Objective: Patient-reported outcomes (PROs) are measures collected from a patient to determine how he/she feels or functions in regards to a health condition. Longitudinal PROs, which are collected at multiple occasions from the same individual, may be affected by response shift (RS). RS is a change in a person’s self-evaluation of a target construct. Latent variable models (LVMs) are statistical models that relate observed variables to latent variables (LV). LVMs are used to analyze PROs and detect RS. LVs are random variables whose realizations are not observable. Factor scores are estimates of LVs for each individual and can be estimated from parameter estimates of LVMs. Factor scoring methods to estimate factor scores include: Thurstone, Bartlett, and sum scores. This simulation study examines the effects of RS on factor scores used to test for change in the LV means and recommend a factor scoring method least affected by RS. Methods: Data from two time points were fit to three confirmatory factor analysis (CFA) models. CFA models are a type of LVM. Each CFA model had different sets of parameters that were invariant over time. The unconstrained (Uncon) CFA model had no invariant parameters, the constrained (Con) model had all the parameters invariant, and the partially constrained (Pcon) model had some of the parameters invariant over time. Factor scores were estimated and tested for change over time via paired t-test. The Type I error, power, and factor loading (the regression coefficient between an observed and LV) and factor score bias were estimated to determine if RS influenced the test of change over time and factor score estimation. Results: The results depended on the true LV mean. The Type I error and power were similar for all factor scoring methods and CFA models when the LV mean was 0 at time 1. For LV mean of 0.5 at time 1 the Type I error and power increased as RS increased for all factor scores except for scores estimated from the Uncon model and Bartlett method. The biases of the factor loadings were unaffected by RS when estimated from an Uncon model. The factor scores estimated from the Uncon model and the Bartlett and sum scores method had the smallest factor score biases. Conclusion: The factor scores estimated from the Uncon model and the Bartlett method was least affected by RS and performed best in all measures of Type I error, statistical power, factor loading and factor score bias. Estimating factor scores from PROs data that ignores RS may result in erroneous (or biased) estimates.

BACKGROUND. The study of patient reported outcomes (PROs) has seen an exponential increase in recent years. In order to be useful in practice, PRO questionnaires should be evaluated for validity, reliability, and responsiveness. Responsiveness, which assesses a questionnaire’s ability to capture changes in quality of life (QOL) when they occur, has not formally been evaluated in hemophilia-specific questionnaires. PRIMARY OBJECTIVE. To evaluate the responsiveness of the Patient Reported Outcomes, Burdens, and Experiences (PROBE) questionnaire in individuals living with hemophilia A or B following events of interest. SECONDARY OBJECTIVES. To evaluate the responsiveness of PROBE over periods in which no events occur. To explore the use of regression analysis in aiding interpretability. To assess the presence of response shift in the study population. METHODS. Participants will be asked to complete PROBE, as well as questions indicating changes in QOL, following a bleed or surgical intervention, and every 6 months. Responses will be evaluated using anchor-based and distribution-based approaches. OUTCOMES. Minimally important differences (MIDs) and minimally detectable changes (MDCs) will be calculated, graphically represented, and compared to determine a single or small range of MID values. STUDY IMPLICATIONS. Understanding responsiveness will provide increased interpretability of PROBE scores. Using an MID value, one can be confident that a change in PROBE score greater than the MID is beyond measurement error and indicates a change in QOL. This will allow for the use of PROBE in future research trials of drug effectiveness and can offer patients’ perspectives on their changes in QOL when switching to novel therapies. In addition, physicians may be able to use PROBE as a method of tracking and better understanding changes in their patients’ health statuses in the clinical setting. / Thesis / Master of Science (MSc) / This project hopes to identify the responsiveness of the Patient Reported Outcomes, Burdens, and Experiences (PROBE) Questionnaire. The responsiveness of a questionnaire is its ability to detect a change in health status when one has occurred. In order to measure whether PROBE can detect these changes, participants living with hemophilia A or B will be asked to fill out the questionnaire, as well as a few questions aimed at determining if their quality of life has changed, after they have a bleed or a surgery, as well as after 6 months. Collecting this information will help us understand how much the PROBE score needs to change in order for patients to consider a small but important change in health to have occurred. This will help with interpreting the PROBE score, which could then be used in research or in hemophilia clinics across Canada.

