Development of symptom assessments utilising item response theory and computer-adaptive testing--a practical method based on a systematic review

Comparing Patient-Reported Outcomes Measurement Information System Computer Adaptive Testing With Existing Measures After Operative Interventions for Extremity Fractures

INTRODUCTION

Patient-Reported Outcomes Measurement Information System (PROMIS) enables the use of computer adaptive testing (CAT). The aim of this prospective cohort study was to compare the most commonly used disease-specific instruments with PROMIS CAT questionnaires in patients with trauma.

METHODS

All patients with trauma (ages 18-75) who underwent an operative intervention for an extremity fracture between June 1, 2018, and June 30, 2019, were included. The disease-specific instruments were the Quick Disabilities of the Arm, Shoulder, and Hand for upper extremity fractures and the Lower Extremity Functional Scale (LEFS) for lower extremity fractures. Pearson's correlation (r) between the disease-specific instruments and the PROMIS CAT questionnaires (PROMIS Physical Function, PROMIS Pain Interference, and PROMIS Ability to Participate in Social Roles and Activities) was calculated at week 2, week 6, month 3, and month 6. Construct validity and responsiveness were calculated.

RESULTS

A total of 151 patients with an upper extremity fracture and 109 patients with a lower extremity fracture were included. At month 3 and month 6, the correlation was strong between the LEFS and PROMIS Physical Function (r = 0.88 and r = 0.90, respectively), and at month 3, the correlation was strong between the LEFS and PROMIS Social Roles and Activities (r = 0.72). At week 6, month 3, and month 6, there was a strong correlation between the Quick Disabilities of the Arm, Shoulder, and Hand and PROMIS Physical Function (r = 0.74, r = 0.70, and r = 0.76, respectively).

CONCLUSIONS

The PROMIS CAT measures are acceptably related to existing non-CAT instruments and may be a useful tool during follow-up after operative interventions for extremity fractures.

Assessing the experience of the quality of care of patients living with multiple sclerosis and their caregivers: The MusiCare questionnaire

BACKGROUND AND PURPOSE

Patients with a chronic illness, such as multiple sclerosis (MS), and their natural caregivers have a specific experience of healthcare and health services. These experiences need to be assessed to evaluate the quality of care. Our objective was to develop a French-language questionnaire to evaluate the quality of care as experienced by MS patients and their natural caregivers.

METHODS

Eligible patients had been diagnosed with MS according to the McDonald criteria. Eligible caregivers were individuals designated by the patients. The MusiCare questionnaire was developed in two standard phases: (i) item generation, based on interviews with patients and caregivers; and (ii) validation, consisting of validity, reliability, external validity, reproducibility, and responsiveness measures.

RESULTS

In total, 1088 patients (n = 660) and caregivers (n = 488) were recruited. The initial 64-item version of MusiCare was administered to a random subsample (n = 748). The validation process generated a 35-item questionnaire. Internal consistency and scalability were satisfactory. Testing of the external validity revealed expected associations between MusiCare scores and sociodemographic and clinical data. The questionnaire showed good reproducibility and responsiveness.

CONCLUSIONS

The availability of a reliable and validated French-language self-report questionnaire probing the experience of the quality of care for MS will allow the feedback of patients and caregivers to be incorporated into a continuous healthcare quality-improvement strategy.

Modification and verification of the Infant-Toddler Meaningful Auditory Integration Scale: a psychometric analysis combining item response theory with classical test theory

Background

Early prelingual auditory development (EPLAD) is a fundamental and important process in the speech and language development of infants and toddlers. The Infant–Toddler Meaningful Auditory Integration Scale (ITMAIS) is a widely used measurement tool for EPLAD, however it has not yet undergone a comprehensive psychometric analysis. The aim of this research was to modify and verify the psychometric properties of ITMAIS using a combination of Item Response Theory (IRT) and Classical Test Theory (CTT).

Methods

Stage 1—1730 children were retrospectively recruited to enable the application of an IRT model, specifically the graded response model, to modify the ITMAIS. Stage 2—another 450 infants and toddlers with normal hearing or permanent hearing loss before auditory intervention were recruited to verify the psychometric properties of the modified ITMAIS (ITMAIS-m) using the CTT method.

