Jason Sutherland

Chronic rhinosinusitis with nasal polyps (CRSwNP) symptoms are commonly treated with systemic corticosteroids (SCS) despite ambiguous clinical guidelines recommending the therapy’s use. SCS are strong anti-inflammatory medications with important related adverse outcomes.This thesis identified a cohort of CRSwNP patients and pharmacy claims to investigate dispensing patterns of SCS, factors associated with increased use, prescribing practices across healthcare providers, and association of SCS use with the incidence of adverse outcomes. Two studies were derived from this work. The first study was mainly descriptive and revealed variations in SCS prescribing practices among otolaryngologists and potential inappropriate SCS prescribing by general practitioners in the management of CRSwNP. Performing multinomial logistic regression, analyses showed patients residing in small population centres and rural areas were more likely to receive higher SCS doses. The second study, using time to event regression analysis, found that SCS use was associated with a higher risk of SCS-related adverse outcomes. In sensitivity analysis for individual outcomes, statistically significant associations were identified between SCS users compared with non-users and increased risk of pneumonia, hypertension, and cardio/cerebrovascular disease. In the study’s cumulative dose analysis, most adverse health outcomes showed increased risk among SCS users exposed to 0.5g to 1g compared with those exposed to 0.5g. A clear dose-response pattern with increasing cumulative dose was not observed. These findings underscore the need for standardized clinical guidelines, monitoring strategies for the prevention and early detection of adverse outcomes, and further research to identify safe and effective use of SCS among patients with CRSwNP.

View record

Background and purpose. The discipline of quality improvement emphasizes the use of data to improve specific and measurable attributes of performance. Using the example of carotid artery endarterectomy (CEA) for carotid artery stenosis, this thesis examined the use of hospital discharge and physician claims data for two quality improvement use cases. In Study 1, claims data were used to assemble patient cohorts based on symptomatic status at time of CEA. In Study 2, claims data were used to identify process-related causes of delayed symptomatic CEA.Methods. A single hospital’s administrative database was used to assemble a retrospective cohort of participants who had undergone CEA. Chart review data were linked with physician claims and hospital discharge data. In Study 1, a standard method of classification by hospital discharge diagnosis was compared to classification using physician claims and hospital discharge data. In Study 2, delays related to the selection of medical care activities, such as physician visits and diagnostic tests, were investigated using linear regression models of waiting time for surgery and K-means clustering for patterns of activity co-occurrence.Results. We identified 971 participants undergoing CEA at the Vancouver General Hospital (Vancouver, Canada) between January 1, 2008, and December 31, 2016. For Study 1, 729 people met inclusion/exclusion criteria and were included in diagnostic classification models (615 training, 114 test). Classification of symptomatic status using hospital discharge diagnosis codes was 32.8% (95% CI 29% – 37%) sensitive and 98.6% specific (96% – 100%). At matched 98.6% specificity, models that incorporated physician claims data were significantly more sensitive: elastic net 69.4% (59% – 82%) and random forest 78.8% (69% – 88%). In Study 2, K means clustering and linear regression analyses of process delay yielded similar results: early evaluation by emergency physicians or neurologists was associated with reduced delay, whereas carotid ultrasonography and post-imaging follow up with general practitioners or eye specialists was associated with greater delay.Conclusion. Physician claims data can be used within quality improvement projects for cohort definition and process analysis use cases.

View record

Hysterectomy is one of the most common surgical procedures among patients. Despite the high volume of this surgery, it is not well understood what patient and clinical factors are associated with a patient’s health-related quality of life before and after surgery. This thesis used a prospectively recruited cohort of patients scheduled for, or who had received, a hysterectomy for non-cancerous conditions to investigate: 1) the health of patients awaiting hysterectomy and factors associated with self-reported health, 2) patient and clinical factors associated with postoperative health, and 3) perceptions of shared decision-making and its association with postoperative health. These investigations were divided into three sub-studies. Patient-reported outcomes were used to measure pelvic floor health, sexual function, depression, and pain. Multi-variable regression analyses measured the association of age, comorbidities, socioeconomic factors, surgical approach, surgical indication, and shared decision-making with self-reported health. The first investigation found that participants that were younger, had more comorbidities, or had worse symptoms of pelvic health were more likely to report depression and pain before their surgery. The second investigation found that most participants reported improvements in health after their hysterectomy, however, those with lower socioeconomic status were more likely to report more pain. Participants that received a hysterectomy for their endometriosis or had received a hysterectomy through the abdominal approach, were more likely to report postoperative depressive symptoms. The last investigation found that almost half of hysterectomy patients reported less than optimal scores on their shared decision-making process with their surgeon before their hysterectomy. Those with lower scores were more likely to report poor pelvic health postoperatively.

