Search

Aims This study aimed to describe practice variation in the use of total hip arthroplasty (THA) for older patients with femoral neck fracture and to determine the association between patient, surgeon, and institution factors and treatment with THA. Methods We performed a cross-sectional analysis of 49,597 patients aged 60 years and older from Ontario, Canada, who underwent hemiarthroplasty or THA for femoral neck fracture between 2002 and 2017. This population-based study used routinely collected healthcare databases linked through ICES (formerly known as the Institute for Clinical Evaluative Sciences). Multilevel logistic regression modelling was used to quantify the association between patient, surgeon, and institution-level variables and whether patients were treated with THA. Variance partition coefficient and median odds ratios were used to estimate the variation attributable to higher-level variables and the magnitude of effect of higher-level variables, respectively. Results Over the study period, 9.4% of patients (n = 4,638) were treated with THA. Patient factors associated with higher likelihood of treatment by THA included: younger age, male sex, and diagnosis with rheumatoid arthritis. Long-term care residence, use of home care services prior to hip fracture, diagnosis of dementia, higher comorbidity burden, and the most marginalized group were negatively associated with treatment by THA. Treating surgeon and institution accounted for 54.2% and 17.8% of the total variation in treatment with THA, respectively. Surgeon volume of THA procedures in the 365 days prior to surgery was the strongest higher-level predictor of treatment with THA. Specific treating surgeons and institutions still accounted for significant proportions of the variability in treatment with THA (40.3% and 19.5% of total observed variation, respectively) after controlling for available patient, surgeon, and institution-level variables. Conclusion The strongest predictors for treatment of patients with femoral neck fracture with THA were patient age, treating surgeon, and treating institution. This practice variation highlights differential access to care for patients. Cite this article: Bone Joint J 2023;105-B(2):180–189.

Background and ObjectivesThe multiple sclerosis (MS) population's survival from breast cancer and colorectal cancer is compromised. Cancer screening and timely diagnoses affect cancer survival and have not been studied in the MS cancer population. We investigated whether the diagnostic route, cancer stage, or diagnostic interval differed in patients with cancer with and without MS.MethodsWe conducted a matched population-based cross-sectional study of breast cancers (2007–2015) and colorectal cancers (2009–2012) in patients with MS from Ontario, Canada, using administrative data. Exclusion criteria included second or concurrent primary cancers, no health care coverage, and, for the patients without MS, those with any demyelinating disease. We based 1:4 matching of MS to non-MS on birth year, sex (colorectal only), postal code, and cancer diagnosis year (breast only). Cancer outcomes were diagnostic route (screen-detected vs symptomatic), stage (stage I vs all others), and diagnostic interval (time from first presentation to diagnosis). Multivariable regression analyses controlled for age, sex (colorectal only), diagnosis year, income quintile, urban/rural residence, and comorbidity.ResultsWe included 351 patients with MS and breast cancer, 1,404 matched patients with breast cancer without MS, 54 patients with MS and colorectal cancer, and 216 matched patients with colorectal cancer without MS. MS was associated with fewer screen-detected cancers in breast (odds ratio [OR] 0.68 [95% CI 0.52, 0.88]) and possibly colorectal (0.52 [0.21, 1.28]) cancer. MS was not associated with differences in breast cancer stage at diagnosis (stage I cancer, OR 0.81 [0.64, 1.04]). MS was associated with greater odds of stage I colorectal cancer (OR 2.11 [1.03, 4.30]). The median length of the diagnostic interval did not vary between people with and without MS in either the breast or colorectal cancer cohorts. Controlling for disability status attenuated some findings.DiscussionBreast cancers were less likely to be detected through screening and colorectal cancer more likely to be detected at early stage in people with MS than without MS. MS-related disability may prevent people from getting mammograms and colonoscopies. Understanding the pathways to earlier detection in both cancers is critical to developing and planning interventions to ameliorate outcomes for people with MS and cancer.

Background Few reports have evaluated prognostic modelling studies of tools used for surgical decision-making. This systematic review aimed to describe and critically appraise studies that have developed or validated multivariable prognostic models for post-operative liver decompensation following partial hepatectomy. Methods This study was designed using the CHARMS checklist. Following a comprehensive literature search, two reviewers independently screened candidate references for inclusion and abstracted relevant study details. Qualitative assessment was performed using the PROBAST tool. Results We identified 36 prognostic modelling studies; 25 focused on development only, 3 developed and validated models, and 8 validated pre-existing models. None compared routine use of a prognostic model against standard clinical practice. Most studies used single-institution, retrospective cohort designs, conducted in Eastern populations. In total, 15 different outcome definitions for post-operative liver decompensation events were used. Statistical concerns surrounding model overfitting, performance assessment, and internal validation led to high risk of bias for all studies. Conclusions Current prognostic models for post-operative liver decompensation following partial hepatectomy may not be valid for routine clinical use due to design and methodologic concerns. Landmark resources and reporting guidelines such as the TRIPOD statement may assist researchers, and additionally, model impact assessment studies represent opportunities for future research.

Appropriate patient selection for liver resection in hepatocellular carcinoma (HCC) is critical to mitigation of major liver-related postoperative complications. Currently, no standard prognostic tool exists to predict the risk of postoperative liver decompensation events (POLDEs) after partial hepatectomy for patients with cirrhosis and HCC. This study aimed to identify independent preoperative predictors of POLDEs for future development of prognostic tools to improve surgical decision-making. This population-based, retrospective cohort study investigated patients with cirrhosis and incident HCC between 2007 and 2017, identified using administrative health data from Ontario, Canada. The occurrence of a POLDE or death within 2 years after surgery was described. Multivariable Cox regression identified independent predictors of POLDE-free survival, as well as cause-specific hazards for POLDEs and death. Among 611 patients with cirrhosis and HCC who underwent liver resection, 160 (26.2%) experienced at least one POLDE, and 189 (30.9%) died within 2 years after surgery. Diabetes, cirrhosis etiology, major liver resection, and previous non-malignant decompensation were independent predictors of POLDE-free survival. Except for extent of resection, the same risk factors were associated with POLDEs in the cause-specific analysis. In contrast, only age and history of previous non-malignant decompensation were independent predictors of mortality. Among patients with cirrhosis undergoing resection for HCC, patient and disease-related factors are associated with POLDEs and POLDE-free survival. These factors can be used both to inform clinical practice and to advance the development of preoperative prognostic tools, which may lead to improved outcomes for this population.

