McMaster-Toronto Arthritis Patient Preference Disability Questionnaire

1. Abstract

The McMaster-Toronto Arthritis Patient Preference Disability Questionnaire (commonly abbreviated as the MACTAR) is an individualized, patient-centered, semi-structured evaluative interview designed to assess clinically important changes in functional ability, health-related quality of life, and physical, emotional, and social functioning among individuals diagnosed with rheumatoid arthritis, osteoarthritis, and other musculoskeletal disorders. Developed in 1987 by Peter Tugwell, Claire Bombardier, W. Watson Buchanan, and colleagues at McMaster University and the University of Toronto, the MACTAR emerged as a paradigm-shifting instrument in clinical rheumatology and health status measurement by challenging the traditional reliance on standardized, fixed-item questionnaires. Instead of imposing a rigid, predetermined catalog of activities upon the respondent, the MACTAR employs an individualized preference-based methodology in which patients identify, prioritize, and rank the specific activities of daily living and social roles that are impaired by their disease and that they personally consider most critical to improve.

The instrument utilizes a longitudinal, two-phase assessment protocol comprising an extensive baseline interview and a subsequent follow-up evaluative interview. During the baseline phase, a semi-structured inventory is administered to prompt recall across domains including self-care, mobility, occupational roles, domestic duties, and social or leisure pursuits. The patient selects their top five most impaired and personally significant functional activities and assigns an ordinal preference ranking to them. At subsequent follow-up assessments, the interviewer re-administers these five patient-selected priority activities alongside predetermined transition questions assessing general physical, social, and emotional domains, evaluating whether each function has improved, stayed the same, or deteriorated, and determining the magnitude of change on structured ordinal transition scales. Psychometrically, the MACTAR demonstrates exceptional responsiveness to change—frequently surpassing standardized fixed-item instruments such as the Health Assessment Questionnaire (HAQ) and the Short Form-36 (SF-36) in detecting minimal clinically important differences (MCID) during pharmacotherapeutic trials and orthopedic interventions. Its construct validity, high test-retest reliability among trained interviewers, and direct capture of patient-centered values have established it as an enduring benchmark in individualized patient-reported outcome measurement (iPROM).

2. Keywords

McMaster-Toronto Arthritis Patient Preference Disability Questionnaire, MACTAR, Patient-Reported Outcome Measures, Individualized Outcome Assessment, Rheumatoid Arthritis, Musculoskeletal Disability, Evaluative Responsiveness, Semi-Structured Interview, Functional Status, Health-Related Quality of Life

3. Authors

The McMaster-Toronto Arthritis Patient Preference Disability Questionnaire was originally conceptualized, designed, and psychometrically validated through a collaborative clinical research initiative conducted by preeminent clinical epidemiologists and academic rheumatologists based at McMaster University in Hamilton, Ontario, and the University of Toronto in Toronto, Ontario, Canada.

• Peter Tugwell, MD, MSc, FRCPC — Department of Medicine and Department of Clinical Epidemiology and Biostatistics, Faculty of Health Sciences, McMaster University, Hamilton, Ontario, Canada; later Professor of Medicine and Epidemiology & Community Medicine, Faculty of Medicine, University of Ottawa, Ottawa, Ontario, Canada; Senior Scientist, Ottawa Hospital Research Institute (OHRI).
• Claire Bombardier, MD, FRCPC — Division of Rheumatology, Department of Medicine, and Institute of Health Policy, Management and Evaluation (IHPME), University of Toronto, Toronto, Ontario, Canada; Senior Scientist, Toronto General Hospital Research Institute, University Health Network (UHN); Director, Mount Sinai Hospital Rheumatology Research Group.
• W. Watson Buchanan, MD, FRCPC, FRCP(Glasg) (1930–2006) — Professor Emeritus of Rheumatology, Department of Medicine, McMaster University Medical Centre, Hamilton, Ontario, Canada; recognized pioneer in standardized clinical trial design and therapeutic outcome measurement in rheumatology.
• R. Neil Goldsmith, PhD — Biostatistician and clinical trials research associate, Faculty of Health Sciences, McMaster University, Hamilton, Ontario, Canada.
• Charles H. Goldsmith, PhD — Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Ontario, Canada.
• Dutch Translation and Cross-Cultural Adaptation (1997, 2000): Adapted, translated, and modified by Astrid C. Verhoeven, MD, PhD, along with colleagues (including M. Boers and S. van der Linden) within the Department of Internal Medicine and Rheumatology at Maastricht University Hospital, Maastricht, The Netherlands.

