Overlap of Depressive Symptoms with Health-Related Quality-of-Life Measures

Because depressive symptoms are a part of health-related quality-of-life (HRQOL) measures, measures of depression will be empirically associated with HRQOL. We discuss examples of published research where authors ignored or did not fully account for overlap between depressive symptom and HRQOL measures. Future researchers need to recognize when their models include conceptually similar variables on the same side or both sides of the equation. This awareness will lead to more accurate conclusions about the prognostic value of depression and other HRQOL measures for health care utilization, mortality, and other outcomes. It will also result in fewer incorrect claims about the effect of depression on HRQOL.

Do the unlabeled response categories of the Minnesota Living with Heart Failure Questionnaire satisfy the monotonicity assumption of simple-summated scoring?

PURPOSE

Half of the 21-item Minnesota Living with Heart Failure Questionnaire (MLHFQ) response categories are labeled (0 = No, 1 = Very little, 5 = Very much) and half are not (2, 3, and 4). We hypothesized that the unlabeled response options would not be more likely to be chosen at some place along the scale continuum than other response options and, therefore, not satisfy the monotonicity assumption of simple-summated scoring.

METHODS

We performed exploratory and confirmatory factor analyses of the MLHFQ items in a sample of 1437 adults in the Better Effectiveness After Transition-Heart Failure study. We evaluated the unlabeled response options using item characteristic curves from item response theory-graded response models for MLHFQ physical and emotional health scales. Then, we examined the impact of collapsing response options on correlations of scale scores with other variables.

RESULTS

The sample was 46% female; 71% aged 65 or older; 11% Hispanic, 22% Black, 54% White, and 12% other. The unlabeled response options were rarely chosen. The standard approach to scoring and scores obtained by collapsing adjacent response categories yielded similar associations with other variables, indicating that the existing response options are problematic.

CONCLUSIONS

The unlabeled MLHFQ response options do not meet the assumptions of simple-summated scoring. Further assessment of the performance of the unlabeled response options and evaluation of alternative scoring approaches is recommended. Adding labels for response options in future administrations of the MLHFQ should be considered.

General population normative data for the EORTC QLQ-C30 health-related quality of life questionnaire based on 15,386 persons across 13 European countries, Canada and the Unites States

OBJECTIVE

The European Organisation for Research and Treatment of Cancer (EORTC) QLQ-C30 health-related quality of life questionnaire is one of the most widely used cancer-specific health-related quality of life questionnaires worldwide. General population norm data can facilitate the interpretation of QLQ-C30 data obtained from cancer patients. This study aimed at systematically collecting norm data from the general population to develop European QLQ-C30 norm scores and to generate comparable norm data for individual countries in Europe and North America.

METHODS

We collected QLQ-C30 data from the general population across 11 European Union (EU) countries, Russia, Turkey, Canada and United States (n ≥ 1000/country). Representative samples were stratified by sex and age groups (18-39, 40-49, 50-59, 60-69 and ≥ 70 years). After applying weights based on the United Nations population distribution statistics, we calculated QLQ-C30 domain scores to generate a 'European QLQ-C30 Norm' based on the EU countries. Further, we calculated QLQ-C30 norm scores for all 15 individual countries.

RESULTS

A total of 15,386 respondents completed the online survey. For the EU sample, most QLQ-C30 domains showed differences by sex/age, with men scoring somewhat better health than women, while age effects varied across domains. Substantially larger differences were seen in inter-country comparisons, with Austrian and Dutch respondents reporting consistently better health compared with British and Polish respondents.

CONCLUSIONS

This study is the first to systematically collect EORTC QLQ-C30 general population norm data across Europe and North America applying a consistent data collection method across 15 countries. These new norm data facilitate valid intra-country as well as inter-country comparisons and QLQ-C30 score interpretation.

Australian Utility Weights for the EORTC QLU-C10D, a Multi-Attribute Utility Instrument Derived from the Cancer-Specific Quality of Life Questionnaire, EORTC QLQ-C30

BACKGROUND

The EORTC QLU-C10D is a new multi-attribute utility instrument derived from the widely used cancer-specific quality-of-life (QOL) questionnaire, EORTC QLQ-C30. The QLU-C10D contains ten dimensions (Physical, Role, Social and Emotional Functioning; Pain, Fatigue, Sleep, Appetite, Nausea, Bowel Problems), each with four levels. To be used in cost-utility analysis, country-specific valuation sets are required.

OBJECTIVE

The aim of this study was to provide Australian utility weights for the QLU-C10D.

METHODS

An Australian online panel was quota-sampled to ensure population representativeness by sex and age (≥ 18 years). Participants completed a discrete choice experiment (DCE) consisting of 16 choice-pairs. Each pair comprised two QLU-C10D health states plus life expectancy. Data were analysed using conditional logistic regression, parameterised to fit the quality-adjusted life-year framework. Utility weights were calculated as the ratio of each QOL dimension-level coefficient to the coefficient on life expectancy.

RESULTS

A total of 1979 panel members opted in, 1904 (96%) completed at least one choice-pair, and 1846 (93%) completed all 16 choice-pairs. Dimension weights were generally monotonic: poorer levels within each dimension were generally associated with greater utility decrements. The dimensions that impacted most on choice were, in order, Physical Functioning, Pain, Role Functioning and Emotional Functioning. Oncology-relevant dimensions with moderate impact were Nausea and Bowel Problems. Fatigue, Trouble Sleeping and Appetite had relatively small impact. The value of the worst health state was -0.096, somewhat worse than death.

CONCLUSIONS

This study provides the first country-specific value set for the QLU-C10D, which can facilitate cost-utility analyses when applied to data collected with the EORTC QLQ-C30, prospectively and retrospectively.

Sensitivity and responsiveness of the patient-reported TED-QOL to rehabilitative surgery in thyroid eye disease

We tested the sensitivity and responsiveness of the TED-QOL to rehabilitative surgery in thyroid eye disease (TED). The 3-item TED-QOL and 16-item GO-QOL, which assess quality of life (QoL) in TED, were administered to consecutive patients undergoing rehabilitative surgery. The questionnaires were completed pre-and post-operatively to assess sensitivity (ability to discriminate between different surgical groups) and responsiveness (ability to detect within patient changes over time).56 patients underwent 69 procedures for TED (29 orbital decompressions, 15 strabismus operations, 25 eyelid procedures). The differences in scores between the three types of surgery (a measure of sensitivity) were statistically significant at the 5% level pre-operatively and post-operatively for all 3 TED-QOL scales and for both GO-QOL scales, but much more so for the TED-QOL scales in each case. The within-patient changes between the pre- and post-operative scores for the same subjects (a measure of responsiveness) were statistically very highly significant for the TED-QOL overall and appearance scales for each of the surgeries. The pre- and post-operative difference for the TED-QOL functioning scale was highly statistically significant for strabismus surgery but not for decompression or lid surgery. The change between the pre- and post-operative scores for the GO-QOL was significant for the functioning scale with strabismus and lid surgery, and was highly significant for the appearance scale with lid surgery but not for strabismus surgery or decompression. The 3-item TED-QOL is sensitive and responsive to rehabilitative surgery in TED and compares favorably with the lengthier GO-QOL for these parameters.