View record

This thesis is focused on characterizing differences in pharmacological treatment among immigrant and non-immigrant children with Attention-Deficit/Hyperactivity Disorder (ADHD). Multimodal approaches combining psychosocial and pharmacological interventions are recommended for the treatment of ADHD. Among pharmacological treatments, stimulant medications are among the most common and efficacious drug therapy used to manage ADHD symptoms. However, there are concerns around increasing use of medications, such as atypical antipsychotics, that lack substantive evidence and are not currently approved by Health Canada for the treatment of ADHD. This thesis has two main studies: First, scoping review methodology was used to examine longitudinal trends, prevalence, tolerability, and safety of atypical antipsychotics among children with ADHD. No studies were identified relating to atypical antipsychotic use specific to immigrant children with ADHD, even though this review found that many children with ADHD still receive atypical antipsychotics. Atypical antipsychotic medications are often used to manage psychiatric comorbidity and/or aggressive and impulsive symptoms associated with ADHD. However, there is little evidence regarding the safety and tolerability of atypical antipsychotics to support the use of these medications as a first-line intervention among children diagnosed with ADHD. Second, differences in first-line stimulant and atypical antipsychotic prescribing practices were examined among newly diagnosed immigrant and non-immigrant children with ADHD, living in British Columbia, Canada. Immigrant children were less likely than non-immigrant children to be dispensed a stimulant medication prescription in the 30-days following an incident diagnosis of ADHD. Among children, being older, male, and receiving treatment from a psychiatrist was associated with being dispensed an atypical antipsychotic prescription as a first-line intervention within 30-days of diagnosis. A latent class analysis (LCA) was conducted to account for sources of variability that may be associated with ADHD treatment. A high proportion of children with ADHD were not dispensed any prescription medications. Non-migrant boys had a 68% increased likelihood of being treated with atypical antipsychotics compared to their migrant counterparts. Together, the studies incorporated in this thesis underline the importance of examining relationships between structural determinants of health, health equity, and ADHD-related prescribing practices.

View record

Background:Many Coronavirus Disease 2019 (COVID-19) survivors report long-term sequelae. However, few studies have measured patient-reported outcomes and compared them to those of patients who tested negative for severe acute respiratory syndrome coronavirus-2 (SARS-COV-2). This study compares the long-term physical and mental health outcomes of patients presenting to emergency departments who tested positive for SARS-COV-2 with those who tested negative.Methods:This study enrolled consecutive eligible patients presenting to emergency departments participating in the Canadian COVID-19 Emergency Department Rapid Response Network (CCEDRRN) between March 1st 2020 and July 14th 2021. Patients tested for SARS-COV-2 were eligible. Consecutive SARS-COV-2 positive patients were consented for follow-up, and matched with test-negative controls from the same hospital and date. The outcome measures were the Veterans RAND physical health component score (PCS) and mental health component score (MCS). The PCS and MCS of propensity score matched patients were analyzed using linear mixed effects models. Risk factors for PCS and MCS were modelled using linear regression.Results:Our cohort included 1170 SARS-COV-2 positive patients and 3716 test-negative controls. Comparing the groups, the adjusted mean difference in PCS was 0.50 (95%CI: -0.36, 1.36) and -1.01 (95%CI: -1.91, -0.11) for MCS. A World Health Organization Ordinal Outcome Score of 6-7, representing severe SARS-COV-2 disease, was the strongest predictor of PCS (β=-7.4; 95%CI: -9.8, -5.1). Prior mental health illness was the strongest predictor of MCS (β=-5.4; 95%CI: -6.3, -4.5). Conclusion:The mean PCS was similar among SARS-COV-2 positive and negative participants tested under similar circumstances, while mean MCS was worse among SARS-COV-2 positive participants. The mental health sequelae of COVID-19 should be considered when developing long-term support programs for survivors.

