting depression in medically ill patients: Comparative accuracy of four screening questionnaires and physicians' diagnoses in Spanish population

OBJECTIVE

The purpose of this study is to compare the diagnostic accuracy of four depression screening tools commonly used in patients with medical disorders, relative to a reference diagnostic standard-a structured psychiatric interview.

METHODS

The Depression in the Medically Ill-18 (DMI-18) questionnaire was administered to 167 patients with medical disorders; of those, 53 completed the Beck Depression Inventory for Primary Care (BDI-PC), 67 the Hospital Anxiety and Depression Scale (HADS), and 46 the Patient Health Questionnaire-9 (PHQ-9). The entire sample was also interviewed with a structured psychiatric interview conducted by a mental health professional. Sensitivity, specificity, likelihood ratios (LRs), and area under the curve (AUC) were calculated and compared for the different measures.

RESULTS

At their respective recommended cutoff points, sensitivities [95% confidence interval (CI)] were 86% (70-95), 82% (63-94), 93% (86-97), and 68% (47-85) for the HADS-D, BDI-PC, DMI-18, and PHQ-9, respectively, while specificities ranged from 72% (47-90) for BDI-PC to 89% (72-98) for PHQ-9. The sensitivities of DMI-18 were significantly higher compared to those of HADS-D (P=.045) and PHQ-9 (P=.01). The PHQ-9 questionnaire obtained the most favorable positive LR (6.35; 95% CI, 2.48-18.36). In contrast, the DMI-18 showed the best negative LR (0.09; 95% CI, 0.04-0.18). Areas under the curves (95% CI) ranged from 0.92 (0.83-1.02) to 0.84 (0.74-0.94). Statistically significant differences were found between the AUCs of the DMI-10 and the BDI-PC.

CONCLUSION

Our results suggest that all evaluated scales have acceptable abilities and can be used as screening instruments for depression in patients with medical disorders. The DMI stands out for its sensitivity.