Results

Using the metric of the graded response model, by removing item 2 from the ITMAIS, ITMAIS-m demonstrated discrimination parameters ranging from 3.947 to 5.431, difficulty parameters from − 1.146 to 1.150, item information distributed between 4.798 and 9.259 and a test information score of 48.061. None of the items showed differential item functioning. ITMAIS-m was further verified in Stage 2, showing Cronbach’s α of 0.919 and item-total correlations ranging from 0.693 to 0.851. There was good convergent validity of ITMAIS-m with other auditory outcome measure (r = 0.932) and pure tone average thresholds (r ranging from − 0.670 to − 0.909), as well as a high ability to discriminate between different hearing grades (Cohen d ranging from 0.41 to 5.83).

Conclusions

The ITMAIS-m is a reliable and valid tool for evaluating EPLAD in infants and toddlers, which can be efficiently and precisely applied in clinical practice. The combined use of IRT and CTT provides a powerful means to modify psychometrically robust scales aimed at childhood auditory outcome measurements.

Review: Patient-reported outcomes in multiple sclerosis care

Patient-reported outcomes (PROs) are increasingly used in multiple sclerosis (MS) research and clinical practice for understanding the effects that the disease and its treatments have on patients' lives. PROs are captured directly from patients and include symptoms, function, health status, and health-related quality of life. No universal guidance on appropriateness of each applied tool exists. However, collecting clear and comprehensive outcome measures represents the first step of patient centered therapeutic management. The importance of PRO assessment is expected to continue to grow in the future. But in current MS reality, PROs are selected and used without a clear justification, and only few PROs are of adequate psychometric quality. There is a clear need for the development of high-quality; MS-specific PROs that assess the true concerns of patients and that evaluate the impact of both clinical and non-clinical interventions on a variety of outcomes. In this perspective review, we describe the importance of and methods for using PRO in MS by defining and identifying the used PROs in MS. Moreover, we will outline the challenges and key unanswered questions for routine use of PROs in MS discussing potential interventions to accelerate the integration of PROs in the clinical management of MS.

Usability testing of EirV3-a computer-based tool for patient-reported outcome measures in cancer

PURPOSE

Eir version 3 (V3) is an electronic tool for administration of patient-reported outcome measures (Eir-Patient) that immediately presents patient scores on the physician's computer (Eir-Doctor). Perceived usability is an important determinant for successful implementation. The aim of this study was to answer the following research question evaluated at the cancer outpatient clinics, in the patients' home, and at general practitioners' (GPs) offices: What are the number, type, and severity of usability issues evaluated by the patient (Eir-Patient module) and by the physician (Eir-Doctor module)?

METHODS

A usability evaluation using observations, think-aloud sessions, individual interviews and focus group interviews in cancer patients and their physicians was conducted. Identified usability issues were graded on a severity scale from 1 (irritant) to 4 (unusable).

RESULTS

Overall, 73 Eir registrations were performed by 37 patients, and used by 17 physicians in clinical consultations. All patients were able to complete the Eir-Patient symptom registration. Seventy-two usability issues were identified. None of them were graded as unusable. For the Eir-Patient module, 62% of the identified usability issues was graded as irritant (grade 1), 18% as moderate (grade 2), and 20% as severe (grade 3). For the Eir-Doctor module, 46% of the identified usability issues were graded as irritant, 36% as moderate and 18% as severe.

CONCLUSIONS

In the updated Eir version, issues in the severe and moderate categories have been changed, to optimize the usability of using real-time PROMs in clinical practice.

Does service timing matter for psychological outcomes in cardiac rehabilitation? Insights from the National Audit of Cardiac Rehabilitation

Background The presence of mental health conditions in cardiac rehabilitation (CR) patients such as anxiety and depression can lead to reduced programme adherence, increased mortality and increased re-occurrence of cardiovascular events undermining the aims and benefit of CR. Earlier research has identified a relationship between delayed commencement of CR and poorer physical activity outcomes. This study wished to explore whether a similar relationship between CR wait time and mental health outcomes can be found and to what degree participation in CR varies by mental health status. Methods Data from the UK National Audit of Cardiac Rehabilitation, a dataset that captures information on routine CR practice and patient outcomes, was extracted between 2012 and 2016. Logistic and multinomial regression models were used to explore the relationship between timing of CR and mental health outcomes measured on the hospital anxiety and depression scale. Results The results of this study showed participation in CR varied by mental health status, particularly in relation to completion of CR, with a higher proportion of non-completers with symptoms of anxiety (5% higher) and symptoms of depression (8% higher). Regression analyses also revealed that delays to CR commencement significantly impact mental health outcomes post-CR. Conclusion In these analyses CR wait time has been shown to predict the outcome of anxiety and depression status to the extent that delays in starting CR are detrimental. Programmes falling outside the 4-week window for commencement of CR following referral must strive to reduce wait times to avoid negative impacts to patient outcome.