The Edmonton Symptom Assessment System: Poor performance as screener for major depression in patients with incurable cancer

BACKGROUND

Depressive symptoms are prevalent in patients with advanced cancer, sometimes of a severity that fulfil the criteria for a major depressive episode.

AIM

The aim of this study was to investigate how the item on depression in the Edmonton Symptom Assessment System with a 0-10 Numerical Rating Scale performed as a screener for major depressive episode. A possible improved performance by adding the Edmonton Symptom Assessment System-Anxiety item was also examined.

DESIGN

An international cross-sectional study including patients with incurable cancer was conducted. The Edmonton Symptom Assessment System score was compared against major depressive episode as assessed by the Patient Health Questionnaire-9. Screening performance was examined by sensitivity, specificity and the kappa coefficient.

SETTING

Patients with incurable cancer (n = 969), median age 63 years and from eight nationalities provided report. Median Karnofsky Performance Status was 70. Median survival was 229 days (205-255 days).

RESULTS

Patient Health Questionnaire-9 major depressive episode was present in 133 of 969 patients (13.7%). Edmonton Symptom Assessment System-Depression screening ability for Patient Health Questionnaire-9 major depressive episode was limited. Area under the receiver operating characteristic curve was 0.71 (0.66-0.76). Valid detection or exclusion of Patient Health Questionnaire-9 major depressive episode could not be concluded at any Edmonton Symptom Assessment System-Depression cut-off; by the cut-off Numerical Rating Scale ⩾ 2, sensitivity was 0.69 and specificity was 0.60. By the cut-off Numerical Rating Scale ⩾ 4, sensitivity was 0.51 and specificity was 0.82. Combined mean ratings by Edmonton Symptom Assessment System-Depression and Edmonton Symptom Assessment System-Anxiety revealed similar limited screening ability.

CONCLUSION

The depression and anxiety items of the Edmonton Symptom Assessment System, a frequently used assessment tool in palliative care settings, seem to measure a construct other than major depressive episode as assessed by the Patient Health Questionnaire-9 instrument.

QLU-C10D: a health state classification system for a multi-attribute utility measure based on the EORTC QLQ-C30

PURPOSE

To derive a health state classification system (HSCS) from the cancer-specific quality of life questionnaire, the EORTC QLQ-C30, as the basis for a multi-attribute utility instrument.

METHODS

The conceptual model for the HSCS was based on the established domain structure of the QLQ-C30. Several criteria were considered to select a subset of dimensions and items for the HSCS. Expert opinion and patient input informed a priori selection of key dimensions. Psychometric criteria were assessed via secondary analysis of a pooled dataset comprising HRQOL and clinical data from 2616 patients from eight countries and a range of primary cancer sites, disease stages, and treatments. We used confirmatory factor analysis (CFA) to assess the conceptual model's robustness and generalisability. We assessed item floor effects (>75 % observations at lowest score), disordered item response thresholds, coverage of the latent variable and differential item function using Rasch analysis. We calculated effect sizes for known group comparisons based on disease stage and responsiveness to change. Seventy-nine cancer patients assessed the relative importance of items within dimensions.

RESULTS

CFA supported the conceptual model and its generalisability across primary cancer sites. After considering all criteria, 12 items were selected representing 10 dimensions: physical functioning (mobility), role functioning, social functioning, emotional functioning, pain, fatigue, sleep, appetite, nausea, bowel problems.

CONCLUSIONS

The HSCS created from QLQ-C30 items is known as the EORTC Quality of Life Utility Measure-Core 10 dimensions (QLU-C10D). The next phase of the QLU-C10D's development involves valuation studies, currently planned or being conducted across the globe.

Symptom Clusters in Advanced Cancer Patients: An Empirical Comparison of Statistical Methods and the Impact on Quality of Life

CONTEXT

Symptom clusters in advanced cancer can influence patient outcomes. There is large heterogeneity in the methods used to identify symptom clusters.

OBJECTIVES

To investigate the consistency of symptom cluster composition in advanced cancer patients using different statistical methodologies for all patients across five primary cancer sites, and to examine which clusters predict functional status, a global assessment of health and global quality of life.

METHODS

Principal component analysis and exploratory factor analysis (with different rotation and factor selection methods) and hierarchical cluster analysis (with different linkage and similarity measures) were used on a data set of 1562 advanced cancer patients who completed the European Organization for the Research and Treatment of Cancer Quality of Life Questionnaire-Core 30.

RESULTS

Four clusters consistently formed for many of the methods and cancer sites: tense-worry-irritable-depressed (emotional cluster), fatigue-pain, nausea-vomiting, and concentration-memory (cognitive cluster). The emotional cluster was a stronger predictor of overall quality of life than the other clusters. Fatigue-pain was a stronger predictor of overall health than the other clusters. The cognitive cluster and fatigue-pain predicted physical functioning, role functioning, and social functioning.

CONCLUSIONS

The four identified symptom clusters were consistent across statistical methods and cancer types, although there were some noteworthy differences. Statistical derivation of symptom clusters is in need of greater methodological guidance. A psychosocial pathway in the management of symptom clusters may improve quality of life. Biological mechanisms underpinning symptom clusters need to be delineated by future research. A framework for evidence-based screening, assessment, treatment, and follow-up of symptom clusters in advanced cancer is essential.

Using a discrete choice experiment to value the QLU-C10D: feasibility and sensitivity to presentation format

PURPOSE

To assess the feasibility of using a discrete choice experiment (DCE) to value health states within the QLU-C10D, a utility instrument derived from the QLQ-C30, and to assess clarity, difficulty, and respondent preference between two presentation formats.

METHODS

We ran a DCE valuation task in an online panel (N = 430). Respondents answered 16 choice pairs; in half of these, differences between dimensions were highlighted, and in the remainder, common dimensions were described in text and differing attributes were tabulated. To simplify the cognitive task, only four of the QLU-C10D's ten dimensions differed per choice set. We assessed difficulty and clarity of the valuation task with Likert-type scales, and respondents were asked which format they preferred. We analysed the DCE data by format with a conditional logit model and used Chi-squared tests to compare other responses by format. Semi-structured telephone interviews (N = 8) explored respondents' cognitive approaches to the valuation task.

RESULTS

Four hundred and forty-nine individuals were recruited, 430 completed at least one choice set, and 422/449 (94 %) completed all 16 choice sets. Interviews revealed that respondents found ten domains difficult but manageable, many adopting simplifying heuristics. Results for clarity and difficulty were identical between formats, but the "highlight" format was preferred by 68 % of respondents. Conditional logit parameter estimates were monotonic within domains, suggesting respondents were able to complete the DCE sensibly, yielding valid results.