Background and ObjectivesWe tested the hypothesis that overall and cancer-specific survival after a colorectal cancer diagnosis is lower in persons with multiple sclerosis (MS) than in those without MS using a retrospective matched cohort design.MethodsUsing population-based administrative data in Manitoba and Ontario, we identified persons with MS from a validated case definition and linked these cohorts to cancer registries to identify those with colorectal cancer. We selected persons with colorectal cancer and without MS, matching 4:1 on birth year, sex, cancer diagnosis year, and region. We used Cox proportional hazards regression to compare all-cause survival between cohorts, adjusting for age at cancer diagnosis, cancer diagnosis year, income, region, and Elixhauser comorbidity score. We compared cancer-specific survival between cohorts using a cause-specific hazards model. We pooled findings across provinces using random-effects meta-analysis. Complementary analyses using a subcohort from Ontario, adjusted for cancer stage and disability status, as measured from the use of home care or long-term care services.ResultsWe included 338 MS cases and 1,352 controls with colorectal cancer. The mean (SD) age at cancer diagnosis was 64.7 (11.1) years. After adjustment, MS was associated with an increased hazard for all-cause death that was highest 6 months after diagnosis (hazard ratio [HR] 1.45, 95% confidence interval [CI] 1.19–1.76) and then declined over time (HR [95% CI] at 1 year 1.34 [1.09–1.63], 2 years 1.24 [0.99–1.56], 5 years 1.10 [0.80–1.50]). MS was associated with increased cancer-specific death at 6 months after diagnosis only (HR 1.29, 95% CI 1.04–1.61). After adjustment for cancer stage, MS was associated with an increased hazard of death due to any cause (1.60, 95% CI 1.16–2.21) and with cancer-specific death (HR 1.47, 95% CI 1.02–2.12). The association of MS and all-cause death was partially attenuated after adjustment for disability status (HR 1.37, 95% CI 0.97–1.92), as was the association with cancer-specific death (HR 1.34, 95% CI 0.91–1.97).DiscussionOverall and cancer-specific survival was lower in persons with than without MS in the early period after colorectal cancer diagnosis. Further study is warranted to determine what factors underlie these worse outcomes.

Mortality secondary to cirrhosis in North America is increasing. We describe the incidence of cirrhosis stratified by birth cohort and cirrhosis etiology and project disease burden to 2040.This is a retrospective cohort study in Ontario, Canada, using population-based administrative health care data. Individuals with incident cirrhosis (2000-2017) were identified, and etiology was defined as HCV, HBV, NAFLD, alcohol-associated liver disease (ALD), or autoimmune liver disease/other using validated case definitions. Annual age/sex-adjusted cirrhosis incidence rate per 100,000 person-years was calculated with incidence projection to 2040 using age-period-cohort modeling along with average annual percent change (AAPC) in cirrhosis incidence stratified by birth cohort and etiology. In total, 159,549 incident cases of cirrhosis were identified. Incidence increased by 26% with an AAPC of 2%/year (95% CI, 1.6-2.4; P 0.001). The largest increases were for HCV (AAPC, 4.1%/year; 95% CI, 2.6-5.7; P 0.001) and NAFLD (AAPC, 3.3%/year; 95% CI, 2.6-4.1%; P 0.001). ALD and HCV cirrhosis in those born1980 increased by 11.6%/year (95% CI, 9.3-13.9; P 0.001) and 9.5%/year (95% CI, 6.2-13.0; P 0.001), respectively. However, by 2040, cirrhosis incidence is projected to continue to increase, driven mostly by NAFLD, especially in postmenopausal women, and ALD in individuals born1980.Cirrhosis incidence will continue to increase over the next two decades secondary to NAFLD with a worrisome rapid rise in ALD cirrhosis among young adults. Public education, policy, and intervention targeting NAFLD risk factors and alcohol use in young adults are urgently needed.

ObjectiveTo test the hypotheses that overall survival and cancer-specific survival after breast cancer diagnosis would be lower in persons with multiple sclerosis (MS) as compared to persons without MS using a retrospective matched cohort design.MethodsWe applied a validated case definition to population-based administrative data in Manitoba and Ontario, Canada, to identify women with MS. We linked the MS cohorts to cancer registries to identify women with breast cancer. Then we selected 4 breast cancer controls without MS matched on birth year, cancer diagnosis year, and region. We compared all-cause survival between cohorts using Cox proportional hazards regression adjusting for age at cancer diagnosis, cancer diagnosis period, income quintile, region, and Elixhauser comorbidity score. We compared cancer-specific survival between cohorts using a multivariable cause-specific hazards model. We pooled findings between provinces using meta-analysis.ResultsWe included 779 patients with MS and 3,116 controls with breast cancer. Most patients with stage data (1,976/2,822 [70.0%]) were diagnosed with stage I or II breast cancer and the mean (SD) age at diagnosis was 57.8 (10.7) years. After adjustment for covariates, MS was associated with a 28% increased hazard for all-cause mortality (hazard ratio [HR] 1.28; 95% confidence interval [CI] 1.08–1.53), but was not associated with altered cancer-specific survival (HR 0.98; 95% CI 0.65–1.46).ConclusionWomen with MS have lower all-cause survival after breast cancer diagnosis than women without MS. Future studies should confirm these findings in other populations and identify MS-specific factors associated with worse prognosis.

This study aimed to identify colorectal cancer (CRC) diagnostic pathways and describe patients in those pathway groups.This was a cross-sectional study of CRC patients in Ontario, Canada, diagnosed 2009-2012 that used linked administrative data at ICES. We used cluster analysis on 11 pathway variables characterising patient presentation, symptoms, procedures and referrals. We assessed associations between patient- and disease-related characteristics and diagnostic pathway group. We further characterised the pathways by diagnostic interval and number of related physician visits.Six diagnostic pathways were identified, with three adhering to provincial diagnostic guidelines: screening (N = 4494), colonoscopy (N = 10,066) and imaging plus colonoscopy (N = 3427). Non-adherent pathways were imaging alone (N = 2238), imaging and emergency presentation (N = 2849) and no pre-diagnostic workup (N = 887). Patients in adherent pathways were younger, had fewer comorbidities, lived in less deprived areas and had earlier stage disease. The median diagnostic interval length varied across pathways from 12 to 126 days, correlating with the number of CRC-related visits.This study demonstrated substantial variations in real-world CRC diagnostic pathways and 25% were diagnosed through non-adherent pathways. Those patients were older, had more comorbid disease and had higher stage cancer. Further research needs to identify and describe the reasons for divergent diagnostic processes.