Computerized adaptive testing with decision regression trees: an alternative to item response theory for quality of life measurement in multiple sclerosis

BACKGROUND

The aim of this study was to propose an alternative approach to item response theory (IRT) in the development of computerized adaptive testing (CAT) in quality of life (QoL) for patients with multiple sclerosis (MS). This approach relied on decision regression trees (DRTs). A comparison with IRT was undertaken based on precision and validity properties.

MATERIALS AND METHODS

DRT- and IRT-based CATs were applied on items from a unidi-mensional item bank measuring QoL related to mental health in MS. The DRT-based approach consisted of CAT simulations based on a minsplit parameter that defines the minimal size of nodes in a tree. The IRT-based approach consisted of CAT simulations based on a specified level of measurement precision. The best CAT simulation showed the lowest number of items and the best levels of precision. Validity of the CAT was examined using sociodemographic, clinical and QoL data.

RESULTS

CAT simulations were performed using the responses of 1,992 MS patients. The DRT-based CAT algorithm with minsplit = 10 was the most satisfactory model, superior to the best IRT-based CAT algorithm. This CAT administered an average of nine items and showed satisfactory precision indicators (R = 0.98, root mean square error [RMSE] = 0.18). The DRT-based CAT showed convergent validity as its score correlated significantly with other QoL scores and showed satisfactory discriminant validity.

CONCLUSION

We presented a new adaptive testing algorithm based on DRT, which has equivalent level of performance to IRT-based approach. The use of DRT is a natural and intuitive way to develop CAT, and this approach may be an alternative to IRT.

Using PROMs to guide patients and practitioners through the head and neck cancer journey

The measurement of patient-reported outcome measures (PROMs) following head and neck cancer (HNC) has the capacity to substantially enhance the care of patients and their care-givers following the diagnosis and treatment of HNC. Literature concerning PROMs has increased exponentially in the past 2 decades, producing a vast array of data upon which the multidisciplinary team can reflect. For this review, “Handle On QOL” has been used as a source of references to illustrate the points raised. PROMs are contextualized by considering the clinically-distinct key stages that cancer patients endure: diagnosis, treatment, acute toxicity, early recovery, late effects, recurrence, and palliation. The PROMs are considered in six main categories: 1) those addressing cornucopia of issues not specific to cancer; 2) those addressing issues common to all cancers; 3) questionnaires with items specific to HNC; 4) questionnaires that focus on a particular aspect of head and neck function; 5) those measuring psychological concerns, such as depression, anxiety, or self-esteem; and 6) item prompt lists. Potential benefits of PROMs in clinical practice are discussed, as are barriers to use. The way forward in integrating PROMs into routine HNC care is discussed with an emphasis on information technology.

A computer adaptive testing version of the Addiction Severity Index-Multimedia Version (ASI-MV): The Addiction Severity CAT

The purpose of this study was to develop and validate a computer adaptive testing (CAT) version of the Addiction Severity Index-Multimedia Version (ASI-MV), the Addiction Severity CAT. This goal was accomplished in 4 steps. First, new candidate items for Addiction Severity CAT domains were evaluated after brainstorming sessions with experts in substance abuse treatment. Next, this new item bank was psychometrically evaluated on a large nonclinical (n = 4,419) and substance abuse treatment (n = 845) sample. Based on these results, final items were selected and calibrated for the creation of the Addiction Severity CAT algorithms. Once the algorithms were developed for the entire assessment, a fully functioning prototype of an Addiction Severity CAT was created. CAT simulations were conducted, and optimal termination criteria were selected for the Addiction Severity CAT algorithms. Finally, construct validity of the CAT algorithms was evaluated by examining convergent and discriminant validity and sensitivity to change. The Addiction Severity CAT was determined to be valid, sensitive to change, and reliable. Further, the Addiction Severity CAT's time of completion was found to be significantly less than the average time of completion for the ASI-MV composite scores. This study represents the initial validation of an Addiction Severity CAT based on item response theory, and further exploration of the Addiction Severity CAT is needed. (PsycINFO Database Record

