Development of computerized adaptive testing (CAT) for the EORTC QLQ-C30 physical functioning dimension

Purpose

Computerized adaptive test (CAT) methods, based on item response theory (IRT), enable a patient-reported outcome instrument to be adapted to the individual patient while maintaining direct comparability of scores. The EORTC Quality of Life Group is developing a CAT version of the widely used EORTC QLQ-C30. We present the development and psychometric validation of the item pool for the first of the scales, physical functioning (PF).

Methods

Initial developments (including literature search and patient and expert evaluations) resulted in 56 candidate items. Responses to these items were collected from 1,176 patients with cancer from Denmark, France, Germany, Italy, Taiwan, and the United Kingdom. The items were evaluated with regard to psychometric properties.

Results

Evaluations showed that 31 of the items could be included in a unidimensional IRT model with acceptable fit and good content coverage, although the pool may lack items at the upper extreme (good PF). There were several findings of significant differential item functioning (DIF). However, the DIF findings appeared to have little impact on the PF estimation.

Conclusions

We have established an item pool for CAT measurement of PF and believe that this CAT instrument will clearly improve the EORTC measurement of PF.     

Psychometric characteristics of daily diaries for the Patient-Reported Outcomes Measurement Information System (PROMIS®): a preliminary investigation

Purpose

The Patient-Reported Outcomes (PRO) Measurement Information System (PROMIS^®) has developed assessment tools for numerous PROs, most using a 7-day recall format. We examined whether modifying the recall period for use in daily diary research would affect the psychometric characteristics of several PROMIS measures.

Methods

Daily versions of short-forms for three PROMIS domains (pain interference, fatigue, depression) were administered to a general population sample (n = 100) for 28 days. Analyses used multilevel item response theory (IRT) models. We examined differential item functioning (DIF) across recall periods by comparing the IRT parameters from the daily data with the PROMIS 7-day recall IRT parameters. Additionally, we examined whether the IRT parameters for day-to-day within-person changes are invariant to those for between-person (cross-sectional) differences in PROs.

Results

Dimensionality analyses of the daily data suggested a single dimension for each PRO domain, consistent with PROMIS instruments. One-third of the daily items showed uniform DIF when compared with PROMIS 7-day recall, but the impact of DIF on the scale level was minor. IRT parameters for within-person changes differed from between-person parameters for 3 depression items, which were more sensitive for measuring change than between-person differences, but not for pain interference and fatigue items. Notably, mean scores from daily diaries were significantly lower than the PROMIS 7-day recall norms.

Conclusions

The results provide initial evidence supporting the adaptation of PROMIS measures for daily diary research. However, scores from daily diaries cannot be directly interpreted on PROMIS norms established for 7-day recall.     

Measuring social health in the patient-reported outcomes measurement information system (PROMIS): item bank development and testing

Purpose

To develop a social health measurement framework, to test items in diverse populations and to develop item response theory (IRT) item banks.

Methods

A literature review guided framework development of Social Function and Social Relationships sub-domains. Items were revised based on patient feedback, and Social Function items were field-tested. Analyses included exploratory factor analysis (EFA), confirmatory factor analysis (CFA), two-parameter IRT modeling and evaluation of differential item functioning (DIF).

Results

The analytic sample included 956 general population respondents who answered 56 Ability to Participate and 56 Satisfaction with Participation items. EFA and CFA identified three Ability to Participate sub-domains. However, because of positive and negative wording, and content redundancy, many items did not fit the IRT model, so item banks do not yet exist. EFA, CFA and IRT identified two preliminary Satisfaction item banks. One item exhibited trivial age DIF.

Conclusion

After extensive item preparation and review, EFA-, CFA- and IRT-guided item banks help provide increased measurement precision and flexibility. Two Satisfaction short forms are available for use in research and clinical practice. This initial validation study resulted in revised item pools that are currently undergoing testing in new clinical samples and populations.     