View record

End-stage ankle arthritis (ESAA) is a debilitating condition that negatively affects patients’ health-related quality of life (HRQoL). If patients fail conservative treatment, they are treated with ankle arthrodesis (AA) or total ankle replacement (TAR). These surgical procedures have been associated with changes in general and ankle-specific HRQoL. However, there is little understanding of how these treatments affect other aspects of HRQoL, such as depression symptoms and pain, or whether these changes are meaningful to patients.This thesis is based on a cohort of prospectively recruited patients treated with AA or TAR for their ESAA. This thesis aims to 1) measure changes in aspects of HRQoL that have not been previously investigated, such as depression symptoms and pain, 2) determine if changes in HRQoL over the peri-operative period are meaningful to patients, and 3) determine the effects of potentially modifiable health-system factors, such as wait times and utilization of post-operative physiotherapy, on patients’ changes in HRQoL.Out of 190 eligible patients, 89 were included in the analysis. Participants with the worst pre-operative HRQoL experienced the largest changes. Changes in ankle-specific HRQoL were correlated with changes in pain and depression symptoms. The effects of wait times and post-operative physiotherapy on changes in HRQoL were found to be minimal. However, while most participants saw statistically significant changes in all aspects of HRQoL that were measured, not all participants found these changes to be meaningful. More research is required to understand why certain ESAA patients treated with AA and TAR do not see meaningful changes to aspects of their HRQoL.

View record

The International Consultation on Incontinence Questionnaire Short-Form (ICIQ-UI-SF) is a four-item patient-reported outcome (PRO) measure. Its intended use is screening for incontinence, assessing impact of incontinence on quality of life, and facilitating patient-clinician discussions. Evaluations of this instrument to date have relied on a simple set of analytical tools—limiting user’s confidence of the instrument’s validity and reliability. The purpose of this thesis was to conduct a comprehensive evaluation of the ICIQ-UI-SF. The analyses were conducted on 177 completed ICIQ-UI-SF instruments by men with chronic urinary incontinence waitlisted for urological surgery for treatment of their condition. This comprehensive evaluation included application of the following methods: confirmatory factor analysis (CFA), principal component analysis, measures of reliability (classical test theory (CTT) and McDonald’s coefficient), item response theory (IRT), and differential item functioning (DIF). A supplemental investigation examined previously constructed ICIQ-UI-SF severity categories. Specific goals included assessing: instrument characteristics (dimensionality, ceiling effects), reliability, performance of individual items, whether socioeconomic status influences patients’ ICIQ-UI-SF scores, and concordance with other commonly collected PROs (EQ-5D-3L, Visual Analogue Scale). Responses to all items were left skewed and ceiling effects were identified. Model fit could not be assessed through the CFA, however the factor loadings of items one and two differed significantly (p 0.0002) from item three indicating possible multidimensionality. The PCA contrastingly provided some, albeit limited evidence that the ICIQ-UI-SF is unidimensional. Reliability was low/moderate as measured by Cronbach’s alpha (0.63) and McDonald’s coefficient (0.65). The IRT revealed the third item’s reliability may be improved by collapsing response levels. The instrument does not discriminate between individuals with high incontinence burden. There was no DIF by socioeconomic status. Supplemental investigations demonstrated the ICIQ-UI-SF discriminates between surgical patients with mild/moderate incontinence versus severe/very severe—but not beyond these dichotomies. Directly comparing ICIQ-UI-SF scores of urological surgery patients with high incontinence burden is not recommended. If unchanged, the ICIQ-UI-SF can be used as a complement to other data, such as reporting aggregated surgical outcomes, or as a starting point for patient-clinician discussions when applied to a surgical population. For application to surgical triage, this analysis recommends amendments to the ICIQ-UI-SF.