A prolonged time from first presentation to cancer diagnosis has been associated with worse disease-related outcomes. This study evaluated potential determinants of a long diagnostic interval among symptomatic breast cancer patients. This was a population-based, cross-sectional study of symptomatic breast cancer patients diagnosed in Ontario, Canada from 2007 to 2015 using administrative health data. The diagnostic interval was defined as the time from the earliest breast cancer-related healthcare encounter before diagnosis to the diagnosis date. Potential determinants of the diagnostic interval included patient, disease and usual healthcare utilization characteristics. We used multivariable quantile regression to evaluate their relationship with the diagnostic interval. We also examined differences in diagnostic interval by the frequency of encounters within the interval. Among 45,967 symptomatic breast cancer patients, the median diagnostic interval was 41 days (interquartile range 20–92). Longer diagnostic intervals were observed in younger patients, patients with higher burden of comorbid disease, recent immigrants to Canada, and patients with higher healthcare utilization prior to their diagnostic interval. Shorter intervals were observed in patients residing in long-term care facilities, patients with late stage disease, and patients who initially presented in an emergency department. Longer diagnostic intervals were characterized by an increased number of physician visits and breast procedures. The identification of groups at risk of longer diagnostic intervals provides direction for future research aimed at better understanding and improving breast cancer diagnostic pathways. Ensuring that all women receive a timely breast cancer diagnosis could improve breast cancer outcomes.

ObjectiveTo determine whether cancer risk differs in people with and without multiple sclerosis (MS), we compared incidence rates and cancer-specific mortality rates in MS and matched cohorts using population-based data sources.MethodsWe conducted a retrospective matched cohort study using population-based administrative data from Manitoba and Ontario, Canada. We applied a validated case definition to identify MS cases, then selected 5 controls without MS matched on birth year, sex, and region. We linked these cohorts to cancer registries, and estimated incidence of breast, colorectal, and 13 other cancers. For breast and colorectal cancers, we constructed Cox models adjusting for age at the index date, area-level socioeconomic status, region, birth cohort year, and comorbidity. We pooled findings across provinces using meta-analysis.ResultsWe included 53,983 MS cases and 269,915 controls. Multivariable analyses showed no difference in breast cancer risk (pooled hazard ratio [HR] 0.92 [95% confidence interval (CI) 0.78–1.09]) or colorectal cancer risk (pooled HR 0.83 [95% CI 0.64–1.07]) between the cohorts. Mortality rates for breast and colorectal did not differ between cohorts. Bladder cancer incidence and mortality rates were higher among the MS cohort. Although the incidence of prostate, uterine, and CNS cancers differed between the MS and matched cohorts, mortality rates did not.ConclusionThe incidence of breast and colorectal cancers does not differ between persons with and without MS; however, the incidence of bladder cancer is increased. Reported differences in the incidence of some cancers in the MS population may reflect ascertainment differences rather than true differences.

OBJECTIVE: Breast cancer screening aims to identify cancers in early stages when prognosis is better and treatments less invasive. We describe inter- and intra-provincial variation in the percentage of screen-detected cases under publicly funded healthcare systems and factors related to having screen- vs non-screen-detected breast cancer across five Canadian provinces. METHODS: Women aged 40+ diagnosed with incident breast cancer from 2007 to 2012 in five Canadian provinces were identified from their respective provincial cancer registries. Standardized provincial datasets were created linking screening, health administrative, and claims data. Province-specific logistic regression models were used to evaluate the association of demographic and healthcare utilization factors in each province with the odds of screen-detected cancer. RESULTS: There was significant inter- and intra-provincial variation by age. Screen detection ranged from 42% to 52% in ages 50–69 but women aged 50–59 had approximately 4–8% lower screen detection than those aged 60–69 in all provinces. Screening associations with income quintile and rurality varied across provinces. Those least likely to be screen-detected within a province were consistently in the lowest income quintile; OR ranged from 0.62–0.89 relative to highest income quintile/urban patients aged 50–69. Lack of visits to primary care 30 months prior to diagnosis was also consistently associated with lower odds of screen detection (OR range, 0.37–0.76). CONCLUSION: Breast cancer screen detection rates in the Canadian provinces examined are relatively high. Associations with income-rurality indicate a need for greater attention and/or targeted outreach to specific communities and/or provincial regions to improve access to breast cancer screening services intra-provincially.

The cancer diagnostic process can be protracted, and it is a time of great anxiety for patients. The objective of this study was to examine inter- and intra-provincial variation in diagnostic intervals and explore factors related to the variation. This was a multi-province retrospective cohort study using linked administrative health databases. All females with a diagnosis of histologically confirmed invasive breast cancer in British Columbia (2007–2010), Manitoba (2007–2011), Ontario (2007–2010), Nova Scotia (2007–2012), and Alberta (2004–2010) were included. The start of the diagnostic interval was determined using algorithms specific to whether the patient’s cancer was detected through screening. We used multivariable quantile regression analyses to assess the association between demographic, clinical and healthcare utilization factors with the diagnostic interval outcome. We found significant inter- and intra-provincial variation in the breast cancer diagnostic interval and by screen-detection status; patients who presented symptomatically had longer intervals than screen-detected patients. Interprovincial diagnostic interval variation was 17 and 16 days for screen- and symptom-detected patients, respectively, at the median, and 14 and 41 days, respectively, at the 90th percentile. There was an association of longer diagnostic intervals with increasing comorbid disease in all provinces in non-screen-detected patients but not screen-detected. Longer intervals were observed across most provinces in screen-detected patients living in rural areas. Having a regular primary care provider was not associated with a shorter diagnostic interval. Our results highlight important findings regarding the length of the breast cancer diagnostic interval, its variation within and across provinces, and its association with comorbid disease and rurality. We conclude that diagnostic processes can be context specific, and more attention should be paid to developing tailored processes so that equitable access to a timely diagnosis can be achieved.

ObjectivesRegional variation in cancer survival is an important health system performance measurement. We evaluated if regional variation in colon cancer survival may be driven by differences in the patient population, their health and healthcare utilisation, and/or cancer care delivery.DesignPopulation-based retrospective cohort study using routinely collected linked health administrative data.SettingOntario, Canada.ParticipantsPatients with colon cancer diagnosed between 1 January 2009 and 31 December 2012.OutcomeCancer-specific survival was compared across the province’s 14 health regions. Using accelerated failure time models, we assessed whether regional survival variations were mediated through differences in case mix, including age, sex, comorbidities, stage at diagnosis and colon subsite, potential marginalisation and/or prediagnosis healthcare.ResultsThe study population included 16 895 patients with colon cancer. There was statistically significant regional variation in cancer-specific survival. Three regions had cancer-specific survival that was between 30% (95% CI 1.03 to 1.65) and 39% (95% CI 1.13 to 1.71) longer and one region had cancer-specific survival that was 26% shorter (95% CI 0.58 to 0.93) than the reference region. For three of these regions, case mix explained between 26% and 56% of the survival variation. Further adjustment for rurality explained 22% of the remaining survival variation in one region. Adjustment for continuity of primary care and the diagnostic interval length explained 10% and 11% of the remaining survival variation in two other regions. Socioeconomic marginalisation, recent immigration and colonoscopy history did not explain colon cancer survival variation.ConclusionsCase mix accounted for much of the regional variation in colon cancer survival, indicating that efforts to monitor the quality of cancer care through survival metrics should consider case mix when reporting regional survival differences. Future work should repeat this approach in other settings and other cancer sites considering a broad range of potential mediators.