PROMIS? Parent Proxy Report Scales: an item response theory analysis of the parent proxy report item banks

Objective

The objective of the present study is to describe the item response theory (IRT) analysis of the National Institutes of Health (NIH) Patient Reported Outcomes Measurement Information System (PROMIS^?) pediatric parent proxy-report item banks and the measurement properties of the new PROMIS^? Parent Proxy Report Scales for ages 8?C17?years.

Methods

Parent proxy-report items were written to parallel the pediatric self-report items. Test forms containing the items were completed by 1,548 parent?Cchild pairs. CCFA and IRT analyses of scale dimensionality and item local dependence, and IRT analyses of differential item functioning were conducted.

Results

Parent proxy-report item banks were developed and IRT parameters are provided. The recommended unidimensional short forms for the PROMIS^? Parent Proxy Report Scales are item sets that are subsets of the pediatric self-report short forms, setting aside items for which parent responses exhibit local dependence. Parent proxy-report demonstrated moderate to low agreement with pediatric self-report.

Conclusions

The study provides initial calibrations of the PROMIS^? parent proxy-report item banks and the creation of the PROMIS^? Parent Proxy-Report Scales. It is anticipated that these new scales will have application for pediatric populations in which pediatric self-report is not feasible.     

Assessment of response shift using two structural equation modeling techniques

Objective

To identify response shift using two structural equation modeling (SEM) techniques.

Study design and setting

Hypertensive patients (n = 909) with coronary artery disease (CAD) completed SF-36 surveys at both baseline and 1-year follow-up. Response shift was identified using Oort and Schmitt SEM techniques. The type of response shift linked to changes in various parameters of the SEM measurement model is defined differently for both SEM approaches. Effect sizes were calculated for the impact of response shift on the change of SF-36 domain scores when using the Oort approach.

Results

Both Oort and Schmitt SEM approaches identified response shift only in the SF-36 physical functioning (PF) scale. The effect size of recalibration on the change of PF domain scores when using the Oort approach was ?0.12.

Conclusion

This study showed that hypertensive patients with CAD experienced a response shift over a 1-year period. Both the SEM approaches identified response shift (uniform recalibration using the Oort approach and recalibration using the Schmitt approach); however, both approaches use different parameters to define and test response shift. We found that either the variation in analytic methods or the sample used may influence the identification and type of response shift.     

Physical function mitigates the adverse effects of being thin On mortality in a free-living older Taiwanese cohort

Objectives

To examine the significance of underweight and physical function as well as their interaction on mortality in the aged.

Design

Prospective cohort.

Setting

The Elderly Nutrition and Health Survey in Taiwan during 1999?C2000.

Participants

Total of 1435 representative free-living elders (739 men and 696 women).

Measurements

Body composition was assessed by various anthropometrics. Physical function score (PF, ranged 0?C100) was derived from the SF-36?. Death by December 31, 2006 was the outcome measure.

Results

After 7.9 (median: 7.0) years follow-up, 381 (223 men, 158 women) of 1435 eligible participants had died. Those with the lowest PF (<45) had 3.43 (hazards ratio (HR), 95% confidence interval (CI) = 2.20?C5.36) times the all-cause mortality risk of the highest PF (??58). Interactions for PF and BMI (P =0.02) and for PF and wrist circumference (P =0.09) on death were found after controlling for potential confounders. Jointly, compared to normal-BMI-highest-PF, the greatest HR for death occurred where BMI <18.5 kg/m2 was combined with the lowest-PF after covariate adjustments (HR = 8.67, 95% CI = 3.77?C20.0). Similarly, the lowest arm muscle circumference (MAMC)-PF had a HR of 5.22 compared to mid-MAMC-highest-PF. However, percent and absolute body fat, estimated by bioelectrical impedance, was comparable to non-sarcopenic individuals.

Conclusion

Thin elderly Taiwanese with sarcopenia, and less skeleton, are at the most risk of death, especially if physical function is limited.     