Osteoarthritis (OA) is one of the leading causes of global disability. Surgical intervention in the form of Total Knee Arthroplasty (TKA) has been established as an excellent treatment modality for people with OA who experience joint symptoms that have a substantial impact on their quality of life and are refractory to non-surgical treatment. In the 1970s the concept of implanting TKAs in mechanical alignment (MA) was developed as a compromise to confer mechanical advantage to the prosthesis, ignoring the patient's natural anatomy, to prevent early failure of the implant. Until now, this compromise has not been revisited. Satisfaction following TKA remains inferior to total hip arthroplasty. The cause of this dissatisfaction is not clear. Implant survival is no longer comparable to that of the early designs of TKA, and recent studies have suggested that deviation from neutral alignment does not have the detrimental effect on survivorship as previously thought. In an attempt to improve patient satisfaction following TKA a new technique has been developed whereby the prostheses are implanted in such a way as to recreate the alignment of the knee in the patient's pre-arthritic state. This has been termed natural or kinematic alignment (KA). This thesis examines the impact of KA in TKA with the primary hypothesis that TKA performed utilising KA would lead to improved functional outcome following surgery compared to that of MA. An initial single surgeon proof of concept case series of 25 patients was performed to look at the precision of new patient specific cutting blocks. The results suggested that the cutting blocks were accurate in producing the desired cuts. Following the proof of concept case series, a feasibility study was then performed comparing the new KA technique with the standard MA technique. The feasibility study familiarised the operating surgeons with the new technology in preparation for a Randomised Control Trial (RCT). A prospective blinded RCT was performed to compare the functional outcome of patients implanted with TKA in MA with that of KA. A total of 71 patients undergoing TKA were randomised to either MA (n=35) or KA (n=36). Preand post-operative hip knee ankle (HKA) radiographs were analysed. A number of patient reported outcome measures and functional tests were assessed pre-operatively, 6 weeks, 3 months, 6 months, and at 1 year post-operation. The cutting guides were accurate. There were no statistically significant differences between the MA and KA groups at 1 year. A cohort of post-menopausal women with unilateral osteoarthritis treated with TKA utilising the KA philosophy had dual energy x-ray absorptiometry scans 1.5 years post-operatively using a modified validated densitometric analysis protocol, to assess peri-prosthetic Bone Mineral Density (BMD). The contralateral knee was scanned so that relative bone mineral density could be calculated. Statistical analysis revealed no significant difference in relative peri-prosthetic bone mineral density due to variation in implant position with respect to the Lateral Distal Femoral Angle (LDFA) and the Medial Proximal Tibial Angle (MPTA). There was a significant correlation with overall HKA angle and the relative BMD under the medial side of the tibial tray. KA TKAs appear to have comparable short-term results to MA TKAs with no significant differences in function 1 year post-operatively. Overall HKA angle rather than the individual component position caused change in relative BMD under the tibial tray, therefore aiming for an anatomical joint line may improve kinematics without a detrimental effect on the implant. Further research is required to see if any theoretical long-term functional benefits of KA are realised or if there are any potential effects on implant survival.