Appropriate patient selection for liver resection in hepatocellular carcinoma (HCC) is critical to mitigation of major liver-related postoperative complications. Currently, no standard prognostic tool exists to predict the risk of postoperative liver decompensation events (POLDEs) after partial hepatectomy for patients with cirrhosis and HCC. This study aimed to identify independent preoperative predictors of POLDEs for future development of prognostic tools to improve surgical decision-making.This population-based, retrospective cohort study investigated patients with cirrhosis and incident HCC between 2007 and 2017, identified using administrative health data from Ontario, Canada. The occurrence of a POLDE or death within 2 years after surgery was described. Multivariable Cox regression identified independent predictors of POLDE-free survival, as well as cause-specific hazards for POLDEs and death.Among 611 patients with cirrhosis and HCC who underwent liver resection, 160 (26.2%) experienced at least one POLDE, and 189 (30.9%) died within 2 years after surgery. Diabetes, cirrhosis etiology, major liver resection, and previous non-malignant decompensation were independent predictors of POLDE-free survival. Except for extent of resection, the same risk factors were associated with POLDEs in the cause-specific analysis. In contrast, only age and history of previous non-malignant decompensation were independent predictors of mortality.Among patients with cirrhosis undergoing resection for HCC, patient and disease-related factors are associated with POLDEs and POLDE-free survival. These factors can be used both to inform clinical practice and to advance the development of preoperative prognostic tools, which may lead to improved outcomes for this population.

Chemotherapy has improved outcomes in early-stage breast cancer, but treatment practices vary, and use of acute care is common. We conducted a pan-Canadian study to describe treatment differences and the incidence of emergency department visits (edvs), edvs leading to hospitalization (edvhs), and direct hospitalizations (hs) during adjuvant chemotherapy. The cohort consisted of women diagnosed with early-stage breast cancer (stages i&ndash, iii) during 2007&ndash, 2012 in British Columbia, Manitoba, Ontario, or Nova Scotia who underwent curative surgery. Parallel provincial analyses were undertaken using linked clinical, registry, and administrative databases. The incidences of edvs, edvhs, and hs in the 6 months after treatment initiation were examined for patients treated with adjuvant chemotherapy. The cohort consisted of 50,224 patients. The proportion of patients who received chemotherapy varied by province, with Ontario having the highest proportion (46.4%), and Nova Scotia, the lowest proportion (38.0%). Age, stage, receptor status, comorbidities, and geographic location were associated with receipt of chemotherapy in all provinces. Ontario had the highest proportion of patients experiencing an edv (36.1%), but the lowest proportion experiencing h (6.4%). Conversely, British Columbia had the lowest proportion of patients experiencing an edv (16.0%), but the highest proportion experiencing h (26.7%). The proportion of patients having an edvh was similar across provinces (13.9%&ndash, 16.8%). Geographic location was associated with edvs, edvhs, and hs in all provinces. Intra- and inter-provincial differences in the use of chemotherapy and acute care were observed. Understanding variations in care can help to identify gaps and opportunities for improvement and shared learnings.

Background: Readmission in patients with cirrhosis is common. We aimed to determine the association between early hospital readmission and survival in the general population of patients with cirrhosis. Methods: This retrospective cohort study used routinely collected health care data from Ontario. We identified adults with cirrhosis using a validated case definition, and included those with at least one hospital admission between 1992 and 2016 resulting in discharge. Patients were classified into two groups based on timing of readmission after index admission: 1) ≤90 days, or 2) >90 days or no readmission. We described overall survival (OS) 90 days after the index hospitalization by readmission status using Kaplan–Meier curves and the log-rank test. The association between readmission and OS was evaluated using a multivariate Cox proportional hazards regression model. Results: Our study included 115,081 patients. The median OS was shorter in patients readmitted in ≤90 days (4.1 years, IQR 0.9, 13.1) compared with those readmitted in >90 days or not readmitted during the study period (9.6 years, IQR 3.2, 21.9, p

Recent data show that the prevalence of chronic liver disease and cirrhosis is increasing in adolescents and young adults in the USA. We aimed to describe the epidemiology of cirrhosis using an age-period-cohort approach to define birth-cohort effects on the incidence of cirrhosis in Ontario, Canada.We did a retrospective population-based cohort study in Ontario, Canada, using linked administrative health data from the databases of ICES, formerly the Institute for Clinical Evaluative Sciences. Patients aged at least 18 years with cirrhosis were identified by use of a validated case definition (defined as at least one inpatient or outpatient visit with a diagnosis of cirrhosis or oesophageal varices without bleeding). We calculated annual standardised incidence and prevalence in the general population. We used an age-period-cohort approach to assess the independent association between birth cohort and incidence of cirrhosis in men and women.Between Jan 1, 1997, and Dec 31, 2016, 165 979 individuals with cirrhosis were identified. The age-standardised incidence increased over the study (from 70·6 per 100 000 person-years in 1997 to 89·6 per 100 000 person-years in 2016) as did the prevalence (from 0·42% in 1997 to 0·84% in 2016). Using age-period-cohort modelling and the median birth year as the reference, the incidence of cirrhosis was higher in participants born in 1980 (incidence rate ratio 1·55, 95% CI 1·50-1·59, p0·0001); and in participants born in 1990 (2·16, 95% CI 2·06-2·27, p0·0001) compared with a person of the same age born in 1951. The increase in incidence of cirrhosis was greater in women than in men (eg, women born in 1990: 2·60, 95% CI 2·41-2·79; men born in 1990: 1·98, 1·85-2·12).The incidence of cirrhosis has increased over the past two decades, and more so in younger birth cohorts and in women. Future studies to define the cause and natural history of cirrhosis in these groups are essential to develop strategies that could reverse these trends for future generations.Southeastern Ontario Academic Medical Association New Clinician Scientist Award; American Association for the Study of Liver Disease (AASLD) Foundation Clinical, Translational and Outcomes Research Award in Liver Disease (JAF).