Computerized adaptive testing of population psychological distress: simulation-based evaluation of GHQ-30

PURPOSE

Goldberg's General Health Questionnaire (GHQ) items are frequently used to assess psychological distress but no study to date has investigated the GHQ-30's potential for adaptive administration. In computerized adaptive testing (CAT) items are matched optimally to the targeted distress level of respondents instead of relying on fixed-length versions of instruments. We therefore calibrate GHQ-30 items and report a simulation study exploring the potential of this instrument for adaptive administration in a longitudinal setting.

METHODS

GHQ-30 responses of 3445 participants with 2 completed assessments (baseline, 7-year follow-up) in the UK Health and Lifestyle Survey were calibrated using item response theory. Our simulation study evaluated the efficiency of CAT administration of the items, cross-sectionally and longitudinally, with different estimators, item selection methods, and measurement precision criteria.

RESULTS

To yield accurate distress measurements (marginal reliability at least 0.90) nearly all GHQ-30 items need to be administered to most survey respondents in general population samples. When lower accuracy is permissible (marginal reliability of 0.80), adaptive administration saves approximately 2/3 of the items. For longitudinal applications, change scores based on the complete set of GHQ-30 items correlate highly with change scores from adaptive administrations.

CONCLUSIONS

The rationale for CAT-GHQ-30 is only supported when the required marginal reliability is lower than 0.9, which is most likely to be the case in cross-sectional and longitudinal studies assessing mean changes in populations. Precise measurement of psychological distress at the individual level can be achieved, but requires the deployment of all 30 items.

A Multidimensional Computerized Adaptive Short-Form Quality of Life Questionnaire Developed and Validated for Multiple Sclerosis: The MusiQoL-MCAT

The aim was to develop a multidimensional computerized adaptive short-form questionnaire, the MusiQoL-MCAT, from a fixed-length QoL questionnaire for multiple sclerosis.A total of 1992 patients were enrolled in this international cross-sectional study. The development of the MusiQoL-MCAT was based on the assessment of between-items MIRT model fit followed by real-data simulations. The MCAT algorithm was based on Bayesian maximum a posteriori estimation of latent traits and Kullback-Leibler information item selection. We examined several simulations based on a fixed number of items. Accuracy was assessed using correlations (r) between initial IRT scores and MCAT scores. Precision was assessed using the standard error measurement (SEM) and the root mean square error (RMSE).The multidimensional graded response model was used to estimate item parameters and IRT scores. Among the MCAT simulations, the 16-item version of the MusiQoL-MCAT was selected because the accuracy and precision became stable with 16 items with satisfactory levels (r ≥ 0.9, SEM ≤ 0.55, and RMSE ≤ 0.3). External validity of the MusiQoL-MCAT was satisfactory.The MusiQoL-MCAT presents satisfactory properties and can individually tailor QoL assessment to each patient, making it less burdensome to patients and better adapted for use in clinical practice.

Assessment of anticancer-treatment outcome in patients with metastatic castration-resistant prostate cancer-going beyond PSA and imaging, a systematic literature review

BACKGROUND

In the past years, there has been significant progress in anticancer drug development for patients with metastatic castration-resistant prostate cancer (CRPC). However, the current instruments to assess clinical treatment response have limitations and may not sufficiently reflect patient benefit. Our objective was to systematically identify tools to evaluate both patient benefit and clinical anticancer-treatment response as basis for an international consensus process and development of a specific pragmatic instrument for men with CRPC.

METHODS

PubMed, Embase and CINAHL were searched to identify currently available tools to assess anticancer-treatment benefit, other than standard imaging procedures and prostate-specific antigen measurements, namely quality of life (QoL), detailed pain assessment, physical function and objective measures of other complex cancer-related syndromes in patients with CRPC. Additionally, all CRPC phase III trials published in the last 5 years were reviewed as well as studies using physical function tools in a general cancer population. The PRIMSA statement was followed for the systematic review process.

RESULTS

The search generated 1096 hits, 185 full-text papers were screened and finally 73 publications were included. Additional 89 publications were included by hand-search. We identified a total of 98 tools used in CRPC trials and grouped these into three categories: 22 tools assessing QoL domains and subgroups, 47 tools for pain assessment and 29 tools for objective measures, mainly physical function and assessment of skeletal disease burden.