Electronic data capture technologies, such as interactive voice response (IVR) systems, are emerging as important alternatives for collecting self-reported data. The purpose of this research was to assess the measurement equivalence between the original paper-based versions and the adapted interactive voice response (IVR) versions of the EQ-5D and the QLQ-C30. Furthermore, we examined the test-retest reliability of two consecutive administrations of the IVR versions of the EQ-5D and the QLQ-C30. The comparison of the paper and IVR versions of the EQ-5D was conducted utilizing a crossover design with subjects randomly assigned to one of two assessment orders: 1) paper then IVR or 2) IVR then paper. A convenience sample of in-treatment outpatient cancer clinic patients (n=139) were asked to complete each assessment two days apart. For the test-retest component, outpatient cancer clinic patients (n=127) were asked to complete the IVR-based EQ-5D twice, two days apart. The analyses tested for mean differences (paired t-test) and test-retest reliability (ICC).In the crossover analysis, ten of the fifteen mean differences analyzed for the scales and items of the QLQ-C30 were within the equivalence interval set a priori. The ICCs for the scales and items of the QLQ-C30 ranged from 0.698 to 0.899. Two of the items, insomnia and appetite loss, did not meet our threshold of being statistically different from an ICC of 0.70. The EQ-5D index score means were equivalent between paper and IVR, however the EQ VAS score differences were not wholly contained in the equivalence interval. The ICCs were above 0.890 for the index and the EQ VAS. In the test-retest analysis, the ICCs for the nine multi-item scales for the QLQ-C30 were all above 0.69, ranging from 0.698 to 0.891. Ten of the fifteen mean differences analyzed were within the equivalence interval set a priori. For the EQ-5D, the mean differences were wholly contained within the equivalence intervals for both the index and the EQ VAS and the ICCs were significantly different from 0.70. Overall, the IVR version of the questionnaires provided psychometrically equivalent results to those obtained on the original paper version and showed good stability over time.

Patient reported outcomes (PROs) are becoming increasingly important in cancer studies, particularly with the emphasis on patient centered outcome research. However, multiple PROs, using different scales, with different directions of favorability are often used within a trial, making interpretation difficult. To enhance interpretability, we propose the use of a standardized effect size graph, which shows all PROs from a study on the same figure, on the same scale. Plotting standardized effects with their 95% confidence intervals (CIs) on a single graph clearly showing the null value conveys a comprehensive picture of trial results. We demonstrate how to create such a graph using data from a randomized controlled trial that measured 12 PROs at two time points. The 24 effect sizes and CIs are shown on one graph and clearly indicate that the intervention is effective and sustained.

Les données rapportées par les patients sont maintenant fréquemment utilisées en recherche biomédicale. Ces instruments permettent la mesure de concepts subjectifs tels que la qualité de vie, les niveaux d’anxiété, de douleur, de fatigue. L’interprétation d’une différence de score au cours du temps était basée sur l’hypothèse que le sens des concepts et échelles restai stable au cours du temps dans l’esprit des individus. Cette hypothèse semble aujourd’hui dépassée. L’auto-évaluation d’un concept est maintenant comprise comme contingente de la représentation subjective qu’à un sujet du dit concept, cette représentation pouvant changer au cours du temps, surtout après avoir vécu un évènement de santé : ce phénomène est connu comme le « response shift ». Depuis la fin des années 1990s, l’investigation de ce phénomène est devenue un sujet d’intérêt majeur en psychométrie. Si des développements ont vu le jour, ce sujet reste récent et donc accompagné de débats variés que ce soit sur le plan théorique ou méthodologique. Aussi, l’objectif général de cette thèse est d’investiguer certaines problématiques méthodologiques et statistiques liées au response shift. Ce manuscrit est composé de trois travaux principaux : un état de l’art et une synthèse des travaux conduits à un niveau international depuis que le response shift est étudié, une étude pilote des performances de la procédure d’Oort (une méthode populaire de détection de response shift) par simulations et un travail théorique sur les liens entre response shift et complexité sémantique des concepts mesurés et items utilisés. / Patient-Reported Outcomes are increasingly used in health-related research. These instruments allow the assessment of subjective concepts such as Health-Related Quality of Life, anxiety level, pain or fatigue. Initially, the interpretation of a difference in score over time was based on the assumption that the meaning of concepts and measurement scales remains stable in individuals’ minds over time. This assumption has been challenged. Indeed, the self-assessment of a concept is now understood as a contingency of the subjective meaning a subject has of this concept, which can change over time especially as a result of a salient medical event: the “response shift” phenomenon. Since the end of the 1990s, researches on response shift phenomenon has become of prime interest in the field of health-related research. If developments have been made, it is still a young field with various scientific debates on a theoretical, methodological and statistical level. Thus, the broad objective of this thesis is to investigate some methodological and statistical issues regarding response shift concept, detection and integration into PRO data analyses. The manuscript is composed of three main works: a state of the art and synthesis of the works conducted at an international level since response shift phenomenon is investigated, a pilot study investigating the statistical performances of the Oort’s Procedure (a popular method of response shift detection using Structural Equation Modeling) by simulations and a theoretical work about the links between response shift occurrence and semantic complexity of concepts measured and items used.