ObjectiveThere is substantial variation in colonoscopy use and evidence of long wait times for the procedure. Understanding the role of system-level resources in colonoscopy utilisation may point to a potential intervention target to improve colonoscopy use. This study characterises colonoscopy resource availability in Ontario, Canada and evaluates its relationship with colonoscopy utilisation.DesignWe conducted a population-based study using administrative health data to describe regional variation in colonoscopy availability for Ontario residents (age 18–99) in 2013. We identified 43 colonoscopy networks in the province in which we described variations across three colonoscopy availability measures: colonoscopist density, private clinic access and distance to colonoscopy. We evaluated associations between colonoscopy resource availability and colonoscopy utilisation rates using Pearson correlation and log binomial regression, adjusting for age and sex.ResultsThere were 9.4 full-time equivalent colonoscopists per 100 000 Ontario residents (range across 43 networks 0.0 to 21.8); 29.5% of colonoscopies performed in the province were done in private clinics (range 1.2%–55.9%). The median distance to colonoscopy was 3.7 km, with 5.9% travelling at least 50 km. Lower colonoscopist density was correlated with lower colonoscopy utilisation rates (r=0.53, pConclusionThe substantial variations in colonoscopy resource availability and the relationship demonstrated between colonoscopy resource availability and use provides impetus for health service planners and decision-makers to address these potential inequalities in access in order to support the use of this medically necessary procedure.

Appendectomy is a common emergency procedure. The risks have been reported in previous studies but often are limited to inpatient complications. The purpose of this study was to describe inpatient and outpatient rates of complications associated with appendectomy in a contemporary population-based cohort and explore factors associated with these complications.We conducted a retrospective study using linked data for Ontario within Institute for Clinical Evaluative Sciences databases. Patients who underwent emergent appendectomy between 2009 and 2014 were included. The primary outcome was a complication (death, readmission, emergency department visit, lengthy [7 d] hospital stay, percutaneous abscess drainage, reoperation or 1 of the predefined complication codes) occurring within 30 days of surgery. We used modified Poisson regression to identify predictors of complications.A total of 50 369 patients underwent emergent appendectomy over the study period, of whom 16 953 (33.7%) had a perforated appendix. Overall, 14 451 patients (28.7%) (8428 [25.2%] in the nonperforated group and 6023 [35.5%] in the perforated group) had at least 1 complication. The most common complication was an emergency department visit (7942 patients [15.8%]), followed by surgical site infection (4792 [9.5%]). Increasing age, female sex, rural residence, perforation status, daytime surgery and open surgical technique were associated with increased risk of complications.We found a higher rate of complications after appendectomy than previously reported. The most common complication was presentation to the emergency department. Our definition of complications is more inclusive than in previous studies and provides a deeper understanding of complications after appendectomy.L'appendicectomie est une intervention urgente qui est courante. Les risques qui y sont associés ont déjà fait l'objet d'études, mais se limitent souvent aux complications perhospitalières. Cette étude avait pour but de décrire les taux de complications chez les patients durant et après leur hospitalisation pour appendicectomie dans une cohorte contemporaine basée dans la population et d'explorer les facteurs associés à ces complications.Nous avons procédé à une étude rétrospective à partir des données reliées des bases de données de l'Institute for Clinical Evaluative Sciences pour l'Ontario. Les patients soumis à une appendicectomie urgente entre 2009 et 2014 ont été inclus. Le paramètre principal était la survenue d'une complication (décès, réadmission, consultation aux urgences, séjour hospitalier prolongé [7 j], drainage d'abcès percutané, réintervention ou l'un des codes de complications prédéfinis) dans les 30 jours suivant la chirurgie. Nous avons utilisé la régression de Poisson pour identifier les facteurs prédicteurs de complications.En tout, 50 369 patients ont subi une appendicectomie urgente au cours de la période de l'étude, dont 16 953 (33,7 %) présentaient un appendice perforé. Globalement, 14 451 patients (28,7 %) (8 428 [25,2 %] dans le groupe ayant l'appendice non perforé et 6 023 [35,5 %] dans le groupe ayant l'appendice perforé) ont eu au moins une complication. La complication la plus fréquente a été une consultation aux urgences (7942 patients [15,8 %]), suivie de l'infection du site opératoire (4792 [9,5 %]). L'âge avancé, le fait d'être de sexe féminin, de vivre en milieu rural, un appendice perforé, la chirurgie effectuée durant le jour et la technique chirurgicale ouverte ont tous été associés à un risque accru de complications.Nous avons observé un taux de complications post-appendicectomie plus élevé comparativement aux rapports précédents. La complication la plus fréquente était la consultation aux urgences. Notre définition du terme complications est plus inclusive que celle des précédentes études et permet une compréhension plus approfondie des complications post-appendicectomie.

Introduction Colorectal cancer (CRC) is the third most common cancer worldwide (Ferlay et al., 2015, International Journal of Cancer, 136, E359), and delayed diagnosis is associated with mortality (Torring et al., 2011, British Journal of Cancer, 104, 934; Torring et al., 2012, Journal of Clinical Epidemiology, 65, 669). The purpose of this review was to determine the factors associated with time to diagnosis in symptomatic CRC using scoping review methods. Methods We performed database and citation searches to identify studies which examine the length of any interval from symptom presentation to diagnosis. Study selection was conducted by two independent reviewers. Factors contributing to time to diagnosis were extracted from selected articles and mapped onto a conceptual framework consisting of four levels: patient and disease factors, provider factors, organisation/setting factors and sectors of influence. Results From the 31 studies included in this review, we identified 138 unique factors, 17 of which were investigated by at least three studies and 11 of which had consistent results. Patient and disease factors were most commonly studied. Patient perception that their symptoms were benign, a non-urgent referral, female sex and rectal tumour location were each associated with a longer time to diagnosis. Conclusion Thus far, the literature has focused on patient or disease-related factors, while other levels of influence have been relatively understudied.

IntroductionDespite recommendations, most studies examining health inequalities fail to report both absolute and relative summary measures. We examine colorectal cancer (CRC) survival for patients with and without severe psychiatric illness (SPI) to demonstrate the use and importance of relative and absolute effects. Objectives and ApproachWe conducted a retrospective cohort study of CRC patients diagnosed between 01/04/2007 and 31/12/2012, using linked administrative databases. SPI was defined as diagnoses of major depression, bipolar disorder, schizophrenia, and other psychotic illnesses six months to five years preceding cancer diagnosis and categorized as inpatient, outpatient or none. Associations between SPI history and risk of death were examined using Cox Proportional Hazards regression to obtain hazard ratios and Aalen’s semi-parametric additive hazards regression to obtain absolute differences. Both models controlled for age, sex, primary tumour location, and rurality. ResultsThe final cohort included 24,507 CRC patients, 482 patients had an outpatient SPI history and 258 patients had an inpatient SPI history. 58.1% of patients with inpatient SPI history died, and 47.1% of patients with outpatient SPI history died. Patients with an outpatient SPI history had a 40% (HR 1.40, 95% CI: 1.22-1.59) increased risk of death and patients with an inpatient SPI history had a 91% increased risk of death (HR 1.91, 95% CI: 1.63-2.25), relative to no history of a mental illness. An outpatient SPI history was associated with an additional 33 deaths per 1000 person years, and an inpatient SPI was associated with an additional 82 deaths per 1000 person years after controlling for confounders. Conclusion / ImplicationsWe demonstrated that reporting of both relative and absolute effects is possible and calculating risk difference is relatively simple using Aalen models. We encourage future studies examining inequalities with time-to-event data to use this method and report both relative and absolute effect measures.