CONCLUSION

A DCE valuation task in which only four of the QLU-C10D's ten dimensions differed in any choice set is feasible for deriving utility weights for the QLU-C10D.

Deriving a preference-based utility measure for cancer patients from the European Organisation for the Research and Treatment of Cancer's Quality of Life Questionnaire C30: a confirmatory versus exploratory approach

Background

Multi attribute utility instruments (MAUIs) are preference-based measures that comprise a health state classification system (HSCS) and a scoring algorithm that assigns a utility value to each health state in the HSCS. When developing a MAUI from a health-related quality of life (HRQOL) questionnaire, first a HSCS must be derived. This typically involves selecting a subset of domains and items because HRQOL questionnaires typically have too many items to be amendable to the valuation task required to develop the scoring algorithm for a MAUI. Currently, exploratory factor analysis (EFA) followed by Rasch analysis is recommended for deriving a MAUI from a HRQOL measure.

Aim

To determine whether confirmatory factor analysis (CFA) is more appropriate and efficient than EFA to derive a HSCS from the European Organisation for the Research and Treatment of Cancer’s core HRQOL questionnaire, Quality of Life Questionnaire (QLQ-C30), given its well-established domain structure.

Methods

QLQ-C30 (Version 3) data were collected from 356 patients receiving palliative radiotherapy for recurrent/metastatic cancer (various primary sites). The dimensional structure of the QLQ-C30 was tested with EFA and CFA, the latter informed by the established QLQ-C30 structure and views of both patients and clinicians on which are the most relevant items. Dimensions determined by EFA or CFA were then subjected to Rasch analysis.

Results

CFA results generally supported the proposed QLQ-C30 structure (comparative fit index =0.99, Tucker–Lewis index =0.99, root mean square error of approximation =0.04). EFA revealed fewer factors and some items cross-loaded on multiple factors. Further assessment of dimensionality with Rasch analysis allowed better alignment of the EFA dimensions with those detected by CFA.

Conclusion

CFA was more appropriate and efficient than EFA in producing clinically interpretable results for the HSCS for a proposed new cancer-specific MAUI. Our findings suggest that CFA should be recommended generally when deriving a preference-based measure from a HRQOL measure that has an established domain structure.

Is progression-free survival associated with a better health-related quality of life in patients with lung cancer? Evidence from two randomised trials with afatinib

OBJECTIVE

Progression-free survival (PFS) is frequently used as an efficacy end point in oncology clinical trials. However, there is limited evidence to support a positive association between improvement in PFS and improvement in health-related quality of life (HRQoL). The association between PFS and HRQoL was evaluated in two randomised trials.

MATERIALS AND METHODS

Data from two randomised controlled trials in patients with non-small cell lung cancer (NSCLC; LUX-Lung 1 and LUX-Lung 3) were used to investigate HRQoL in patients to determine whether tumour progression is accompanied by worsening HRQoL. HRQoL was assessed using the cancer-specific European Organization for Research and Treatment of Cancer (EORTC) core questionnaire QLQ-C30, the EuroQol EQ-5D overall utility and EuroQol EQ visual analogue scale. In both studies, progression was evaluated by independent review using RECIST criteria (primary end point) and also by investigator assessment. The relationship between tumour progression and HRQoL was evaluated using analysis of covariance and a longitudinal model.

RESULTS

Compliance with HRQoL questionnaire completion was high. In both studies, patients with progression consistently experienced numerically poorer HRQoL at the time of progression than patients without progression. Differences in mean scores were statistically significant (p<0.05) between patients with and without progression at week 4 in all analyses in LUX-Lung 1 and at multiple time points in LUX-Lung 3. Results from the longitudinal analysis showed that progression (by independent review and investigator assessment) appears to have consistent negative impact on all three HRQoL measures (all p<0.0001).

CONCLUSIONS

Tumour progression in patients with NSCLC was associated with statistically significant worsening in HRQoL. These findings confirm the value of PFS as a patient-relevant end point.

Use of methods for specifying the target difference in randomised controlled trial sample size calculations: Two surveys of trialists' practice

BACKGROUND

Central to the design of a randomised controlled trial (RCT) is a calculation of the number of participants needed. This is typically achieved by specifying a target difference, which enables the trial to identify a difference of a particular magnitude should one exist. Seven methods have been proposed for formally determining what the target difference should be. However, in practice, it may be driven by convenience or some other informal basis. It is unclear how aware the trialist community is of these formal methods or whether they are used.

PURPOSE

To determine current practice regarding the specification of the target difference by surveying trialists.

METHODS

Two surveys were conducted: (1) Members of the Society for Clinical Trials (SCT): participants were invited to complete an online survey through the society's email distribution list. Respondents were asked about their awareness, use of, and willingness to recommend methods; (2) Leading UK- and Ireland-based trialists: the survey was sent to UK Clinical Research Collaboration registered Clinical Trials Units, Medical Research Council UK Hubs for Trial Methodology Research, and the Research Design Services of the National Institute for Health Research. This survey also included questions about the most recent trial developed by the respondent's group.

RESULTS

Survey 1: Of the 1182 members on the SCT membership email distribution list, 180 responses were received (15%). Awareness of methods ranged from 69 (38%) for health economic methods to 162 (90%) for pilot study. Willingness to recommend among those who had used a particular method ranged from 56% for the opinion-seeking method to 89% for the review of evidence-base method. Survey 2: Of the 61 surveys sent out, 34 (56%) responses were received. Awareness of methods ranged from 33 (97%) for the review of evidence-base and pilot methods to 14 (41%) for the distribution method. The highest level of willingness to recommend among users was for the anchor method (87%). Based upon the most recent trial, the target difference was usually one viewed as important by a stakeholder group, mostly also viewed as a realistic difference given the interventions under evaluation, and sometimes one that led to an achievable sample size.

LIMITATIONS

The response rates achieved were relatively low despite the surveys being short, well presented, and having utilised reminders.

CONCLUSION

Substantial variations in practice exist with awareness, use, and willingness to recommend methods varying substantially. The findings support the view that sample size calculation is a more complex process than would appear to be the case from trial reports and protocols. Guidance on approaches for sample size estimation may increase both awareness and use of appropriate formal methods.

Weight loss, appetite loss and food intake in cancer patients with cancer cachexia: three peas in a pod? - analysis from a multicenter cross sectional study

BACKGROUND

How to assess cachexia is a barrier both in research and in clinical practice. This study examines the need for assessing both reduced food intake and loss of appetite, to see if these variables can be used interchangeably. A secondary aim is to assess the variance explained by food intake, appetite and weight loss by using tumor-related factors, symptoms and biological markers as explanatory variables.

MATERIAL AND METHODS

One thousand and seventy patients with incurable cancer were registered in an observational, cross sectional multicenter study. A total of 885 patients that had complete data on food intake (PG-SGA), appetite (EORTC QLQ-C30) and weight loss were included in the present analysis. The association between reduced food intake and appetite loss was assessed using Spearman's correlation. To find the explained variance of the three symptoms a multivariate analysis was performed.