CONCLUSION

A wide variety of assessment tools and also efforts to standardize and harmonize patient-reported outcomes and pain assessment were identified. However, the specific needs of the increasing CRPC population living longer with their incurable cancer are insufficiently captured and objective physical outcome measures are under-represented. In the age of new anticancer drug targets and principles, new methods to monitor patient relevant outcomes of antineoplastic therapy are of utmost importance.

The Adults and Older Adults Functional Assessment Inventory

Functional assessment methods are an important element in multidimensional neuropsychological evaluations, particularly in older adults. The Adults and Older Adults Functional Assessment Inventory is a new measure of basic and instrumental activities of daily living. Rasch model analyses were used to analyze the psychometric characteristics of the instrument in a sample of 803 participants. The original categories did not provide an optimal assessment of functional incapacity. The scale was dichotomized to achieve a better reliability score and item fit. The final 50 items revealed a moderately high variability in item difficulty, acceptable fits to items and persons, and a good Person Separation Reliability score. The scores were able to discriminate between normal controls and clinical patients. None of the items showed Differential Item Functioning associated with age, gender, or education. The instrument is able to achieve measures of functional incapacity with the useful properties of the Rasch model.

Psychosocial consequences of cancer cachexia: the development of an item bank

CONTEXT

Cancer cachexia syndrome (CCS) is often accompanied by psychosocial consequences (PSC). To alleviate PSC, a systematic assessment method is required. Currently, few assessment tools are available (e.g., Functional Assessment of Anorexia/Cachexia Therapy). There is no systematic assessment tool that captures the PSC of CCS.

OBJECTIVES

To develop a pilot item bank to assess the PSC of CCS.

METHODS

A total of 132 questions, generated from patient answers in a previous study, were reduced to 121 items by content analysis and evaluation by multidisciplinary experts (doctor, nutritionists, and nurses). In our two-step, cross-sectional study, patients, judged by staff to have PSC of CCS, were included, and the questions were randomly allocated to the patients. Questions were evaluated for understandability and triggering emotions, and patients were asked to provide a response using a four-point Likert scale. Subsequently, problematic questions were revised, reformulated, and retested.

RESULTS

A total of 20 patients with a variety of tumor types participated. Of the 121 questions, 31 had to be reformulated after Step 1 and were retested in Step 2, after which seven were again evaluated as not being perfectly comprehensible. In Step 1, 22 questions were found to trigger emotions, but no item required remodeling. Item performance using the Likert scale revealed no consistent floor or ceiling effects. Our final pilot question bank comprised 117 questions.

CONCLUSION

The final item bank contains questions that are understood and accepted by the patients. This item bank now needs to be developed into a measurement tool that groups items into domains and can be used in future research studies.

Development of a computerized adaptive testing system of the Fugl-Meyer motor scale in stroke patients

OBJECTIVE

To develop a computerized adaptive testing system of the Fugl-Meyer motor scale (CAT-FM) to efficiently and reliably assess motor function in patients with stroke.

DESIGN

First, a simulation study was used to develop and examine the psychometric properties of the CAT-FM. Second, a field study was employed to determine the administration efficiency of the CAT-FM.

SETTING

One medical center and 1 teaching hospital.

PARTICIPANTS

Patients' responses (n=301) were used for the simulation study; 49 patients participated in the field study.

INTERVENTIONS

Not applicable.

MAIN OUTCOME MEASURES

The 2 CAT-FM item banks (upper extremity and lower extremity) include 37 items from the original Fugl-Meyer scale. The reliability, validity, and responsiveness of the CAT-FM were determined by the simulation study.

RESULTS

Two stopping rules (reliability ≥0.9 or an increase of reliability <.01 after testing an item) were used. The simulation study showed that the CAT-FM had high reliability (≥.93 for upper-extremity and lower-extremity subscales) and concurrent validity (Pearson r≥.91 for the upper-extremity and lower-extremity subscales and motor scale). The responsiveness was moderate (standardized response mean for the upper extremity=.67, lower extremity=.79, and motor=.77) for the 226 patients who completed both assessments at 14 and 90 days after stroke. The field study found that, on average, the time needed to administer the CAT-FM was 242 seconds with 4.7 items.