BACKGROUND: Randomized trials have shown that transjugular intrahepatic portosystemic shunt (TIPS) improves control of ascites compared with serial large volume paracentesis (LVP) in patients with refractory ascites. However, the effect of TIPS on liver transplant-free (LTF) survival is controversial. Our objective was to compare TIPS versus serial LVP on LTF survival in the general population of patients with refractory ascites. METHODS: This is a retrospective, population-based cohort study using linked administrative health data from Ontario, Canada. Adult patients identified with refractory ascites from January 1, 2008 to December 31, 2016 were included and followed until December 31, 2017. A propensity score was used to match patients treated with serial LVP to those who received TIPS in a 2:1 ratio. LTF survival was evaluated using Kaplan–Meier analysis and Cox proportional hazards regression with TIPS treated as a time-varying exposure. RESULTS: Overall, 4,935 patients with refractory ascites were identified and 488 patients were matched (325 serial LVP, 163 TIPS). The mean age was 58 years, 70% were male, 50% had viral hepatitis, the median model for end-stage liver disease (MELD) score was 12, 13% received liver transplant and the 1-year LTF survival was 72%. After TIPS, 80 patients (49%) had no further requirement for LVP by 6 months and 61 patients (37%) never required a repeat paracentesis. In survival analysis, there was marginally worse LTF survival in patients receiving TIPS (TIPS HR 1.29, 95% CI 1.00–1.67; p = .052). CONCLUSION: In this population-based study of patients with refractory ascites, TIPS was associated with improved control of ascites but not improved LTF survival.

A prolonged time from first presentation to cancer diagnosis has been associated with worse disease-related outcomes. This study evaluated potential determinants of a long diagnostic interval among symptomatic breast cancer patients.This was a population-based, cross-sectional study of symptomatic breast cancer patients diagnosed in Ontario, Canada from 2007 to 2015 using administrative health data. The diagnostic interval was defined as the time from the earliest breast cancer-related healthcare encounter before diagnosis to the diagnosis date. Potential determinants of the diagnostic interval included patient, disease and usual healthcare utilization characteristics. We used multivariable quantile regression to evaluate their relationship with the diagnostic interval. We also examined differences in diagnostic interval by the frequency of encounters within the interval.Among 45,967 symptomatic breast cancer patients, the median diagnostic interval was 41 days (interquartile range 20-92). Longer diagnostic intervals were observed in younger patients, patients with higher burden of comorbid disease, recent immigrants to Canada, and patients with higher healthcare utilization prior to their diagnostic interval. Shorter intervals were observed in patients residing in long-term care facilities, patients with late stage disease, and patients who initially presented in an emergency department. Longer diagnostic intervals were characterized by an increased number of physician visits and breast procedures.The identification of groups at risk of longer diagnostic intervals provides direction for future research aimed at better understanding and improving breast cancer diagnostic pathways. Ensuring that all women receive a timely breast cancer diagnosis could improve breast cancer outcomes.

Cancer care is complex and exists within the broader healthcare system. The CanIMPACT team sought to enhance primary cancer care capacity and improve integration between primary and cancer specialist care, focusing on breast cancer. In Canada, all medically-necessary healthcare is publicly funded but overseen at the provincial/territorial level. The CanIMPACT Administrative Health Data Group’s (AHDG) role was to describe inter-sectoral care across five Canadian provinces: British Columbia, Alberta, Manitoba, Ontario and Nova Scotia. This paper describes the process used and challenges faced in creating four parallel administrative health datasets. We present the content of those datasets and population characteristics. We provide guidance for future research based on ‘lessons learned’. The AHDG conducted population-based comparisons of care for breast cancer patients diagnosed from 2007-2011. We created parallel provincial datasets using knowledge from data inventories, our previous work, and ongoing bi-weekly conference calls. Common dataset creation plans (DCPs) ensured data comparability and documentation of data differences. In general, the process had to be flexible and iterative as our understanding of the data and needs of the broader team evolved. Inter-sectoral data inconsistencies that we had to address occurred due to differences in: 1) healthcare systems, 2) data sources, 3) data elements and 4) variable definitions. Our parallel provincial datasets describe the breast cancer diagnostic, treatment and survivorship phases and address ten research objectives. Breast cancer patient demographics reflect inter-provincial general population differences. Across provinces, disease characteristics are similar but underlying health status and use of healthcare services differ. Describing healthcare across Canadian jurisdictions assesses whether our provincial healthcare systems are delivering similar high quality, timely, accessible care to all of our citizens. We have provided a description of our experience in trying to achieve this goal and include a list of ‘lessons learned’ and a study process checklist for future use. Key FindingsThe conduct of inter-sectoral research using linked administrative health data requires a committedteam that is adequately resourced and has a set of clear, feasible objectives at the start. Guiding principles include: maximization of sectoral participation by including single-jurisdictionexpertise and making the most inclusive data decisions; use of living documents that track all datadecisions and careful consideration about data quality and availability differences. Inter-sectoral research requires a good understanding of the local healthcare system and othercontextual issues for appropriate interpretation of observed differences.

Breast assessment sites (bass) were developed to provide expedited and coordinated care for patients being evaluated for breast cancer (bca) in Ontario. We compared the diagnostic and treatment intervals for patients diagnosed at a bas and for those diagnosed through a usual care (uc) route. This population-based, cross-sectional study of patients diagnosed with bca in Ontario during 2007&ndash, 2015 used linked administrative data. &ldquo, Diagnostic interval&rdquo, was the time from the earliest cancer-related health care encounter before diagnosis to diagnosis, &ldquo, treatment interval&rdquo, was the time from diagnosis to treatment. Diagnosis at a bas was determined from the patient&rsquo, s biopsy and mammography institutions. Interval lengths for the bas and uc groups were compared using multivariable quantile regression, stratified by detection method. The diagnostic interval was shorter for patients who were bas-diagnosed than for those who were uc-diagnosed, with adjusted median differences of &minus, 4.0 days [95% confidence interval (ci): &minus, 3.2 days to &minus, 4.9 days] for symptomatic patients and &minus, 5.4 days (95% ci: &minus, 4.7 days to &minus, 6.1 days) for screen-detected patients. That association was modified by stage at diagnosis, with larger differences in patients with early-stage cancers. In contrast, the treatment interval was longer in patients who were bas-diagnosed than in those who were uc-diagnosed, with adjusted median differences of 4.2 days (95% ci: 3.8 days to 4.7 days) for symptomatic patients and 4.2 days (95% ci: 3.7 days to 4.8 days) for screen-detected patients. Diagnosis of bca through a bas was associated with a shorter diagnostic interval, but a longer treatment interval. Although efficiencies in the diagnostic interval might help to reduce distress experienced by patients, the longer treatment intervals for patients who are bas-diagnosed remain a cause for concern.