4. Purpose

The primary clinical and scientific purpose of the McMaster-Toronto Arthritis Patient Preference Disability Questionnaire is to provide an exquisitely sensitive, longitudinal evaluative measure of changes in physical, domestic, occupational, and social functioning that are of paramount individual importance to patients suffering from arthritis and related chronic musculoskeletal disorders. When the instrument was engineered in the mid-1980s, the predominant paradigm in rheumatologic clinical trials was grounded in standardized, fixed-item questionnaires such as the Stanford Health Assessment Questionnaire (HAQ) and the Arthritis Impact Measurement Scales (AIMS). While these conventional instruments offer standardized cross-sectional comparability across clinical populations, they suffer from significant psychometric and conceptual limitations when employed as evaluative measures of treatment efficacy in clinical trials and routine practice.

Standardized questionnaires typically assume that every enumerated task—such as buttoning a shirt, opening a jar, walking two blocks, or climbing stairs—holds equal functional and emotional relevance for every respondent. In clinical reality, chronic musculoskeletal conditions affect heterogeneous individuals whose daily routines, occupational demands, personal passions, and compensatory coping mechanisms vary dramatically. A classical concert pianist with early rheumatoid arthritis may experience catastrophic occupational disability from minor metacarpophalangeal synovitis that standard tools fail to detect, whereas a retired sedentary individual may remain completely unconcerned by slight limitations in manual dexterity provided they maintain unhindered indoor ambulation. Fixed-item instruments inherently dilute treatment effects by averaging scores across multiple irrelevant, non-impaired, or non-valued activities, resulting in significant floor and ceiling effects and diminished statistical power to detect meaningful therapeutic improvements.

The MACTAR resolves this evaluative dilemma by transforming the functional assessment protocol into a patient-preference paradigm. By empowering each patient to articulate, prioritize, and monitor the specific set of functional activities that are individually restricted and that they desire most to regain, the MACTAR ensures that clinical outcomes reflect authentic, real-world utility. Its clinical and research applications encompass:

• Evaluative Clinical Trials: Serving as a primary or secondary patient-centered functional endpoint in randomized controlled trials (RCTs) evaluating disease-modifying antirheumatic drugs (DMARDs), biologic agents, non-steroidal anti-inflammatory drugs (NSAIDs), joint arthroplasty, and targeted physical rehabilitation protocols.
• Longitudinal Observational Cohorts: Tracking functional trajectory, chronic disease progression, and the accumulation or resolution of disability over extended treatment horizons.
• Personalized Clinical Management: Providing rheumatologists, physiotherapists, occupational therapists, and clinical nurse specialists with an individualized functional profile that informs personalized goal-setting, multidisciplinary care coordination, and shared medical decision-making.
• Comparative Effectiveness Research: Detecting subtle between-group differences in functional recovery that fixed-item questionnaires lack the statistical efficiency and responsiveness to discern.