Establishing a common metric for self-reported pain: linking BPI Pain Interference and SF-36 Bodily Pain Subscale scores to the PROMIS Pain Interference metric

Purpose

The study purposes were to mathematically link scores of the Brief Pain Inventory Pain Interference Subscale and the Short Form-36 Bodily Pain Subscale (legacy pain interference measures) to the NIH Patient-Reported Outcome Measurement Information System (PROMIS^®) Pain Interference (PROMIS-PI) metric and evaluate results.

Methods

Linking was accomplished using both equipercentile and item response theory (IRT) methods. Item parameters for legacy items were estimated on the PROMIS-PI metric to allow for pattern scoring. Crosswalk tables also were developed that associated raw scores (summed or average) on legacy measures to PROMIS-PI scores. For each linking strategy, participants’ actual PROMIS-PI scores were compared to those predicted based on their legacy scores. To assess the impact of different sample sizes, we conducted random resampling with replacement across 10,000 replications with sample sizes of n = 25, 50, and 75.

Results

Analyses supported the assumption that all three scales were measuring similar constructs. IRT methods produced marginally better results than equipercentile linking. Accuracy of the links was substantially affected by sample size.

Conclusions

The linking tools (crosswalks and item parameter estimates) developed in this study are robust methods for estimating the PROMIS-PI scores of samples based on legacy measures. We recommend using pattern scoring for users who have the necessary software and score crosswalks for those who do not.

     

Factor analytic and item response theory evaluation of the Penn State Worry Questionnaire in women with cancer

Purpose

Cancer survivors frequently experience worry about a variety of topics, including fear of recurrence. However, general measures of worry still require examination of reliability for this vulnerable population. This study utilized modern psychometric methods to examine the reliability of a worry measure in women with breast or gynecologic cancer.

Methods

Women with cancer (n = 332) completed the 16-item Penn State Worry Questionnaire (PSWQ), which has an abbreviated 8-item version (PSWQ-A). Categorical confirmatory factor analysis (CCFA) was used to determine the factor structure and item response theory (IRT) was used to examine score reliability.

Results

CCFA supported a two-factor structure with 11 positively worded items and the 5 negatively worded items loading on different factors. IRT analysis of the 11 positively worded items showed that each was contributing meaningful information to the overall scores. The 11 positively worded items and the PSWQ-A produced the most reliable scores for levels of worry ranging from one θ below to two θ above the mean.

Conclusions

The 11 positively worded items of the PSWQ and the 8-item PSWQ-A were suitable for use in cancer patients while the full PSWQ was unsuitable due to inclusion of the negatively worded items. Future research should consider measuring worry when examining distress in cancer survivors.     

Validity of the Thai EQ-5D in an occupational population in Thailand

Purpose

To assess the construct validity of the Thai EuroQoL (EQ-5D) among an occupational population in Thailand.

Methods

Data were derived from a large cohort study among employees of the Electricity Generating Authority of Thailand. In 2008 and 2009, 4,850 participants completed the Thai EQ-5D and Short-Form 36 version 2 (SF-36v2). Thai preferences weights were used to convert EQ-5D health states into EQ-5D index scores. Construct validity of the Thai EQ-5D was examined by specifying and testing hypotheses about the relationships between the EQ-5D, SF-36v2, and participants’ demographic and medical characteristics.

Results

Construct validity of the Thai EQ-5D was supported by expected relationships with SF-36v2 scale and summary scores. For example, SF-36v2 scores on the mental health scale were much lower for participants who reported having problems on the EQ-5D anxiety/depression dimension compared to those reporting no problems (mean norm-based SF-36v2 scores: 52.9 vs. 41.8, p < 0.001). Additionally, reporting a problem in a given EQ-5D dimension was generally associated with lower SF-36v2 summary scores. The EQ-5D index score distinguished between groups of participants in the expected manner, on the basis of sex, age, education and self-reported health, thus providing evidence of known-groups validity.