RESULTS

The mean age was 62 years with a mean survival of 247 days and a mean Karnofsky performance status of 72. Thirteen percent of the patients who reported eating less than normal had good appetite and 25% who had unchanged or increased food intake had reduced appetite. Correlation between appetite loss and food intake was 0.50. Explained variance for the regression models was 44% for appetite loss, 27% for food intake and only 13% for weight loss.

CONCLUSION

Both appetite loss and food intake should be assessed in cachectic patients since conscious control of eating may sometimes overcome appetite loss. The low explained variance for weight loss is probably caused by the need for more knowledge about metabolism and inflammation, and is consistent with the cancer cachexia definition that claims that in cachexia weight loss is not caused by reduced food intake alone. The questions concerning appetite loss from EORTC-QLQ C30 and food intake from PG-SGA seem practical and informative when dealing with advanced cancer patients.

Should linking replace regression when mapping from profile-based measures to preference-based measures?

BACKGROUND

Profile instruments are frequently used to assess health-related quality of life and other patient-reported outcomes. However, preference-based measures are required for health-economic cost-utility evaluations.

RESULTS

Although regression-based approaches are commonly used to map from profile measures to preference measures, we show that this results in biased estimates because of regression to the mean.

CONCLUSIONS

Linking (scale-aligning) is proposed as an alternative.

Don't middle your MIDs: regression to the mean shrinks estimates of minimally important differences

Minimal important differences (MIDs) for patient-reported outcomes (PROs) are often estimated by selecting a clinical variable to serve as an anchor. Then, differences in the clinical anchor regarded as clinically meaningful or important can be used to estimate the corresponding value of the PRO. Although these MID values are sometimes estimated by regression techniques, we show that this is a biased procedure and should not be used; alternative methods are proposed.

ISOQOL recommends minimum standards for patient-reported outcome measures used in patient-centered outcomes and comparative effectiveness research

PURPOSE

An essential aspect of patient-centered outcomes research (PCOR) and comparative effectiveness research (CER) is the integration of patient perspectives and experiences with clinical data to evaluate interventions. Thus, PCOR and CER require capturing patient-reported outcome (PRO) data appropriately to inform research, healthcare delivery, and policy. This initiative's goal was to identify minimum standards for the design and selection of a PRO measure for use in PCOR and CER.

METHODS

We performed a literature review to find existing guidelines for the selection of PRO measures. We also conducted an online survey of the International Society for Quality of Life Research (ISOQOL) membership to solicit input on PRO standards. A standard was designated as "recommended" when >50 % respondents endorsed it as "required as a minimum standard."

RESULTS

The literature review identified 387 articles. Survey response rate was 120 of 506 ISOQOL members. The respondents had an average of 15 years experience in PRO research, and 89 % felt competent or very competent providing feedback. Final recommendations for PRO measure standards included: documentation of the conceptual and measurement model; evidence for reliability, validity (content validity, construct validity, responsiveness); interpretability of scores; quality translation, and acceptable patient and investigator burden.

CONCLUSION

The development of these minimum measurement standards is intended to promote the appropriate use of PRO measures to inform PCOR and CER, which in turn can improve the effectiveness and efficiency of healthcare delivery. A next step is to expand these minimum standards to identify best practices for selecting decision-relevant PRO measures.

Comparing higher order models for the EORTC QLQ-C30

PURPOSE

To investigate the statistical fit of alternative higher order models for summarizing the health-related quality of life profile generated by the EORTC QLQ-C30 questionnaire.

METHODS

A 50% random sample was drawn from a dataset of more than 9,000 pre-treatment QLQ-C30 v 3.0 questionnaires completed by cancer patients from 48 countries, differing in primary tumor site and disease stage. Building on a "standard" 14-dimensional QLQ-C30 model, confirmatory factor analysis was used to compare 6 higher order models, including a 1-dimensional (1D) model, a 2D "symptom burden and function" model, two 2D "mental/physical" models, and two models with a "formative" (or "causal") formulation of "symptom burden," and "function."

RESULTS

All of the models considered had at least an "adequate" fit to the data: the less restricted the model, the better the fit. The RMSEA fit indices for the various models ranged from 0.042 to 0.061, CFI's 0.90-0.96, and TLI's from 0.96 to 0.98. All chi-square tests were significant. One of the Physical/Mental models had fit indices superior to the other models considered.

CONCLUSIONS

The Physical/Mental health model had the best fit of the higher order models considered, and enjoys empirical and theoretical support in comparable instruments and applications.

Is there a genetic cause of appetite loss?-an explorative study in 1,853 cancer patients

BACKGROUND

Appetite loss has a major impact on cancer patients. It is exceedingly prevalent, is a prognostic indicator and is associated with inferior quality of life. Cachexia is a multi-factorial syndrome defined by a negative protein and energy balance, driven by a variable combination of reduced food intake and abnormal metabolism. Not all cancer patients that experience weight loss have appetite loss, and the pathophysiology between cachexia and appetite loss may thus be different. Knowledge of pathophysiology of appetite loss in cancer patients is still limited. The primary object of this study was to explore the association with 93 predefined candidate single-nucleotide polymorphisms (SNPs) and appetite loss in cancer patients to possibly generate new theories of the pathophysiology of the condition.

METHODS

A total of 1,853 cancer patients were phenotyped according to appetite loss and then genotyped.

RESULTS

After allowing for multiple testing, there was no statistically significant association between any of the SNPs analysed and appetite loss. The ten most significant SNPs in the co-dominant model had observed odds ratios varying from 0.72 to 1.28.

CONCLUSIONS

This large exploratory study could not find any associations with loss of appetite and 93 SNPs with a potential to be involved in appetite loss in cancer patients. This does not however rule out genes putative role in the development of the symptom, but the observed odds ratios are close to one which makes it unlikely that any of the individual SNPs explored in the present study have great importance.

Is there a genetic cause for cancer cachexia? - a clinical validation study in 1797 patients

BACKGROUND

Cachexia has major impact on cancer patients' morbidity and mortality. Future development of cachexia treatment needs methods for early identification of patients at risk. The aim of the study was to validate nine single-nucleotide polymorphisms (SNPs) previously associated with cachexia, and to explore 182 other candidate SNPs with the potential to be involved in the pathophysiology.

METHOD

A total of 1797 cancer patients, classified as either having severe cachexia, mild cachexia or no cachexia, were genotyped.

RESULTS

After allowing for multiple testing, there was no statistically significant association between any of the SNPs analysed and the cachexia groups. However, consistent with prior reports, two SNPs from the acylpeptide hydrolase (APEH) gene showed suggestive statistical significance (P=0.02; OR, 0.78).