INTRODUCTION: Immune checkpoint inhibitors are a new form of immunotherapy that has transformed the treatment landscape for an ever-increasing number of malignancies. While these medications utilize and enhance the immune system to treat malignancies, they can also have significant side effects, termed immune related adverse events, that in many ways resemble autoimmune disease states. One such example is inflammatory arthritis, which has been found to resemble a number of different presentations, including rheumatoid arthritis and seronegative spondyloarthropathies. In addition to these traditional inflammatory arthritis phenotypes, worsening of pre-existing arthritis is another subgroup of inflammatory arthritis that has previously not been considered in this population. Furthermore, while the effects of these autoimmune arthropathies on functionality is well-documented, it is not known whether there is a significant effect on functionality in patients that experience immune checkpoint inhibitor-induced arthritis. Given that patient reported outcomes are a validated and routinely utilized measure of functionality and quality of life, the Health Assessment Questionnaire, pain visual analogue scale and Patient Global Assessment were used to measure these outcomes following diagnosis. Our aim here is to explore the subtypes of inflammatory arthritis that result from this type of treatment and its overall effect on functionality and quality of life. METHODS: This study was a retrospective review of patients at one academic center who experienced an inflammatory arthritis resulting from immune checkpoint inhibition and required a referral to a rheumatologist for further work-up. Patients were evaluated in clinic at which time they also completed a Health Assessment Questionnaire as part of standard of care. Once patients were evaluated, their inflammatory arthritis was classified based on which clinical arthritis that it matched most closely, including polymyalgia rheumatica, rheumatoid arthritis, seronegative spondyloarthritis, or an exacerbation or osteoarthritis. Other demographic information such as gender, age, and race were also collected. Patient questionnaires were scored and compared to the type of inflammatory arthritis to assess for any correlations. RESULTS: We found 30 patients that had an inflammatory arthritis resulting from immune checkpoint inhibition, with 12 having a polyarthritis similar to rheumatoid arthritis, 11 patients having osteoarthritis exacerbation, 4 patients with a polymyalgia rheumatica arthritis phenotype and 3 patients with a spondyloarthopathy. In terms of the patient reported outcomes, the overall score was 0.57 ± 0.47, indicating that there was little effect of these arthropathies on overall functionality. The pain visual analogue scale had an average score of 41.8 ± 31.4 mm and the Patient Global Assessment had an overall score of 25.6 ± 26.7 mm. DISCUSSION: The overall results suggest that the inflammatory arthritis phenotypes do not significantly impact the functionality or quality of life of most patients who experience this side effect. Given that the use of immune checkpoint inhibitors will continue to grow, the overall need to better understand the resulting arthritis presentations is key. This is perhaps most true for those with pre-existing osteoarthritis, given the widespread nature of the disease in the general population and the prominence of the exacerbation as seen in our cohort.

Objectives: Patient-reported outcomes (PROs; self-report assessments) are increasingly important in evaluating medical care and treatment efficacy. Electronic administration of PROs via computer is becoming widespread. This article reviews the literature addressing whether computer-administered tests are equivalent to their paper-and-pencil forms. Methods: Meta-analysis was used to synthesize 65 studies that directly assessed the equivalence of computer versus paper versions of PROs used in clinical trials. A total of 46 unique studies, evaluating 278 scales, provided sufficient detail to allow quantitative analysis. Results: Among 233 direct comparisons, the average mean difference between modes averaged 0.2% of the scale range (e.g., 0.02 points on a 10-point scale), and 93% were within ±5% of the scale range. Among 207 correlation coefficients between paper and computer instruments (typically intraclass correlation coefficients), the average weighted correlation was 0.90; 94% of correlations were at least 0.75. Because the cross-mode correlation (paper vs. computer) is also a test-retest correlation, with potential variation because of retest, we compared it to the within-mode (paper vs. paper) test-retest correlation. In four comparisons that evaluated both, the average cross-mode paper-to-computer correlation was almost identical to the within-mode correlation for readministration of a paper measure (0.88 vs. 0.91). Conclusions: Extensive evidence indicates that paper- and computer-administered PROs are equivalent.