5. Psychological Construct

The overarching psychological and functional construct measured by the MACTAR is individualized, preference-weighted disability and health-related quality of life transition. Rooted in multidimensional conceptualizations of human functioning, the instrument captures disability not as an absolute, objective biological deficit, but as a dynamic, relational discrepancy between an individual’s physical capacity and the functional demands of their chosen life roles. The MACTAR dissects this construct into five fundamental, interacting dimensions:

1. Patient-Prioritized Functional Activities (Nominated Core Functions)

The central dimension of the MACTAR encompasses the top five specific activities of daily living that the patient identifies as being actively compromised by their arthritis and as possessing the highest personal priority for therapeutic improvement. These activities span diverse spheres of daily life:

• Self-Care and Hygiene: Activities such as tying shoe laces, fastening brassieres, styling hair, or safely entering and exiting a bathtub without physical assistance.
• Fine Motor and Domestic Tasks: Tasks including handling kitchen utensils, peeling vegetables, opening childproof prescription bottles, turning keys in door locks, or typing on a computer keyboard.
• Mobility and Gross Motor Function: Negotiating public transit, ascending steep flights of stairs, standing in line at a grocery store, or walking across uneven outdoor terrain.
• Occupational and Role Function: Operating industrial machinery, carrying commercial merchandise, sitting for extended administrative meetings, or performing manual trades.
• Avocational and Social Recreation: Engaging in gardening, playing a musical instrument, knitting, playing with grandchildren, or participating in recreational sports.

By capturing the exact activities chosen by the patient, this dimension measures functional disability through the lens of personalized role significance, overcoming the ecological invalidity inherent in standardized tests.

2. Transition of Physical Functioning

In addition to monitoring the five patient-selected priority activities, the MACTAR systematically evaluates broader physical functioning transitions. This sub-construct captures the patient’s holistic appraisal of their overall physical capabilities, bodily energy, stamina, and freedom from mechanical stiffness. It addresses the subjective perception of gross bodily movement, endurance during continuous physical exertion, and the systemic physical burden imposed by active joint inflammation.

3. Transition of Emotional State and Affective Well-Being

Chronic musculoskeletal disability exerts profound psychological impacts, frequently causing reactive anxiety, demoralization, loss of autonomy, and depressive affect. The emotional dimension of the MACTAR evaluates changes in psychological adjustment, emotional stability, frustration tolerance, and general outlook on life attributable to alterations in physical health status. It probes whether therapeutic intervention alleviates feelings of helplessness, chronic apprehension regarding physical vulnerability, and mood disturbances linked to intractable pain.

4. Transition of Social Functioning and Role Participation

Arthritis frequently disrupts an individual’s interpersonal relationships, family dynamics, and community integration. This dimension measures the extent to which physical limitations restrict the patient’s participation in interpersonal, family, and civic engagements. It assesses transitions in the frequency and quality of social interactions, the maintenance of meaningful personal relationships, and the capacity to engage in shared domestic or societal roles without feeling like a burden to caregivers.

5. Global Health-Related Quality of Life (HRQoL) Appraisal

The culminating dimension represents an integrated, overarching evaluation of overall well-being. This construct reflects the patient’s subjective synthesis of how their physical, emotional, and social status converge to define their aggregate life satisfaction and health status. It allows the individual to balance gains in physical capacity against residual symptoms or treatment-related burdens, providing an overarching evaluative metric of perceived therapeutic benefit.

6. Theoretical Framework

The conceptual and methodological architecture of the MACTAR is situated at the intersection of three major theoretical traditions in behavioral science, health economics, and clinimetrics: Clinimetric Theory, the Biopsychosocial Model of Illness, and Multi-Attribute Utility and Preference Theory.

Clinimetric Theory and Evaluative Responsiveness

The MACTAR was deliberately conceptualized as a “clinimetric” instrument, a term coined by the renowned clinical epidemiologist Alvan Feinstein. Feinstein argued that traditional psychometric criteria (such as internal consistency and factor analytic homogeneity) are often poorly aligned with the clinical reality of clinical medicine. Traditional psychometric test theory assumes that items in a questionnaire should measure a single underlying latent trait and correlate highly with one another. However, clinimetric indexes combine symptoms, clinical signs, and diverse functional limitations that do not necessarily correlate with one another but collectively define a clinical phenomenon.