CONCLUSION

This study failed to detect any significant association between any of the SNPs analysed and cachexia; although two SNPs from the APEH gene had a trend towards significance. The APEH gene encodes the enzyme APEH, postulated to be important in the endpoint of the ubiquitin system and thus the breakdown of proteins into free amino acids. In cachexia, there is an extensive breakdown of muscle proteins and an increase in the production of acute phase proteins in the liver.

Responsiveness and minimal important score differences in quality-of-life questionnaires: a comparison of the EORTC QLQ-C30 cancer-specific questionnaire to the generic utility questionnaires EQ-5D and 15D in patients with multiple myeloma

OBJECTIVES

The aims of this study were to (i) compare the responsiveness of the EORTC QLQ-C30 cancer-specific questionnaire and the generic questionnaires EQ-5D and 15D used for economic evaluation of healthcare interventions and (ii) determine the minimal important differences (MIDs) in these questionnaires. The MID is the smallest change in a quality-of-life score considered important to patients.

METHODS

Between 2006 and 2008, 239 patients with multiple myeloma completed the questionnaires at inclusion (T1) and after 3 months (T2). At T2, patients were asked whether they had noticed any change in their quality of life. Responsiveness and MIDs were determined by mean score changes (T2-T1) for patients who, in the interview, stated they had improved, deteriorated, or were unchanged. Responsiveness was also assessed using standardized response means. Wilcoxon tests for pair differences were used to evaluate the statistical significance of the changes.

RESULTS

Patients who improved had significantly (P < 0.01) higher scores at T2 in all three questionnaires. Patients who deteriorated reported lower scores at T2; however, for the 15D, the differences in score were not statistically significant. The MIDs for the QLQ-C30, EQ-5D, and 15D were 8, 0.08, and 0.03 in patients who improved and 12, 0.10 and 0.02 in patients who deteriorated, respectively.

CONCLUSIONS

All three questionnaires showed an acceptable responsiveness in patients who improved. However, the 15D did not respond optimally in patients who deteriorate and cannot be recommended for use in patients with myeloma.

Studies comparing Numerical Rating Scales, Verbal Rating Scales, and Visual Analogue Scales for assessment of pain intensity in adults: a systematic literature review

CONTEXT

The use of unidimensional pain scales such as the Numerical Rating Scale (NRS), Verbal Rating Scale (VRS), or Visual Analogue Scale (VAS) is recommended for assessment of pain intensity (PI). A literature review of studies specifically comparing the NRS, VRS, and/or VAS for unidimensional self-report of PI was performed as part of the work of the European Palliative Care Research Collaborative on pain assessment.

OBJECTIVES

To investigate the use and performance of unidimensional pain scales, with specific emphasis on the NRSs.

METHODS

A systematic search was performed, including citations through April 2010. All abstracts were evaluated by two persons according to specified criteria.

RESULTS

Fifty-four of 239 papers were included. Postoperative PI was most frequently studied; six studies were in cancer. Eight versions of the NRS (NRS-6 to NRS-101) were used in 37 studies; a total of 41 NRSs were tested. Twenty-four different descriptors (15 for the NRSs) were used to anchor the extremes. When compared with the VAS and VRS, NRSs had better compliance in 15 of 19 studies reporting this, and were the recommended tool in 11 studies on the basis of higher compliance rates, better responsiveness and ease of use, and good applicability relative to VAS/VRS. Twenty-nine studies gave no preference. Many studies showed wide distributions of NRS scores within each category of the VRSs. Overall, NRS and VAS scores corresponded, with a few exceptions of systematically higher VAS scores.

CONCLUSION

NRSs are applicable for unidimensional assessment of PI in most settings. Whether the variability in anchors and response options directly influences the numerical scores needs to be empirically tested. This will aid in the work toward a consensus-based, standardized measure.

Evidence-Based Guidelines for Determination of Sample Size and Interpretation of the European Organisation for the Research and Treatment of Cancer Quality of Life Questionnaire Core 30

Purpose

To use published literature to estimate large, medium, and small differences in quality of life (QOL) data from the European Organisation for the Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30).

Methods

An innovative method combining systematic review of published studies, expert opinions, and meta-analysis was used to estimate large, medium, and small differences for QLQ-C30 scores. Published mean data were identified from the literature. Differences (contrasts) between groups (eg, between treatment groups, age groups, and performance status groups) were reviewed by 34 experts in QOL measurement and cancer treatment. The experts, blinded to actual QOL results, were asked to predict these differences. A large difference was defined as one representing unequivocal clinical relevance. A medium difference was defined as likely to be clinically relevant but to a lesser extent. A small difference was one believed to be subtle but nevertheless clinically relevant. A trivial difference was used to describe circumstances unlikely to have any clinical relevance. Actual QOL results were combined using meta-analytic techniques to estimate differences corresponding to small, medium, or large effects.

Results

Nine hundred eleven articles were identified, leading to 152 relevant articles (2,217 contrasts) being reviewed by at least two experts. Resulting estimates from the meta-analysis varied depending on the subscale. Thus, the recommended minimum to detect medium differences ranges from 9 (cognitive functioning) to 19 points (role functioning).

Conclusion

Guidelines for the size of effects are provided for the QLQ-C30 subscales. These guidelines can be used for sample size calculations for clinical trials and can also be used to aid interpretation of differences in QLQ-C30 scores.

Interpretation of differential item functioning analyses using external review

Differential item functioning (DIF) analyses are used to determine whether certain groups respond differently to a particular item of a test or questionnaire; however, these do not explain the reasons for observed response differences. Many studies have used external reviews of items, sometimes using blinded reviewers, to help interpret these results. The authors conducted a literature review of this topic to describe the current usage of external reviews alongside DIF analyses. It concentrated on studies of health-related quality of life instruments, but studies in other fields were also considered. Relatively few examples of blinded item reviews were identified, and these were mostly from educational studies. A case study using blinded bilingual reviewers alongside translation DIF analyses of a health-related quality of life instrument is described. Future researchers should consider conducting external item reviews alongside DIF analyses.

Differential item functioning (DIF) analyses of health-related quality of life instruments using logistic regression

BACKGROUND

Differential item functioning (DIF) methods can be used to determine whether different subgroups respond differently to particular items within a health-related quality of life (HRQoL) subscale, after allowing for overall subgroup differences in that scale. This article reviews issues that arise when testing for DIF in HRQoL instruments. We focus on logistic regression methods, which are often used because of their efficiency, simplicity and ease of application.

METHODS

A review of logistic regression DIF analyses in HRQoL was undertaken. Methodological articles from other fields and using other DIF methods were also included if considered relevant.

RESULTS

There are many competing approaches for the conduct of DIF analyses and many criteria for determining what constitutes significant DIF. DIF in short scales, as commonly found in HRQL instruments, may be more difficult to interpret. Qualitative methods may aid interpretation of such DIF analyses.

CONCLUSIONS

A number of methodological choices must be made when applying logistic regression for DIF analyses, and many of these affect the results. We provide recommendations based on reviewing the current evidence. Although the focus is on logistic regression, many of our results should be applicable to DIF analyses in general. There is a need for more empirical and theoretical work in this area.