Conclusion

The study demonstrated good construct validity of the Thai EQ-5D in a large occupational population in Thailand.     

Development and psychometric evaluation of the public health surveillance well-being scale

Purpose

To develop and psychometrically evaluate the brief Public Health Surveillance Well-Being Scale (PHS-WB) that captures mental, physical, and social components of well-being.

Methods

Using data from 5,399 HealthStyles survey respondents, we conducted bi-factor, item response theory, and differential item functioning analyses to examine the psychometric properties of a pool of 34 well-being items. Based on the statistical results and content considerations, we developed a brief 10-item well-being scale and assessed its construct validity through comparisons of demographic subgroups and correlations with measures of related constructs.

Results

Based on the bi-factor analyses, the items grouped into both an overall factor and individual domain-specific factors. The PHS-WB scale demonstrated good internal consistency (alpha?=?0.87) and correlated highly with scores for the entire item pool (r?=?0.94). The well-being scale scores differed as expected across demographic groups and correlated with global and domain-specific measures of similar constructs, supporting its construct validity.

Conclusion

The 10-item PHS-WB scale demonstrates good psychometric properties, and its high correlation with the item pool suggests minimal loss of information with the use of fewer items. The brief PHS-WB allows for well-being assessment on national surveys or in other situations where a longer form may not be feasible.     

Nutrition,Daily Walking and Resilience are Associated with Physical Function in the Oldest Old Men

Background

Maintaining good physical functioning in old age is of utmost importance for healthy and active aging. We examined physical function and associated factors in the oldest-old men.

Subjects and methods

The participants of this cross sectional analysis of a longitudinal study were the oldest old men( n=394, mean age 88 years, range 82–97 years) from the Helsinki Business Men cohort who responded to a postal health and nutrition survey in 2016. Physical function was defined using the respective subscale (Physical Function, PF) in the RAND-36 health-related quality of life (HRQoL) instrument. Resilience was measured with validated Finnish version of Resilience scale. Diet quality was assessed using Mediterranean diet adherence score (MeDi) and Diet quality index (DQI) which is designed to show adherence to Finnish dietary recommendations. Food and dietary intakes were retrieved using 3-day food records (obtained from a sub-group of the respondents). The participants were divided into quartiles corresponding to their PF scores and health and nutrition indicators were calculated into these PF quartiles. Furthermore, a linear regression model was used to determine factors associated with PF.

Results

PF quartiles were positively associated with lower age, daily walking habit, cognition, diet quality, resilience, alcohol use and negatively associated with blood glucose levels, weight loss, body weight (BW) and falls. Polyunsaturated to saturated fat ratio and berry intake were also associated with PF. In a linear regression model PF was positively associated (p <.001, adjusted R2 =.560) with MeDi, cognition, resilience, vitality (RAND-36), and negatively with age and BW..

Conclusions

MeDi, exercise, resilience, cognition, use of alcohol, fat quality and lower age were positively associated with PF in the oldest-old men. Weight loss, falls and interestingly BW were negatively associated with PF.

     

Predicting preference-based SF-6D index scores from the SF-8 health survey

Purpose

To develop and test functions for predicting the preference-based SF-6D index scores from the SF-8 health survey.

Methods

This study was a secondary analysis of data collected in a population health survey in which respondents (n = 7,529) completed both the SF-36 and the SF-8 questionnaires. We examined seven ordinary least-square estimators for their performance in predicting SF-6D scores from the SF-8 at both the individual and the group levels.

Results

In general, all functions performed similarly well in predicting SF-6D scores, and the predictions at the group level were better than predictions at the individual level. At the individual level, 42.5–51.5 % of prediction errors were smaller than the minimally important difference (MID) of the SF-6D scores, depending on the function specifications, while almost all prediction errors of the tested functions were smaller than the MID of SF-6D at the group level. At both individual and group levels, the tested functions predicted lower than actual scores at the higher end of the SF-6D scale.