Background Advanced cancer stage at diagnosis may explain high cancer mortality among patients with a severe psychiatric illness (SPI). Studies to date investigating advanced stage cancer at diagnosis as a potential explanation for high cancer mortality in individuals with a history of mental illness have been inconclusive. We examined the relationship between a SPI history and unknown cancer stage at diagnosis in colorectal cancer (CRC) patients. Methods This was a population-based, cross-sectional study using linked administrative databases of CRC patients diagnosed between 01/04/2007 and 31/12/2012. Individuals who had a history of mental illness but did not meet the definition of a SPI were excluded. An SPI was measured in the 5 years prior to the cancer diagnosis and categorized as inpatient, outpatient or no SPI. Individuals with a best stage in Stage 0 to Stage IV were considered staged and absence of staging information was defined as unknown stage. The risk of unknown stage cancer was estimated using modified Poisson regression. Results The final study cohort included 24,507 CRC patients. 258 (1.1%) individuals experienced a history of inpatient SPI and 482 (2.0%) experienced outpatient SPI. After adjusting for confounders, CRC patients with an inpatient or outpatient history of SPI were at greater risk of having missing TNM stage at diagnosis, compared to patients with no history of a mental illness (RR 1.45 (95% CI: 1.14–1.85) and RR1.17 (95% CI 0.95–1.43), respectively). The results did not change when alternate practices to assign SPI history using administrative data were used. Conclusions Individuals with an SPI, especially those with a psychiatric admission, were more likely to have missing stage data compared to individuals without a history of a mental illness. Incomplete and low quality cancer staging data likely undermines the quality of cancer care following initial diagnosis. Understanding why patients with an SPI are missing this information is a critical first step to providing excellent care to this vulnerable population.

Introduction Le cancer du sein est détecté soit par un examen de dépistage, soit à l’aide de signes et symptômes. Au Canada, les mammographies pour le dépistage du cancer du sein sont offertes dans le cadre de programmes organisés ainsi qu'en contexte indépendant (dépistage opportuniste). Les unités d’évaluation diagnostique (UED) de la province de l’Ontario sont des programmes en établissement qui fournissent des services diagnostiques coordonnés pour le cancer du sein. à la différence des soins habituels où le fournisseur de soins de première ligne organise les examens et les consultations. Cette étude décrit les méthodes de détection, l’intervalle diagnostique et l’utilisation des UED pour le cancer du sein en Ontario. Méthodologie L’étude a porté sur une cohorte de 6 898 femmes ayant reçu un diagnostic de cancer du sein envahissant en 2011. Nous avons utilisé le Registre d’inscription des cas de cancer de l’Ontario jumelé à des bases de données administratives sur les soins de santé. Nous avons déterminé la méthode de détection à l’aide des données du Programme ontarien de dépistage du cancer du sein (PODCS) et des demandes de règlement des médecins. L’intervalle diagnostique a été défini comme le temps écoulé entre le dépistage initial, l’aiguillage vers un spécialiste ou la première épreuve diagnostique et le diagnostic de cancer lui-même. Le parcours diagnostique (qu'il passe par les UED ou les soins habituels) a été déterminé en fonction des dossiers du PODCS et du lieu de biopsie ou d’intervention chirurgicale. Nous avons cartographié l’intervalle diagnostique et de la couverture des UED en fonction du lieu de résidence des femmes. Résultats En 2011, 36 % des cas de cancer du sein en Ontario ont été détectés par dépistage, dont 48 % chez des femmes de 50 à 69 ans. L’intervalle diagnostique provincial médian était de 32 jours, les médianes par comté variant entre 15 et 65 jours. À l’échelle provinciale, 48,4 % des cas ont été diagnostiqués dans une UED, ce pourcentage variant entre 0 et 100 % selon les comtés. Conclusion Le taux de détection au dépistage correspondant aux patientes admissibles du fait de leur âge s'est révélé inférieur au taux de dépistage officiel du cancer du sein pour l’ensemble de la population. La répartition géographique de l’intervalle diagnostique et du recours aux UED révèle des variations régionales dans les soins diagnostiques en oncologie qu'il est nécessaire de corriger.

Delays in surgery for patients with acute appendicitis may increase the risk of perforation and complications. These risks must be balanced with the theoretical risk of nighttime operations.To determine the effect of (i) time of day of surgery and (ii) time between triage presentation and surgery, on surgical complications in patient undergoing appendectomy for appendicitis.Population level data from Ontario from 2009 to 2015 was used to conduct this retrospective cohort study. The primary exposure was (i) time of day of surgery and (ii) time between triage and surgery. The primary outcome was a composite outcome of perioperative complications. Conditional logistic regression adjusted for patient factors was used to estimate adjusted odds ratios (OR) between the exposure and complications.Twelve thousand nine hundred thirty-seven matched pairs (N = 25,874) of adult patients were included. Compared to day operations, the odds of complications were lower during the evening (OR, 0.90; 95% confidence interval [CI], 0.84-0.96) and no different at night (OR, 0.90; 95% CI, 0.77-1.04). The odds of complications were higher in patients who waited longer than 24 hours for surgery compared with those who waited less than 6 hours (OR, 1.27; 95% CI, 1.14-1.43) with no increased risk in those who waited 6 hours to 12 hours or 12 hours to 24 hours.This large population study is the first to assess surgical risks using a case-crossover design to account for the surgeon as a potential confounder. Nighttime surgery and delaying surgery up to 24 hours were not associated increased risks of complications.Therapeutic, level III.