Minimal important differences and response shift in health-related quality of life; a longitudinal study in patients with multiple myeloma

Background

We previously reported that changes of 6-17 percent in the EORTC QLQ-C30 scores are regarded important by patients with multiple myeloma and thus may be considered as Minimal Important Differences (MIDs). However, patients' internal standard of measurement may have changed over time (response shift, RS). In the present work, we evaluated whether myeloma patients experience RS and if this could affect the MID-estimates.

Methods

Between 2006 and 2008, 239 patients with multiple myeloma completed the EORTC QLQ-C30 at baseline (T1) and after three months (T2). At T2, patients were asked if they had noticed any change in the domains pain, fatigue, physical function and global quality of life. They were also asked to give a retrospective judgment of their baseline values on all the four domains.

Results

We found clear evidence of RS in myeloma patients. However, there were differences in both magnitude and direction between patients who stated that they improved and those who deteriorated. Deteriorating patients retrospectively reported better health-related quality of life at T1 for the domains pain, fatigue and physical function. In these patients, MIDs adjusted for RS were observed to increase up to 12 percentage points. In contrast, for patients stating that they improved, we only found evidence of statistically significant RS in the domain global quality of life.

Conclusions

MIDs estimated from pre-test/post-test data appeared to be robust against RS in patients reporting improvement over 3-months. This could indicate that RS has a minimal impact on the results in patients who respond to treatment, and that RS may not have an important impact on interpretation of changes reported in clinical trials where an improvement occurs.

Although the effect sizes of the RSs were small, RS in deteriorating patients may have an important impact on the interpretation of changes reported in clinical trials.

Trial registration

The study is registered at clinicaltrials.gov, identifier NCT00290095.

The clinical and psychometric validation of a questionnaire to assess the quality of life of adult patients treated with long-term parenteral nutrition

BACKGROUND

Home parenteral nutrition (HPN) is an established treatment for the management of patients with chronic intestinal failure. No quality-of-life assessment tools have been developed and validated specifically for this patient population, and previous studies have used generic instruments or techniques not validated in HPN. The assessment of quality of life (QOL) should produce clinically relevant data reflecting patients' issues. The HPN-QOL was designed to assess the QOL of HPN patients. The purpose of this study was to test the hypothesized scale structure of the questionnaire with regard to reliability and validity in a sample of patients.

METHODS

A provisional questionnaire was prepared following recognized guidelines and then subjected to field testing. The questionnaire was administered to 100 adult patients receiving HPN. Psychometric tests examined the reliability and validity of the questionnaire, and patients' debriefing comments were analyzed.

RESULTS

The provisional questionnaire was adapted using evidence from quantitative and qualitative analysis. Multitrait scaling analysis and face validity refined the questionnaire to 48 items. Compliance rates were high, and the questionnaire was well accepted.

CONCLUSIONS

A method of objectively assessing the QOL of patients treated with HPN has been developed. The HPN-QOL has been rigorously prepared and demonstrates psychometric and clinical validity to assess the QOL of long-term HPN patients.

The practical impact of differential item functioning analyses in a health-related quality of life instrument

INTRODUCTION

Differential item functioning (DIF) analyses are commonly used to evaluate health-related quality of life (HRQoL) instruments. There is, however, a lack of consensus as to how to assess the practical impact of statistically significant DIF results.

METHODS

Using our previously published ordinal logistic regression DIF results for the Fatigue scale of a HRQoL instrument as an example, the practical impact on a particular Norwegian clinical trial was investigated. The results were used to determine the difference in mean Fatigue scores assuming that the same trial was conducted in the UK. The results were then compared with published information on what would be considered a clinically important change in scores.

RESULTS

The item with the largest DIF effect resulted in differences between the mean English and Norwegian Fatigue scores that, although small, could be considered clinically important. Sensitivity analyses showed that larger differences were found for shorter scales, and when the proportions in each response category were equal.

DISCUSSION

Our scenarios suggest that translation differences in an item can result in small, but clinically important, differences at the scale score level. This is more likely to be problematic for observational studies than for clinical trials, where randomised groups are stratified by centre.

Investigating the missing data mechanism in quality of life outcomes: a comparison of approaches

BACKGROUND

Missing data is classified as missing completely at random (MCAR), missing at random (MAR) or missing not at random (MNAR). Knowing the mechanism is useful in identifying the most appropriate analysis. The first aim was to compare different methods for identifying this missing data mechanism to determine if they gave consistent conclusions. Secondly, to investigate whether the reminder-response data can be utilised to help identify the missing data mechanism.

METHODS

Five clinical trial datasets that employed a reminder system at follow-up were used. Some quality of life questionnaires were initially missing, but later recovered through reminders. Four methods of determining the missing data mechanism were applied. Two response data scenarios were considered. Firstly, immediate data only; secondly, all observed responses (including reminder-response).

RESULTS

In three of five trials the hypothesis tests found evidence against the MCAR assumption. Logistic regression suggested MAR, but was able to use the reminder-collected data to highlight potential MNAR data in two trials.

CONCLUSION

The four methods were consistent in determining the missingness mechanism. One hypothesis test was preferred as it is applicable with intermittent missingness. Some inconsistencies between the two data scenarios were found. Ignoring the reminder data could potentially give a distorted view of the missingness mechanism. Utilising reminder data allowed the possibility of MNAR to be considered.

Differential item functioning (DIF) in the EORTC QLQ-C30: a comparison of baseline, on-treatment and off-treatment data

INTRODUCTION

Differential item functioning (DIF) analyses can be used to explore translation, cultural, gender or other differences in the performance of quality of life (QoL) instruments. These analyses are commonly performed using "baseline" or pretreatment data. We previously reported DIF analyses to examine the pattern of item responses for translations of the European Organisation for Research and Treatment of Cancer (EORTC) QLQ-C30 QoL instrument, using only data collected prior to cancer treatment. We now compare the consistency of these results with similar analyses of on-treatment and off-treatment assessments and explore whether item relationships differ from those at baseline.

METHODS

Logistic regression DIF analyses were used to examine the translation of each item in each multi-item scale at the three time points, after controlling for the overall scale score and other covariates. The consistency of results at the three time points was explored.

RESULTS

For most EORTC QLQ-C30 subscales, the DIF results were very consistent across the three time points. Results for the Nausea and Vomiting scale varied the most across assessments.

DISCUSSION

The results indicated that DIF analyses were stable across each time point and that the same DIF effects were usually found regardless of the treatment status of the respondent.

Meaning behind measurement: self-comparisons affect responses to health-related quality of life questionnaires

PURPOSE

The subjective nature of quality of life is particularly pertinent to the domain of health-related quality of life (HRQOL) research. The extent to which participants' responses are affected by subjective information and personal reference frames is unknown. This study investigated how an elderly population living with a chronic metabolic bone disorder evaluated self-reported quality of life.