Conclusions

Our study developed functions to generate preference-based SF-6D index scores from the SF-8 health survey, the first of its kind. Further research is needed to evaluate the performance and validity of the prediction functions.     

Psychometric properties of the PROMIS® pediatric scales: precision,stability, and comparison of different scoring and administration options

Objectives

The objectives of the present study are to investigate the precision of static (fixed-length) short forms versus computerized adaptive testing (CAT) administration, response pattern scoring versus summed score conversion, and test–retest reliability (stability) of the Patient-Reported Outcomes Measurement Information System (PROMIS^®) pediatric self-report scales measuring the latent constructs of depressive symptoms, anxiety, anger, pain interference, peer relationships, fatigue, mobility, upper extremity functioning, and asthma impact with polytomous items.

Methods

Participants (N = 331) between the ages of 8 and 17 were recruited from outpatient general pediatrics and subspecialty clinics. Of the 331 participants, 137 were diagnosed with asthma. Three scores based on item response theory (IRT) were computed for each respondent: CAT response pattern expected a posteriori estimates, short-form response pattern expected a posteriori estimates, and short-form summed score expected a posteriori estimates. Scores were also compared between participants with and without asthma. To examine test–retest reliability, 54 children were selected for retesting approximately 2 weeks after the first assessment.

Results

A short CAT (maximum 12 items with a standard error of 0.4) was found, on average, to be less precise than the static short forms. The CAT appears to have limited usefulness over and above what can be accomplished with the existing static short forms (8–10 items). Stability of the scale scores over a 2-week period was generally supported.

Conclusion

The study provides further information on the psychometric properties of the PROMIS pediatric scales and extends the previous IRT analyses to include precision estimates of dynamic versus static administration, test–retest reliability, and validity of administration across groups. Both the positive and negative aspects of using CAT versus short forms are highlighted.     

Difference in method of administration did not significantly impact item response: an IRT-based analysis from the Patient-Reported Outcomes Measurement Information System (PROMIS) initiative

Purpose

To test the impact of method of administration (MOA) on the measurement characteristics of items developed in the Patient-Reported Outcomes Measurement Information System (PROMIS).

Methods

Two non-overlapping parallel 8-item forms from each of three PROMIS domains (physical function, fatigue, and depression) were completed by 923 adults (age 18–89) with chronic obstructive pulmonary disease, depression, or rheumatoid arthritis. In a randomized cross-over design, subjects answered one form by interactive voice response (IVR) technology, paper questionnaire (PQ), personal digital assistant (PDA), or personal computer (PC) on the Internet, and a second form by PC, in the same administration. Structural invariance, equivalence of item responses, and measurement precision were evaluated using confirmatory factor analysis and item response theory methods.

Results

Multigroup confirmatory factor analysis supported equivalence of factor structure across MOA. Analyses by item response theory found no differences in item location parameters and strongly supported the equivalence of scores across MOA.

Conclusions

We found no statistically or clinically significant differences in score levels in IVR, PQ, or PDA administration as compared to PC. Availability of large item response theory-calibrated PROMIS item banks allowed for innovations in study design and analysis.     

Validity and reliability of Persian version of chronic liver disease questionnaire (CLDQ)

Purpose

The aim of this study is to test the psychometric properties of the Persian version of the Chronic Liver Disease Questionnaire (CLDQ) in Iranian candidates for liver transplantation.

Method

One hundred and fifty-five consecutive adult patients awaiting liver transplantation completed the Persian version of CLDQ and the short-form health survey (SF-36). The etiology of cirrhosis, Child Pugh classification and Model for End stage Liver Disease (MELD) scores were taken from medical records.

Results

The scaling success rate for convergent validity was 100% for all domains, and the success rate for item discriminant validity was 95.8% (139/145). The internal consistency (Cronbach ??) for the domains ranged from 0.65 to 0.89. Multitrait?Cmultimethod correlation matrix and factor analysis revealed that the CLDQ and SF-36 measure different constructs of quality of life.