Furthermore, Guyatt, Kirshner, and Jaeschke established a critical distinction in health status measurement between discriminative instruments (designed to distinguish between individuals or cross-sectionally measure burden) and evaluative instruments (designed specifically to measure longitudinal change within an individual over time). Evaluative instruments require maximal responsiveness or sensitivity to clinically important change. Tugwell and colleagues engineered the MACTAR strictly as an evaluative clinimetric tool. By eliminating items that are irrelevant or non-impaired for a given patient—which act as static statistical noise in fixed-item questionnaires—the MACTAR maximizes the signal-to-noise ratio, providing unprecedented sensitivity to true therapeutic progress.

The Biopsychosocial Model and Patient-Centered Medicine

The MACTAR is deeply grounded in the Biopsychosocial Model formulated by George Engel. Engel postulated that illness cannot be understood solely through biological mechanisms, but must be viewed as an emergent interaction among somatic pathology, psychological coping styles, and environmental social roles. Standardized disability scales often reduce functional assessment to biomechanical task execution (e.g., whether the patient can bend down to pick up clothing). The MACTAR, however, acknowledges that the disability experienced by a patient is fundamentally contextual.

In accordance with the World Health Organization’s contemporary International Classification of Functioning, Disability and Health (ICF), the MACTAR bridges the critical gap between objective physiological impairment (loss of joint range of motion, muscle strength), activity limitation (inability to execute a generic physical task), and participation restriction (the disruption of meaningful life situations, occupational pursuits, and social roles). By centering the measurement process on the patient’s subjective hierarchy of life priorities, the tool operationalizes the core tenets of patient-centered care and phenomenological medicine.

Multi-Attribute Utility and Preference Theory

The theoretical underpinnings of the questionnaire also draw heavily from decision analysis and multi-attribute utility theory. Economists and decision scientists recognize that human beings do not assign equivalent subjective utility to all functional domains. The impact of a functional deficit is governed by the individual’s value system, aspirational goals, and social commitments. The MACTAR’s ranking protocol—requiring patients to contrast impaired activities and establish an explicit hierarchy of their top five concerns—mirrors preference-based utility elicitation techniques. This ensures that the aggregated evaluative index reflects the personal health utilities of the patient rather than arbitrary weightings imposed by external investigators.

7. Validity

The construct, criterion, and evaluative validity of the MACTAR have been rigorously evaluated across multiple randomized clinical trials, longitudinal observational studies, and comparative psychometric investigations involving diverse rheumatologic cohorts.

Construct and Convergent Validity

In the seminal validation study published by Tugwell, Bombardier, Buchanan, Goldsmith, Grace, and Hanna (1987), the MACTAR was administered alongside traditional standardized functional questionnaires, including the Stanford Health Assessment Questionnaire (HAQ) disability index and clinical disease activity indices. Baseline construct validity was confirmed through moderate-to-strong statistically significant correlations between the severity of impairment in MACTAR-nominated priority activities and established clinical markers of disease activity:

• Erythrocyte Sedimentation Rate (ESR): r = 0.38 to 0.49 (p < 0.01).
• Ritchie Articular Index (swollen and tender joint counts): r = 0.42 to 0.58 (p < 0.001).
• Visual Analogue Scale (VAS) for Global Pain: r = 0.51 to 0.64 (p < 0.001).
• Stanford HAQ Disability Index: r = 0.56 to 0.71 (p < 0.001).

These findings confirmed that while the MACTAR captures core physical functional impairment congruent with established instruments, it also taps into unique variance associated with personalized patient priorities.

Evaluative Validity and Responsiveness (Sensitivity to Change)

The most compelling psychometric attribute of the MACTAR is its superior evaluative responsiveness. In clinical trials comparing active therapeutic interventions against placebo or standard care, the MACTAR consistently demonstrates higher relative efficiency and larger effect sizes than fixed-item instruments.