CONCLUSIONS

The CAT-FM is an efficient, reliable, valid, and responsive clinical tool for assessing motor function in patients with stroke.

A proposed method to investigate reliability throughout a questionnaire

Background

Questionnaires are used extensively in medical and health care research and depend on validity and reliability. However, participants may differ in interest and awareness throughout long questionnaires, which can affect reliability of their answers. A method is proposed for "screening" of systematic change in random error, which could assess changed reliability of answers.

Methods

A simulation study was conducted to explore whether systematic change in reliability, expressed as changed random error, could be assessed using unsupervised classification of subjects by cluster analysis (CA) and estimation of intraclass correlation coefficient (ICC). The method was also applied on a clinical dataset from 753 cardiac patients using the Jalowiec Coping Scale.

Results

The simulation study showed a relationship between the systematic change in random error throughout a questionnaire and the slope between the estimated ICC for subjects classified by CA and successive items in a questionnaire. This slope was proposed as an awareness measure - to assessing if respondents provide only a random answer or one based on a substantial cognitive effort. Scales from different factor structures of Jalowiec Coping Scale had different effect on this awareness measure.

Conclusions

Even though assumptions in the simulation study might be limited compared to real datasets, the approach is promising for assessing systematic change in reliability throughout long questionnaires. Results from a clinical dataset indicated that the awareness measure differed between scales.

Comparison of Different Test Construction Strategies in the Development of a Gender Fair Interest Inventory Using Verbs

Three test construction strategies are described and illustrated in the development of the Verb Interest Test (VIT), an inventory that assesses vocational interests using verbs. Verbs might be a promising alternative to the descriptions of occupational activities used in most vocational interest inventories because they are context-independent, timesaving, and applicable across educational levels. Three test construction strategies are implemented and compared. The first construction method follows the rules of classical test theory (CTT), the second is within the framework of CTT as well but also takes gender differences in mean scores into account, and the third strategy is guided by item response theory (IRT) and controls for differential item functioning for men and women. The three VIT versions resulting from the different construction methods are compared regarding their construct and criterion validity. For practical use and career counseling, test development following the IRT approach seems most useful since it allows maximal occupational exploration and precise trait estimation.

Cautionary tales in the clinical interpretation of trials assessing therapy-induced changes in health status

Trials assessing the effects of therapies on symptoms, functional capacity, health-related quality of life and other aspects of health status are becoming more common in an era of chronic disease management. Such trials involve instruments for measuring health status whose reliability, validity and responsiveness need to be understood by clinicians and policy-makers in interpreting trial results. Deciding whether a treatment is clinically efficacious requires prior determination, based on empirical evidence, of what constitutes a minimal important difference (MID) between active treatment and control groups in the change in health status between study start and end. This MID should be used to calculate the sample size that will confer adequate power to detect a treatment effect if it truly exists. Many trials assessing health status have major methodological flaws: use of inappropriate or psychometrically unsound measurement instruments, lack of specification of MID, assumption that statistically significant results represent clinically significant treatment effects, and statement of conclusions inconsistent with observed results. This article provides guidance to clinicians in interpreting results of such trials in regard to clinical decision-making.

Applying an extended theoretical framework for data collection mode to health services research

Background

Over the last 30 years options for collecting self-reported data in health surveys and questionnaires have increased with technological advances. However, mode of data collection such as face-to-face interview or telephone interview can affect how individuals respond to questionnaires. This paper adapts a framework for understanding mode effects on response quality and applies it to a health research context.

Discussion

Data collection modes are distinguished by key features (whether the survey is self- or interviewer-administered, whether or not it is conducted by telephone, whether or not it is computerised, whether it is presented visually or aurally). Psychological appraisal of the survey request will initially entail factors such as the cognitive burden upon the respondent as well as more general considerations about participation. Subsequent psychological response processes will further determine how features of the data collection mode impact upon the quality of response provided. Additional antecedent factors which may further interact with the response generation process are also discussed. These include features of the construct being measured such as sensitivity, and of the respondent themselves (e.g. their socio-demographic characteristics). How features of this framework relate to health research is illustrated by example.

Summary

Mode features can affect response quality. Much existing evidence has a broad social sciences research base but is of importance to health research. Approaches to managing mode feature effects are discussed. Greater consideration must be given to how features of different data collection approaches affect response from participants in studies. Study reports should better clarify such features rather than rely upon global descriptions of data collection mode.