View record

Introduction: Non-invasive prenatal testing (NIPT) is an accurate and safe screening test for detection of Down syndrome, trisomy 18, and trisomy 13, which can be used to reduce the need for invasive diagnostic testing in high-risk women. In November 2015, British Columba (BC) introduced publicly-funded NIPT as a contingent screening test for women at an increased risk of having a trisomic pregnancy. The objective of this research was to quantify the extent to which the introduction of publicly-funded NIPT influenced the utilization of invasive diagnostic testing in BC.Methods: We used linked population-based provincial prenatal biochemistry and cytogenetic laboratories data maintained by the BC Prenatal Genetic Screening Program and Perinatal Services BC. Our study population included all known singleton pregnancies in BC between April 1, 2011 and April 30, 2017 with a positive result from provincially-funded serum integrated prenatal screen (SIPS), integrated prenatal screen (IPS), and quad marker screen (Quad). We divided this timeframe into three periods: period 1 with no NIPT (April 1, 2011-January 31, 2013), period 2 with self-pay NIPT only (February 1, 2013-October 31, 2015), and period 3 with publicly-funded NIPT (November 1, 2015-April 30, 2017). We performed an interrupted time series analysis using log-binomial regression to evaluate the change in use of invasive diagnostic testing after the introduction of self-pay NIPT and publicly-funded NIPT. Models were adjusted for maternal age and self-reported maternal race.Results: Among the 8,649 pregnancies included in the study, the estimated rate of invasive diagnostic testing decreased by 12.7% (95% CI: 0.79 to 0.97) when self-pay NIPT was introduced. With the introduction of publicly-funded NIPT, the estimated rate of invasive diagnostic testing decreased by 47.8% (95% CI: 0.40 to 0.68). The estimated trend of decrease in the rate of invasive diagnostic testing utilization was 1.7% per month (95% CI: 0.98 to 0.99) within the second period with self-pay NIPT.Conclusions: The introduction of publicly-funded NIPT was associated with a significant decrease in the number of invasive diagnostic tests performed in BC among singleton pregnancies screened positive with provincially-funded SIPS, IPS, and Quad. Future cost-consequences studies including detailed healthcare costs would be valuable.

View record

The Fecal Incontinence Quality of Life scale (FIQL) is a patient reported outcome measure (PROM) that is used to measure the effect that fecal incontinence has on quality of life, and has previously demonstrated high reliability and validity. It measures four domains of quality of life: lifestyle, coping/behavior, depression/self-perception, and embarrassment. Despite its wide use, previous studies have not applied rigorous modern methods to evaluate the FIQL's psychometric properties at the item and test level. This thesis used a cohort of prospectively recruited patients from an elective surgical registry and applied methods from classical test theory (CTT), exploratory factor analysis (EFA), item response theory (IRT), and differential item and test functioning (DIF) to identify strengths and weaknesses in the FIQL. Specifically, this thesis aims to 1) confirm the reliability of the instrument, 2) describe the domains of quality of life measured by the instrument, 3) identify high and low quality items, and 4) determine whether one's score on the FIQL is influenced by gender or surgical procedure. Out of 317 completed questionnaires from 880 total eligible patients, 236 were included for analysis. Reliability for all four domains was high as measured by Cronbach's α. Exploratory factor analysis failed to identify the four domains the FIQL claims to measure. Individual items demonstrated high discrimination but most had low difficulty. Items 2c, 2l, 3a, and 3h failed to demonstrate good separation between response categories. Five item pairs demonstrated local item dependence, most from question 3. Only item 2g demonstrated differential item functioning, based on gender. Differential test functioning was minimal. The FIQL demonstrated a high degree of reliability, and the lifestyle domain can be used as is or with minor improvements. The FIQL can be improved by making response options consistent, distributing items from different domains evenly throughout the instrument, adding items with higher difficulty and better response separation, and removing items 2c, 2l, 3a, and 3h. Further research is needed before the FIQL can be used confidently as a stand-alone measure of fecal incontinence-related quality of life.

View record