BACKGROUND Colonoscopy is a key resource used to diagnose colorectal cancer (CRC). This study evaluated the relationship between colonoscopy availability and the length of the CRC diagnostic interval. METHODS This is a cross-sectional study of CRC patients diagnosed in Ontario, Canada, in 2008-2012. We used administrative health data to characterise colonoscopist density, private colonoscopy clinic access, distance to the closest colonoscopist and the diagnostic interval, defined as the time from patients' first cancer-related healthcare encounter to their cancer diagnosis date. We used multivariable quantile regression to evaluate the association between colonoscopy availability and the diagnostic interval, modelling the median and 90th percentile. RESULTS The median diagnostic interval was 84 days (90th percentile 323 days). The diagnostic interval was longer in patients residing in areas with lower colonoscopists density or private clinic access (adjusted median difference = 9 and 19 days, respectively), with evidence of effect modification by symptom status. Increased distance to a colonoscopist was associated with a longer diagnostic interval in asymptomatic patients, but a shorter diagnostic interval in symptomatic patients (adjusted median difference = 29 and -25 days, respectively). CONCLUSIONS This study demonstrated that reduced colonoscopy resource availability is associated with longer diagnostic intervals for CRC patients.

Mortality reduction attributable to organized breast screening is modest. Screening may be less effective at detecting more aggressive cancers at an earlier stage. This study was conducted to determine the relative efficacy of screening mammography to detect cancers at an earlier stage by molecular phenotype. We identified 2882 women with primary invasive breast cancer diagnosed between January 1, 2008 and December 31, 2012 and who had a mammogram through the Ontario Breast Screening Program in the 28 months before diagnosis. Five tumor phenotypes were defined by expression of estrogen (ER) and progesterone (PR) receptors and HER2/neu oncogene. We conducted univariable and multivariable analyses to describe the predictors of detection as an interval cancer. Additional analyses identified predictors of detection at stages II, III, or IV compared with stage I, by phenotype. Analyses were adjusted for the effects of age, grade, and breast density. ER negative and HER2 positive tumors were over-represented among interval cancers, and triple negative cancers were more likely than ER +/HER2 − cancers to be detected as interval cancers OR 2.5 (95% CI 2.0–3.2, p

Aims To describe the quality of the non-technical component of the care (personal care) of patients receiving radical radiotherapy for prostate cancer and to identify elements of personal care that should be priorities for quality improvement. Materials and methods One hundred and eight patients undergoing radiotherapy for localised prostate cancer completed a self-administered questionnaire that asked them to rate the importance of 143 non-technical elements of care and to rate the quality of their own care with respect to each element. The elements that a patient rated as both ‘very important' and less than ‘very good' were deemed to be his priorities for improvement. The priorities of the population were established by ranking the elements based on the percentage of patients who identified them as a priority (importance/quality analysis). Results The response rate was 65%. The percentage of elements rated ‘very good' varied from patient to patient: median 79% (interquartile range 69–92%). The percentage of elements rated either ‘very good' or ‘good' was higher: median 96% (interquartile range 86–98%). Nonetheless, almost every patient rated at least some elements of his care as less than optimal, regardless of the cut-off point used to define optimal quality. Patients assigned their lowest quality ratings to elements relating to the quality of the treatment environment and comprehensiveness of additional services available to them. However, patients rated most of these elements as relatively unimportant, and importance/quality analysis identified elements of care relating to communication of information about the disease and its treatment as the highest priorities for quality improvement. Conclusions Most patients rated most elements of their personal care as very good, but almost all were able to identify some elements that were less than optimal. When ratings of quality were integrated with ratings of importance, elements relating to communication emerged as the patients' highest priorities for quality improvement.

336 Background: Pancreatic cancer is a leading cause of cancer death, largely due to vague presenting symptoms and late stage at diagnosis. Population-based administrative data can be a valuable resource for studying the diagnostic interval. The objective of this study was to determine the first encounter in the diagnostic interval and to calculate that interval in a cohort of patients with pancreatic cancer using an empirical approach. Methods: This is a retrospective, cohort study of patients diagnosed with pancreatic ductal adenocarcinoma (PDAC) from 2007 – 2015 in Alberta, Canada. We used the Alberta Cancer Registry (ACR), physician billing claims, hospital discharge and emergency room visits to identify health encounters that occurred more frequently in the 3 months prior to diagnosis compared to those in the 3-24 months prior to diagnosis. We used statistical control charts to define the lookback period for each encounter category and identify the earliest encounter that represented the start of the diagnostic interval (index contact date). The end of the interval was the diagnosis date. Quantile regression was used to determine factors associated with the diagnostic interval. Results: We identified 3142 patients with PDAC. Median age of diagnosis was 71 (IQR 61-80). We identified an index contact date in 96.5% of the patients. The median length of the diagnostic interval was 76 days (IQR 21-191; 90th percentile 276 days). A higher Elixhauser comorbidity score (+18.57 days/ 1 point increase, 95% CI 16.07-21.07, p < 0.001) and stage 3 disease (+22.55 days, 95% CI 5.02-40.08, p = 0.01) was associated with a longer diagnostic interval. Conclusions: In this cohort of patients with pancreatic cancer, there was a wide range in the diagnostic interval with 10% of patients having a diagnostic interval approaching one year. Diagnostic interval research using administrative databases can understand variations in diagnosis times, can inform early detection efforts and can improve quality of care.

Introduction There is an evidence gap regarding the duration of SARS-CoV-2 shedding and of its variability across different care settings and by age, sex, income, and co-morbidities. Such evidence is part of understanding of infectivity and reinfection. We examine direct measures of viral shedding using a linked population-based health administrative dataset. Methods Laboratory and sociodemographic databases for Ontario, Canada were linked to identify those testing positive (RT-PCR) between Jan. 15 and April 30, 2020 who underwent subsequent testing by May 31, 2020. To maximise use of available data, we computed two shedding duration estimates defined as the time between initial positive and most recent positive (documented shedding) or second of two negative tests (documented resolution). We also report multivariable results using quantile regression to examine subgroup differences. Results In Ontario, of the 16,595 who tested positive before April 30, 2020, 6604 had sufficient subsequent testing to allow shedding duration calculation. Documented shedding median duration calculated in 4,889 (29% of 16,595) patients was 19 days (IQR 12–28). Documented resolution median duration calculated in 3,219 (19% of the 16,595) patients was 25 days (IQR 18–34). Long-term care residents had 3–5 day longer shedding durations using both definitions. Shorter documented shedding durations of 2–4 days were observed in those living in higher income neighbourhoods. Shorter documented resolution durations of 2–3 days were observed at the 25th% of the distribution in those aged 20–49. Only 11.5% of those with definitive negative test results reverted to negative status by day 14. Conclusions Viral shedding continued well beyond 14 days among this large subset of a population-based group with COVID-19, and longer still for long-term care residents and those living in less affluent neighborhoods. Our findings do not speak to duration of infectivity but are useful for understanding the expected duration of RT-PCR positivity and for identifying reinfection.