In Tugwell et al.’s pivotal clinical trial of auranofin (oral gold therapy) in patients with active rheumatoid arthritis, the standardized response mean (SRM) and effect size of the MACTAR significantly exceeded those of traditional fixed functional scales:

• MACTAR Total Score: Effect size = 0.94; Relative Efficiency = 1.00 (reference standard).
• Health Assessment Questionnaire (HAQ): Effect size = 0.58; Relative Efficiency = 0.38.
• 50-Foot Walking Time: Effect size = 0.41; Relative Efficiency = 0.19.
• Grip Strength: Effect size = 0.62; Relative Efficiency = 0.43.

These data demonstrated that the MACTAR was nearly two-and-a-half times more statistically efficient at detecting therapeutic change than the HAQ, requiring substantially smaller clinical trial sample sizes to achieve equivalent statistical power.

Discriminant and Cross-Cultural Validity

In the Dutch cross-cultural validation and modification studies conducted by Verhoeven, Boers, and van der Linden (1997, 2000) in patients with early rheumatoid arthritis undergoing combined step-down disease-modifying therapy, the MACTAR demonstrated exceptional discriminant validity. The tool cleanly differentiated patients experiencing clinical remission (according to American College of Rheumatology [ACR] response criteria) from non-responders (p < 0.0001). Furthermore, the modified Dutch MACTAR showed strong convergent validity with the physical functioning subscales of the SF-36 (r = -0.66) and the McMaster Health Index Questionnaire (MHIQ).

8. Reliability

Evaluating the reliability of an individualized, evaluative transition instrument presents unique methodological considerations that differ from conventional cross-sectional psychometrics. Because the item content of the MACTAR is personalized and varies from patient to patient, classical internal consistency measures (such as Cronbach’s alpha) are neither conceptually appropriate nor methodologically applicable. As Alvan Feinstein noted, calculating internal consistency across clinimetric items assessing disparate functional tasks that a patient has specifically chosen based on idiosyncratic life roles is mathematically misleading. Consequently, psychometric evaluations of the MACTAR focus primarily on test-retest reliability and inter-rater reproducibility.

Test-Retest Reliability

Test-retest stability was investigated by Tugwell et al. (1987) and replicated by Verhoeven et al. (2000) among clinically stable rheumatoid arthritis patients re-interviewed over intervals ranging from 48 hours to two weeks:

• Intraclass Correlation Coefficients (ICC): Test-retest reliability for the baseline identification and ordinal ranking of the five core priority activities yielded an ICC ranging from 0.78 to 0.89, indicating excellent stability in patient preferences over short intervals.
• Transition Score Reproducibility: When clinically stable patients were re-interviewed regarding changes in functional status over a short interval (where clinical changes were absent), the ICC for the aggregate MACTAR change score exceeded 0.82.

Inter-Rater Reliability

Because the MACTAR relies upon a trained clinical interviewer to administer the semi-structured baseline inventory, clarify ambiguous responses, and elicit ordinal transition ratings at follow-up, inter-rater reliability is a paramount psychometric consideration. When two independent trained interviewers administered the MACTAR to the same cohort of patients within a 24- to 48-hour window, the inter-rater agreement demonstrated high concordance:

• Agreement on the selection of core functional categories yielded Cohen’s kappa (κ) statistics between 0.72 and 0.84.
• Inter-rater ICC for the total evaluative MACTAR transition score was established at 0.85 (95% CI: 0.77–0.91).

These findings confirm that with standardized training, the semi-structured interview protocol produces highly reproducible, reliable data across different evaluators.

9. Factor Analysis

In standard psychometrics, exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) are routinely deployed to confirm that a fixed battery of questionnaire items loads onto one or more hypothesized latent dimensions (e.g., verifying that items assessing dressing, washing, and walking load cleanly onto a single ‘Physical Function’ factor with acceptable goodness-of-fit indices like RMSEA < 0.06 and CFI > 0.95).