Conclusion

The Persian version of the CLDQ, a disease-specific questionnaire for measuring health-related quality of life, is accepted by liver transplantation candidates with adequate reliability and validity. There is no significant correlation of Child Pugh classification and MELD score with quality of life.     

Validation of the Brazilian version of Chronic Liver Disease Questionnaire

Purpose

The aim of this study was to validate the Chronic Liver Disease Questionnaire (CLDQ) for use in Brazilian population.

Method

A total of 200 patients with chronic liver disease and varying disease severity answered a socio-demographic questionnaire, t CLDQ, and the Medical Outcome Study Short Form 36 (SF-36). Patients returned in 1–15 days to answer CLDQ again. The Cronbach’s alpha of the total CLDQ score was 0.95 and fluctuated between 0.69 and 0.83 in its six domains.

Results

The intra-class correlation between total CLDQ scores in two evaluations was 0.97 and in all domains was >0.93. CLDQ was moderately correlated with the SF-36, 0.63 (total CLDQ vs. vitality, SF-36), 0.62 (CLDQ and mental health, SF-36), 0.62 (preoccupation, CLDQ, vs. General Health, SF-36), 0.59 (fatigue, CLDQ, vs. vitality, SF-36), 0.59 (activity, CLDQ, vs. vitality, SF-36), and 0.59 (fatigue, CLDQ, vs. mental health, SF-36). The highest scores were found in non-cirrhotic group. Child A patients had higher average scores than Child B and C groups in all domains, while patients with MELD <15 scored higher than patients with MELD ≥15.

Conclusion

CLDQ-BR was validated in Brazilian population and was appropriate for use in patients with liver disease of different etiologies and degrees of severity.     

Development of a paper-and-pencil semi-adaptive questionnaire for 5 domains of health-related quality of life (PAT-5D-QOL)

Objectives

To develop a paper-and-pencil semi-adaptive test for 5 domains of health-related quality of life (PAT-5D-QOL) based on item response theory (IRT).

Methods

The questionnaire uses items from previously developed item banks for 5 domains: (1) walking, (2) handling objects, (3) daily activities, (4) pain or discomfort, and (5) feelings. For each domain, respondents are initially classified into 4 functional levels. Depending on the level, they are instructed to respond to a different set of 5 additional questions. IRT scores for each domain and overall health utility scores are obtained using a simple spreadsheet. The questions were selected using psychometric and conceptual criteria. The format of the questionnaire was developed through focus groups and cognitive interviews. Feasibility was tested in two population surveys. A simulation study was conducted to compare PAT-5D-QOL with a computerized adaptive test (CAT-5D-QOL) and a fixed questionnaire, developed from the same item banks, in terms of accuracy, bias, precision, and ceiling and floor effects.

Results

Close to 90 % of the participants in feasibility studies followed the skip instructions properly. In a simulation study, scores on PAT-5D-QOL for all domains tended to be more accurate, more precise, less biased, and less affected by a ceiling effect than scores on a fixed IRT-based questionnaire of the same length. PAT-5D-QOL was slightly inferior to a fully adaptive instrument.

Conclusions

PAT-5D-QOL is a novel, semi-adaptive, IRT-based measure of health-related quality of life with a broad range of potential applications.     

Reliability and validity of the English (Singapore) and Chinese (Singapore) versions of the Short-Form 36 version 2 in a multi-ethnic Urban Asian population in Singapore

Purpose

We aimed to evaluate the measurement properties of the Singapore English and Chinese versions of the Short-Form 36 version 2 (SF-36v2) Questionnaire, an improved version of the widely used SF-36, for assessing health-related quality of life (HRQoL) in a multi-ethnic urban Asian population in Singapore.