METHODS

Participants (n = 1,331) in a multi-centre randomised controlled trial for the treatment of Paget's disease completed annual HRQOL questionnaires, including the SF-36, EQ-5D and HAQ. Supplementary questions were added to reveal implicit reference frames used when making HRQOL evaluations. Twenty-one participants (11 male, 10 female, aged 59-91 years) were interviewed retrospectively about their responses to the supplementary questions, using cognitive interviewing techniques and semi-structured topic guides.

RESULTS

The interviews revealed that participants used complex and interconnected reference frames to promote response shift when making quality of life evaluations. The choice of reference frame often reflected external factors unrelated to individual health. Many participants also stated that they were unclear whether to report general or disease-related HRQOL.

CONCLUSIONS

It is important, especially in clinical trials, to provide instructions clarifying whether 'quality of life' refers to disease-related HRQOL. Information on self-comparison reference frames is necessary for the interpretation of responses to questions about HRQOL.

Evaluating the effectiveness of using PROs in clinical practice: a role for cluster-randomised trials

Many of us believe there are major benefits to be gained by using patient-reported outcomes (PROs) in routine clinical practice. However, demonstrating tangible benefits has frequently proved elusive. Although randomised clinical trials (RCTs) have become accepted as the standard for comparing alternative forms of clinical interventions, when evaluating the effectiveness of PROs there are a number of challenges both in terms of study design and the subsequent analysis. Despite this, to date, of those investigators who have used RCTs, relatively few have used anything other than simple two-group randomisation and comparisons. Most of these trials have also failed to demonstrate convincing benefits to patient outcomes. We suggest that the use of PROs may result in modest yet important improvements to patient outcomes, and that these benefits may be obscured in conventional individual patient trials because of contamination effects. The advantages of alternative designs such as cross-over studies and in particular cluster-randomised trials are illustrated.

A simulation study provided sample size guidance for differential item functioning (DIF) studies using short scales

OBJECTIVE

Differential item functioning (DIF) analyses are increasingly used to evaluate health-related quality of life (HRQoL) instruments, which often include relatively short subscales. Computer simulations were used to explore how various factors including scale length affect analysis of DIF by ordinal logistic regression.

STUDY DESIGN AND SETTING

Simulated data, representative of HRQoL scales with four-category items, were generated. The power and type I error rates of the DIF method were then investigated when, respectively, DIF was deliberately introduced and when no DIF was added. The sample size, scale length, floor effects (FEs) and significance level were varied.

RESULTS

When there was no DIF, type I error rates were close to 5%. Detecting moderate uniform DIF in a two-item scale required a sample size of 300 per group for adequate (>80%) power. For longer scales, a sample size of 200 was adequate. Considerably larger sample sizes were required to detect nonuniform DIF, when there were extreme FEs or when a reduced type I error rate was required.

CONCLUSION

The impact of the number of items in the scale was relatively small. Ordinal logistic regression successfully detects DIF for HRQoL instruments with short scales. Sample size guidelines are provided.

Simple imputation methods were inadequate for missing not at random (MNAR) quality of life data

Objective

QoL data were routinely collected in a randomised controlled trial (RCT), which employed a reminder system, retrieving about 50% of data originally missing. The objective was to use this unique feature to evaluate possible missingness mechanisms and to assess the accuracy of simple imputation methods.

Methods

Those patients responding after reminder were regarded as providing missing responses. A hypothesis test and a logistic regression approach were used to evaluate the missingness mechanism. Simple imputation procedures were carried out on these missing scores and the results compared to the actual observed scores.

Results

The hypothesis test and logistic regression approaches suggested the reminder data were missing not at random (MNAR). Reminder-response data showed that simple imputation procedures utilising information collected close to the point of imputation (last value carried forward, next value carried backward and last-and-next), were the best methods in this setting. However, although these methods were the best of the simple imputation procedures considered, they were not sufficiently accurate to be confident of obtaining unbiased results under imputation.

Conclusion

The use of the reminder data enabled the conclusion of possible MNAR data. Evaluating this mechanism was important in determining if imputation was useful. Simple imputation was shown to be inadequate if MNAR are likely and alternative strategies should be considered.

Quality, interpretation and presentation of European Organisation for Research and Treatment of Cancer quality of life questionnaire core 30 data in randomised controlled trials

AIM

To review reporting standard, presentation and interpretation for quality of life (QOL) outcomes in randomised controlled trials (RCTs) using the European Organisation for Research and Treatment of Cancer quality of life questionnaire core 30 (EORTC QLQ-C30).

METHODS

Cancer RCTs reporting EORTC QLQ-C30 data were identified and reviewed against a reporting quality checklist. Interpretation/presentation methods for QOL data were also recorded.

RESULTS

Eighty-two papers were reviewed. Seventy percent met criteria for high quality reporting; 94% reported mean scores; 84% presented results in tables/graphs; 80% reported p-values or statistical significance. Clinical significance was addressed in 38%. Where clinical significance was not addressed, reliance was usually on statistical significance to interpret the results.

DISCUSSION

EORTC QLQ-C30 results are generally reported well, although it was common to rely on statistical significance alone for interpreting results. Whilst interpretation in terms of clinical significance has improved in recent years, there is still a lack of robust clinical interpretation of QOL results even in papers reported to a high standard.

Doloplus-2, a valid tool for behavioural pain assessment?

BACKGROUND

The Doloplus-2 is used for behavioural pain assessment in cognitively impaired patients. Little data exists on the psychometric properties of the Doloplus-2. Our objectives were to test the criterion validity and inter-rater reliability of the Doloplus-2, and to explore a design for validations of behavioural pain assessment tools.

METHODS

Fifty-one nursing home patients and 22 patients admitted to a geriatric hospital ward were included. All were cognitively impaired and unable to self-report pain. Each patient was examined by an expert in pain evaluation and treatment, who rated the pain on a numerical rating scale. The ratings were based on information from the medical record, reports from nurses and patients (if possible) about pain during the past 24 hours, and a clinical examination. These ratings were used as pain criterion. The Doloplus-2 was administered by the attending nurse. Regression analyses were used to estimate the ability of the Doloplus-2 to explain the expert's ratings. The inter-rater reliability of the Doloplus-2 was evaluated in 16 patients by comparing the ratings of two nurses administrating the Doloplus-2.

RESULTS

There was no association between the Doloplus-2 and the expert's pain ratings (R2 = 0.02). There was an association (R2 = 0.54) between the expert's ratings and the Doloplus-2 scores in a subgroup of 16 patients assessed by a geriatric expert nurse (the most experienced Doloplus-2 administrator). The inter-rater reliability between the Doloplus-2 administrators assessed by the intra-class coefficient was 0.77. The pain expert's ratings were compared with ratings of two independent geriatricians in a sub sample of 15, and were found satisfactory (intra-class correlation 0.74).