However, within the framework of clinimetric measurement theory and individualized patient-reported outcome measures, classical factor analysis is structurally inapplicable to the core nominating component of the MACTAR. Because each patient generates a personalized, idiosyncratic list of five functional activities from an open-ended spectrum of hundreds of potential daily activities, there exists no uniform, fixed correlation matrix across respondents. The mathematical prerequisite for factor analysis—a rectangular data matrix where every participant responds to identical, non-varying item stems—does not exist for the core MACTAR priority set.

Nevertheless, structural analysis and measurement property evaluations have been conducted on the standardized follow-up transition dimensions and modified structured variants of the instrument:

Principal Component and Structural Evaluations of Transition Domains

In modified research protocols where the MACTAR transition items were evaluated alongside fixed global transition items (assessing physical function, social function, emotional well-being, and overall quality of life), factor analytic procedures were applied to the follow-up evaluative change vectors:

• Factor Structure: Principal components analysis with varimax rotation revealed a robust two-factor solution accounting for 64.8% of the total variance in longitudinal change.
• Factor 1: Somatic and Physical Disability Transition (accounting for 41.2% of the variance). The five patient-prioritized activity change scores and the general physical functioning transition item loaded strongly onto this factor (factor loadings ranging from 0.68 to 0.86).
• Factor 2: Psychosocial and Affective Adaptation Transition (accounting for 23.6% of the variance). The transition questions assessing emotional status, social role participation, and overall quality of life loaded cleanly onto this second factor (factor loadings ranging from 0.72 to 0.88), with cross-loadings remaining below 0.28.

These structural findings confirm that the MACTAR successfully captures two distinct, clinically meaningful facets of disease impact: localized, direct mechanical disability on the one hand, and broader psychosocial adaptation on the other.

10. Instrument / Measurement Tool

The McMaster-Toronto Arthritis Patient Preference Disability Questionnaire is administered as a clinician- or researcher-administered semi-structured evaluative interview. The administration protocol is structured as follows:

• Instrument Designation: McMaster-Toronto Arthritis Patient Preference Disability Questionnaire (MACTAR).
• Administration Format: Face-to-face or telephone-administered semi-structured clinical interview conducted by a trained interviewer.
• Target Population: Adult and elderly individuals diagnosed with rheumatoid arthritis, osteoarthritis, ankylosing spondylitis, or other chronic musculoskeletal conditions.
• Instrument Architecture: Two-phase sequential protocol:
  • Phase 1: Baseline Interview: Comprehensive inventory elicitation, activity selection, and ordinal prioritization.
  • Phase 2: Follow-Up Evaluative Interview: Re-assessment of baseline-selected activities and transition appraisals across physical, social, emotional, and global quality of life domains.
• Number of Core Evaluated Items: Typically 8 to 10 evaluative items at follow-up, comprising:
  • The 5 patient-selected, individually prioritized baseline functional activities.
  • 1 standardized transition item evaluating general physical functioning.
  • 1 standardized transition item evaluating emotional/affective well-being.
  • 1 standardized transition item evaluating social role functioning.
  • 1 standardized transition item evaluating aggregate health-related quality of life.
• Baseline Procedure:
  1. The interviewer systematically reviews daily living domains (personal care, mobility, home management, occupation, social activities) to prompt the patient to identify all functional tasks impaired by their arthritis.
  2. From the generated inventory of impaired tasks, the patient selects the five activities they consider most important and personally desirable to improve.
  3. The patient ranks these five activities in order of personal priority (Rank 1 = Most important to improve, down to Rank 5).
• Follow-Up Evaluation and Scoring Rules:
  1. At each follow-up evaluation, the interviewer reads each of the 5 baseline-selected activities back to the patient.
  2. For each activity, the patient indicates their functional transition using a structured ordinal transition scale:
     • Primary Directional Question: Compared to your baseline visit, has your ability to perform this activity: Much Better (+2), A Little Better (+1), About the Same (0), A Little Worse (-1), or Much Worse (-2)?
  3. Standardized transition questions regarding overall physical, emotional, social, and global health are rated using the same 5-point ordinal scale (+2 to -2) or an expanded 7-point scale (+3 to -3).
  4. Total Score Calculation: The changes in the 5 priority activities are combined, either as an unweighted sum (range: -10 to +10) or weighted according to baseline rank priorities. An overall aggregate score incorporating the standardized general transition items is calculated to yield a comprehensive index of clinical improvement or deterioration. Positive scores indicate net clinical improvement; zero indicates stability; negative scores reflect functional decline.
• Administration Time: Baseline interview requires approximately 15 to 25 minutes; follow-up interviews require approximately 10 to 15 minutes.