Methods

SF-36v2 scores and data on medical history, demographic and lifestyle factors from the Singapore Prospective Study Programme were analyzed. Convergent and divergent validity, internal consistency, floor and ceiling effects, known group validity and factor structure of the SF-36v2 were assessed for the English and Chinese versions, respectively.

Results

Complete data for 4,917 participants (45.8 %) out of 10,747 eligible individuals were analyzed (survey language: 4,115 English and 802 Chinese). Item-scale correlations exceeded 0.4 for all items of the English SF-36v2 and for all except one item of the Chinese SF-36v2 (bathe and dress: item-scale correlation: 0.36). In the English SF-36v2, Cronbach’s alpha exceeded 0.70 for all scales. In the Chinese SF-36v2, Cronbach’s alpha exceeded 0.7 on all scales except social functioning (Cronbach’s alpha: 0.68). For known groups validity, respondents with chronic medical conditions expectedly reported lower SF-36v2 score on most English and Chinese SF-36v2 scales. In confirmatory factor analysis, the Singapore three-component model was favored over the United States two-component and Japan three-component models.

Conclusions

The English and Chinese SF-36v2 are valid and reliable for assessing HRQoL among English and Chinese-speaking Singaporeans. Test–retest reliability and responsiveness of the English and Chinese SF-36v2 in Singapore remain to be evaluated.     

Comparison of direct-measured and derived short form six dimensions (SF-6D) health preference values among chronic hepatitis B patients

Purpose

The short form six dimensions (SF-6D) are derived from the SF-36 Health Survey with the intention that item data of the latter are often converted to a preference value, which was subsequently used in economic evaluations of interventions. The aim was to compare the equivalence and sensitivity of health preference values derived from the SF-36/SF-12 Health Surveys to that measured directly by the SF-6D for chronic hepatitis B (CHB) patients.

Methods

This was a secondary analysis of the SF-6D and SF-36 data from a baseline sample of 589 patients with CHB infection with different stages of liver diseases. Degree of agreement (equivalence) between direct-measured and derived SF-6D values was determined using spearman correlation and intra-class correlation. Sensitivity and discriminative power of different SF-6D values were compared by standardized effect size and relative efficiency (RE) statistics.

Results

Significant differences in the direct-measured or derived SF-6D preference values were found between CHB groups. Degree of agreement between SF-6D values was satisfactory. Direct-measured SF-6D was the most efficient, followed by SF-12-derived and the SF-36-derived was the least, based on the standardized effect size and the RE statistics. Sensitivity and discriminative power of direct-measured SF-6D were superior to derived SF-6D among people with different CHB health states.

Conclusions

Although direct-measured and derived SF-6D preference values had satisfactory sensitivity in discriminating between CHB groups, direct-measured SF-6D is the most sensitive and preferable method of obtaining health preference.     

Psychometric evaluation of the EORTC computerized adaptive test (CAT) fatigue item pool

Purpose

Fatigue is one of the most common symptoms associated with cancer and its treatment. To obtain a more precise and flexible measure of fatigue, the EORTC Quality of Life Group has developed a computerized adaptive test (CAT) measure of fatigue. This is part of an ongoing project developing a CAT version of the widely used EORTC QLQ-C30 questionnaire.

Methods

Based on the literature search and evaluations by experts and patients, 41 new fatigue items were developed (in addition to the three QLQ-C30 fatigue items). Psychometric properties of the items, including evaluations of dimensionality, fit to item response theory (IRT) model, and differential item functioning (DIF), were assessed in an international sample of cancer patients.

Results

Responses were obtained from 1,321 cancer patients coming from eight countries. Factor analysis showed that 37 of the items could be included in a unidimensional model (RMSEA = 0.098, TLI = 0.995, CFI = 0.920). Of the 37 items, two were deleted because of poor fit to the IRT model forming the basis for the CAT, and one because of DIF between cancer sites.

Conclusions

We have established a 34-item fatigue bank allowing for more precise and flexible measurement of fatigue, while still being backward compatible with the QLQ-C30 fatigue scale.