CONCLUSION

It was challenging to conduct such a study in patients with cognitive impairment and the study has several limitations. The results do not support the validity of the Doloplus-2 in its present version and they indicate that it demands specific administration skills.

Implicit self-comparisons against others could bias quality of life assessments

OBJECTIVES

To explore how patient-reported health-related quality of life (HRQL) and global health status are affected by use of differing personal reference frames. We hypothesized that implicit comparisons against self at an earlier time, against healthy peers, or against ill patients would greatly affect patients' response values.

STUDY DESIGN AND SETTING

Patients in a randomized trial for treatment of Paget's disease completed annual HRQL questionnaires. Supplementary questions were appended, asking the patients whether they were aware of having made implicit comparisons.

RESULTS

The majority of patients reported considering themselves a year ago (31% at baseline), themselves before becoming ill (23%), or other healthy people (24%), with similar proportions during follow-up. Mean HRQL scores varied substantially according to the declared frame of reference, with differences as big as 19% of the scale score, or a standardized mean effect size of 0.74 standard deviations.

CONCLUSION

Reported reference frames were associated with effects of similar magnitude to the differences in HRQL that are regarded as clinically important. This may be of particular concern in trials that randomize patients to management in different settings, such as treatment at home/in hospital, or surgery/chemotherapy and might bias or obscure HRQL differences.

Applying item response theory and computer adaptive testing: the challenges for health outcomes assessment

OBJECTIVES

We review the papers presented at the NCI/DIA conference, to identify areas of controversy and uncertainty, and to highlight those aspects of item response theory (IRT) and computer adaptive testing (CAT) that require theoretical or empirical research in order to justify their application to patient reported outcomes (PROs).

BACKGROUND

IRT and CAT offer exciting potential for the development of a new generation of PRO instruments. However, most of the research into these techniques has been in non-healthcare settings, notably in education. Educational tests are very different from PRO instruments, and consequently problematic issues arise when adapting IRT and CAT to healthcare research.

RESULTS

Clinical scales differ appreciably from educational tests, and symptoms have characteristics distinctly different from examination questions. This affects the transferring of IRT technology. Particular areas of concern when applying IRT to PROs include inadequate software, difficulties in selecting models and communicating results, insufficient testing of local independence and other assumptions, and a need of guidelines for estimating sample size requirements. Similar concerns apply to differential item functioning (DIF), which is an important application of IRT. Multidimensional IRT is likely to be advantageous only for closely related PRO dimensions.

CONCLUSIONS

Although IRT and CAT provide appreciable potential benefits, there is a need for circumspection. Not all PRO scales are necessarily appropriate targets for this methodology. Traditional psychometric methods, and especially qualitative methods, continue to have an important role alongside IRT. Research should be funded to address the specific concerns that have been identified.

A review of the instruments used to assess the quality of life of adult patients with chronic intestinal failure receiving parenteral nutrition at home

Home parenteral nutrition is an established method of supporting patients with intestinal failure, but this treatment may be life long and imposes severe restrictions on daily life. Impact on quality of life is an important outcome when considering the management of home parenteral nutrition patients. This paper reviews studies in which the quality of life of patients receiving home parenteral nutrition has been assessed. A systematic search of electronic databases and relevant publications was undertaken to identify generic or treatment-specific questionnaires used with home parenteral nutrition patients. Many of the thirty-four reports discovered were small studies. Nineteen used non-specific generic instruments, eight used non-validated questionnaires, four used a combination of both, and three did not use any formal tool. Few systematic patterns emerged. There are few available data on the quality of life of home parenteral nutrition patients, and there is a need for standardised, scientifically validated, treatment-specific instruments to measure quality of life in this population.

Psychological distress and fatigue predicted recurrence and survival in primary breast cancer patients

This study examined whether psychological distress in newly diagnosed breast cancer patients was associated with their survival. We analyzed data from 1,588 breast cancer patients who filled in the EORTC QLQ-C30 questionnaire and the Hospital Anxiety and Depression Scale (HADS) 2 months after their primary operation. The median follow-up time was 12.9 years. Psychological distress (EORTC QLQ-C30 emotional function; HADS anxiety; HADS depression) and EORTC fatigue, physical function, and overall ratings were used to predict recurrence-free and overall survival, controlling for the known clinical and histopathological prognostic factors (biological model) using Cox multivariate regression analysis. Low levels of psychological distress (good EORTC emotional function) and low fatigue independently predicted longer recurrence-free and overall survival, controlling for biological factors. Lack of anxiety (HADS) also predicted longer recurrence-free survival. When added in combination to the biological model, fatigue remained a significant predictor of recurrence-free survival (P = 0.0004; risk ratio 1.32 (1.13-1.54)) and emotional function remained a significant predictor of overall survival (P = 0.0074; risk ratio 0.81 (0.70-0.95)). Low psychological distress and a low level of fatigue may cause a greater cancer resistance or may reflect underlying mental and physical robustness.

Methods for handling missing data in palliative care research

Missing data is a common problem in palliative care research due to the special characteristics (deteriorating condition, fatigue and cachexia) of the population. Using data from a palliative study, we illustrate the problems that missing data can cause and show some approaches for dealing with it. Reasons for missing data and ways to deal with missing data (including complete case analysis, imputation and modelling procedures) are explored. Possible mechanisms behind the missing data are: missing completely at random, missing at random or missing not at random. In the example study, data are shown to be missing at random. Imputation of missing data is commonly used (including last value carried forward, regression procedures and simple mean). Imputation affects subsequent summary statistics and analyses, and can have a substantial impact on estimated group means and standard deviations. The choice of imputation method should be carried out with caution and the effects reported.

Pain assessment tools: is the content appropriate for use in palliative care?

Inadequate pain assessment prevents optimal treatment in palliative care. The content of pain assessment tools might limit their usefulness for proper pain assessment, but data on the content validity of the tools are scarce. The objective of this study was to examine the content of the existing pain assessment tools, and to evaluate the appropriateness of different dimensions and items for pain assessment in palliative care. A systematic search was performed to find pain assessment tools for patients with advanced cancer who were receiving palliative care. An ad hoc search with broader search criteria supplemented the systematic search. The items of the identified tools were allocated to appropriate dimensions. This was reviewed by an international panel of experts, who also evaluated the relevance of the different dimensions for pain assessment in palliative care. The systematic literature search generated 16 assessment tools while the ad hoc search generated 64. Ten pain dimensions containing 1,011 pain items were identified by the experts. The experts ranked intensity, temporal pattern, treatment and exacerbating/relieving factors, location, and interference with health-related quality of life as the most important dimensions. None of the assessment tools covered these dimensions satisfactorily. Most items were related to interference (231) and intensity (138). Temporal pattern (which includes breakthrough pain), ranked as the second most important dimension, was covered by 29 items only. Many tools include dimensions and items of limited relevance for patients with advanced cancer. This might reduce compliance and threaten the validity of the assessment. New tools should reflect the clinical relevance of different dimensions and be user-friendly.