Patient-reported outcomes (PROs) produce meaningful information about patient-perceived health status reported directly by patients. Routine collection of PROs data is particularly important in chronic conditions, such as heart failure (HF). Major cardiovascular societies and regulatory agencies encouraged PRO inclusion in randomized controlled trials (RCTs), but PROs remain underutilized as a key outcome in these studies. In this systematic review, we aimed to evaluate temporal trends and explore trial characteristics associated with PRO inclusion in HF RCTs published in high-impact medical journals. We also assessed the quality of PRO reporting against the Consolidated Standards of Reporting Trials PRO extension. We found that over half of HF RCTs included a PRO. The proportion of RCTs with PROs increased significantly since 2000. A number of RCT characteristics such as multicentre; medium-sized (n = 51-250 participants); trials coordinated in Central and South America; and that tested health services, devices or surgery, exercise and rehabilitation interventions were independently associated with higher odds of PRO inclusion. The quality of PRO reporting was modest, with better reporting in RCTs with PROs a primary or co-primary endpoint. Consistent PRO inclusion and high-quality reporting are necessary to increase the utility of these findings by patients, clinicians, and health care policy makers. / Thesis / Master of Science (MSc)

Background: The management of cleft lip and/or palate (CL/P) includes multidisciplinary care beginning in infancy and continuing through to adulthood. Outcomes of cleft care have been difficult to measure because of the subjective nature of evaluating concepts such as appearance and speech. Including the patient perspective in outcome evaluation through the use of a patient-reported outcome (PRO) measure would provide a more accurate reflection of a patient’s status. The overall objective of this thesis is to show that through adherence to rigorous methods of development, a PRO measure can provide clinically meaningful outcome evaluation in cleft care. Methods: The first paper uses the qualitative method of interpretive description to define a conceptual framework to guide the development of a PRO measure for patients with CL/P, the CLEFT-Q. The second paper describes the protocol for the entire development of the CLEFT-Q. The third paper analyzes the results of the cross-sectional field-test of the CLEFT-Q scales to determine whether or not the CLEFT-Q is able to detect differences between specific cleft types. Results: The qualitative study included 138 patients with CL/P from six countries. The final conceptual framework contained thirteen concepts within the domains of appearance, facial function, and health-related quality of life. The second paper details the process of designing the CLEFT-Q scales. The field-test included 2,434 patients from thirty sites in twelve countries, and CLEFT-Q scores were found to vary with cleft type for all scales. Conclusions: PRO measures need to be rigorously designed in order to provide scientifically sound, clinically meaningful measurement. The CLEFT-Q is able to detect differences between patients with various cleft types, and will be a useful tool to provide the patient perspective in future outcome evaluation in cleft care. / Thesis / Doctor of Philosophy (PhD) / Measuring outcomes of treatment for cleft lip and/or palate (CL/P) should include the patient perspective. The objective of this thesis is to show that through rigorous methods of development, a patient-reported outcome (PRO) measure, the CLEFT-Q, can provide clinically meaningful evaluation of outcomes. First, 136 patients with CL/P from six countries were interviewed to learn what concepts related to having a cleft or its treatment are important to them. A conceptual framework was developed that informed the CLEFT-Q scales. Describing the methodology behind developing the CLEFT-Q then served to inform and engage members of the community. A field-test of the CLEFT-Q scales showed that in a sample of 2,434 patients with CL/P from twelve countries, CLEFT-Q outcomes varied in patients with different types of CL/P. The CLEFT-Q can be used to provide rigorous measurement of PROs in patients with CL/P in the future.