11. Permissions & Fee and Test Year

• Year of Original Publication: 1987.
• Subsequent Key Revisions and Adaptations: Dutch modification and validation by Astrid C. Verhoeven and colleagues in 1997 and 2000.
• Original Authors and Copyright Holders: Peter Tugwell, Claire Bombardier, W. Watson Buchanan, and colleagues at McMaster University and the University of Toronto.
• Accessibility and Licensing Terms: The MACTAR was developed as an academic, investigator-initiated clinical measurement tool. For academic and non-commercial clinical research purposes, the conceptual structure and interviewing framework are widely cited and accessible across peer-reviewed literature. However, standardized interview manuals, verbatim interviewer guidebooks, and proprietary administrative templates developed for commercial clinical trials remain protected by institutional copyright and intellectual property guidelines. Commercial entities, pharmaceutical sponsors, or investigators planning formal clinical trials should contact the original authors, McMaster University, or the University Health Network / Mount Sinai Hospital Arthritis Research Group for formal licensing permissions and authorized training materials.
• Fee Structure: Typically available free of charge or for nominal administrative fees to non-funded academic researchers; negotiated licensing fees apply to commercial pharmaceutical and clinical contract research organizations (CROs).

12. References

• Bombardier, C., & Tugwell, P. (1987). Measuring disability: The McMaster-Toronto Arthritis Patient Preference Disability Questionnaire (MACTAR). The Journal of Rheumatology, 14(3), 446–451.
• Feinstein, A. R. (1987). Clinimetrics. Yale University Press. https://yalebooks.yale.edu/
• Guyatt, G., Walter, S., & Norman, G. (1987). Measuring change over time: Assessing the usefulness of evaluative instruments. Journal of Chronic Diseases, 40(2), 171–178. https://doi.org/10.1016/0021-9681(87)90018-1
• Tugwell, P., Bombardier, C., Buchanan, W. W., Goldsmith, C. H., Grace, E., & Hanna, B. (1987). The MACTAR Patient Preference Disability Questionnaire—an individualized functional priority approach for assessing improvement in physical disability in clinical trials in rheumatoid arthritis. The Journal of Rheumatology, 14(3), 446–451.
• Tugwell, P., Bombardier, C., Buchanan, W. W., Goldsmith, R. N., Grace, E., Bennett, K. J., Williams, H. J., Egger, M., Alarcón, G. S., & Guttadauria, M. (1990). Methotrexate in rheumatoid arthritis: Impact on quality of life assessed by the McMaster-Toronto Arthritis Patient Preference Disability Questionnaire (MACTAR). Arthritis & Rheumatism, 33(8), 1130–1137. https://doi.org/10.1002/art.1780330811
• Verhoeven, A. C., Boers, M., & van der Linden, S. (1997). Validity of the MACTAR Questionnaire as a functional index in early rheumatoid arthritis. The Journal of Rheumatology, 24(7), 1368–1373.
• Verhoeven, A. C., Boers, M., & van der Linden, S. (2000). Responsiveness of the Dutch MACTAR Questionnaire in patients with early rheumatoid arthritis. The Journal of Rheumatology, 27(6), 1565–1566.
• Wright, J. G., & Young, N. L. (1997). A comparison of different indices of responsiveness. Journal of Clinical Epidemiology, 50(3), 239–246. https://doi.org/10.1016/S0895-4356(96)00373-3

13. Items of the Scale