Umeå University's logo

Importance: Population-based colorectal cancer (CRC) screening with fecal occult blood tests (FOBT) is widely implemented. Estimating associations of invitation to and participation in the screening program with CRC mortality is challenging due to lack of valid control groups and uncertainties about optimal follow-up time.

Objective: To estimate associations of invitation to and participation in screening with CRC mortality reduction when adjusting death from CRC diagnosed more than 2 years after cessation of screening and nonadherence.

Design, Setting, and Participants: This prospective cohort study was conducted in the Stockholm-Gotland region of Sweden between 2008 and 2021. Yearly birth cohorts of individuals aged 60 to 69 years (born 1938-1954) were randomly assigned to a study group (SG) invited to screen between 2008 and 2012 or to a control group (CG) who were either invited to screen between 2013 and 2015 or were not invited to screen. Analyses were performed between May 2024 and February 2026.

Exposure: Invitation to and participation in CRC screening, adjusting for contamination in the SG (exposed), invitations in the CG (not exposed), and nonadherence. Main Outcome and Measures: CRC-specific mortality, defined as CRC in the Cancer Register (2008-2020) or the Swedish Quality Register for Colorectal Cancer (2021), with CRC as underlying cause of death in the Cause of Death Register. Poisson regression was used to estimate mortality rate ratios (RRs) with 95% CIs, adjusting for follow-up year, attained age at follow-up, contamination from CRC deaths diagnosed more than 2 years after screening in the SG, invitations in the CG, and nonadherence.

Results: The study included 376 511 individuals, with 203 692 in the SG (99 472 men [48.8%]; mean [SD] age at start of follow-up, 63.6 [3.2] years) and 172 819 in the CG (84 705 men [49.0%]; mean [SD] age at start of follow-up, 62.8 [3.5] years). During follow-up, 856 CRC deaths occurred in the SG and 812 CRC deaths occurred in the CG. After adjustment for CRC deaths diagnosed more than 2 years after the last SG invitation (455 deaths) and invitations in the CG (166 deaths), the RR was 0.74 (95% CI, 0.58-0.94), a 26% reduction. After further adjustment for nonadherence in the SG (196 deaths) and participation in the CG (82 deaths), the RR for participation was 0.57 (95% CI, 0.40-0.80), a 43% reduction.

Conclusions and Relevance: In this cohort study of population-based FOBT screening, screening invitation was associated with a 26% reduction in CRC mortality and participation was associated with a 43% reduction, after accounting for contamination and nonadherence. These findings suggest that bias adjustment is important when evaluating screening programs, and these findings support efforts to increase screening uptake.

Background: The effectiveness of an intervention such as cancer screening can be estimated by conducting a cohort study estimating the rate ratio between an exposed group invited to screening and a control group not invited to screening. A common issue is non-compliance, where not all individuals in the study group are exposed. Ignoring non-compliance can result in biased estimates of the exposure effect, but excluding non-exposed individuals may also be problematic as they may differ in risk profile from those who were exposed. A similar problem arises when members of the control group are inadvertently exposed (contamination). Observational studies face additional challenges due to confounding.

Objective: To report the development of a method to adjust rate ratio estimates in cohort studies for contamination and non-compliance, that also adjusts for confounding.

Method: Derivation of the new method is outlined.

Results: The method is illustrated in two examples.

Conclusion: The results are comparable with a stratified estimate, but through the use of a Poisson regression model the range of possible analyses is extended, for example to tests of confounding factors and interaction.

Aims/hypothesis: The aim of this work was to investigate whether baseline uric acid (UA) is associated with future complications among young individuals with newly diagnosed type 1 diabetes.

Methods: UA levels were analysed in individuals, aged 15–34 years with newly diagnosed type 1 diabetes, from the nationwide Diabetes Incidence Study in Sweden (DISS) cohort to assess the relationship with macro- and microvascular complications later in life. Information on complications was obtained by record linkage to the National Diabetes Registry of Sweden and the National Patient Registry of Sweden. Individuals who developed complications during follow-up (n = 94) were matched for year and age at diagnosis (±2 years), sex and HbA1c with control individuals (n = 94) without complications.

Results: Plasma UA levels at the time of diabetes diagnosis were significantly higher in individuals who later developed diabetes-related complications compared with those who did not, after a median follow-up of 19.0 years (IQR 16.3–21.0): 209.2 ± 68.9 vs 171.7 ± 50.2 µmol/l (p<0.001). The odds of developing complications were 1% higher for every 1 μmol/l rise in baseline UA, and individuals within the highest quartile of UA were more than three times more likely to develop diabetes-related complications later in life after adjusting for age, HbA1c, smoking and eGFR.

Conclusions/interpretation: This study indicates that higher baseline UA levels at the time of type 1 diabetes diagnosis may be linked to both macrovascular and microvascular complications later in life.

Aim: Evaluation of a breast cancer (BC) screening programme is necessary to ensure its quality. Performance measurements might be prioritized considering the association with outcomes related to BC mortality. We piloted an approach to explore the association of selected performance measurements with incidence-based BC mortality (IBM).

Methods: We performed an ecological analysis of aggregated data from regional or national population-based cancer registries and BC screening programmes in Europe, using 13 performance indicators. We built a panel data (longitudinal cross-sectional) regression model to estimate the association between screening performance measurements and IBM rates.

Results: We included data of 9 programmes and registries from Italy, Spain, Norway, Ireland and the Czech Republic. The number of screening years included in the dataset ranged from 5 to 20 years. In adjusted panel analyses, higher screening coverage, breast cancer detection rates (BCDR prevalent and subsequent rounds), node-negative proportion, and episode sensitivity were associated with lower incidence-based mortality (IBM), whereas a higher interval cancer rate was associated with higher IBM. The association for recall rate in subsequent examinations was small and imprecise.

Conclusion: Our pilot approach suggests association of several performance indicators with IBM. These indicators were related to the implementation of the screening programme (screening coverage), sensitivity (BC detection rate), and efficiency (recall rate). Further studies with larger datasets and individual data may confirm these findings.

BACKGROUND AND PURPOSE: Adjuvant endocrine treatment (AET) is crucial in early oestrogen receptor (ER)-positive breast cancer (BC), providing reduced recurrence rate and increased overall survival. The aim of this study was to estimate AET adherence rates by age at diagnosis and region in Sweden.

PATIENTS AND METHODS: In total, 10,422 women diagnosed with ER-positive BC in 2008-2010 were identified in the Swedish National BC Registry. Information on prescriptions and dispensation of AET was gathered through record linkage to the Swedish Prescription Registry. 1, 3- and 5-year medication possession ratios (MPRs) were calculated. Good adherence was set as MPR ≥ 80%.

RESULTS: The 1-, 3- and 5-year AET age-adjusted adherence rates were 94.4, 87.6 and 81.6%, respectively. The 1-, 3- and 5- year adherence rate was significantly highest in the South region (96.2, 90.5 and 86.2%). Regions with an oncologic clinic had higher adherence rate than regions without, 82.8% versus 75.5% at 5-year FU. Women at age 40-64 years (95.6, 89.9 and 84.1%) and 65-74 years at diagnosis (95.7, 89.5 and 84.6%) had significantly higher adherence rate than women ≥ 75 years at diagnosis (89.1, 79.2 and 68.3%).

INTERPRETATIONS: Despite guidelines being national, there were significant differences in adherence between regions in Sweden. As the largest differences were between age groups invited and not invited to mammography screening intervention should focus on women < 40 and ≥ 75 years at diagnosis. Further studies are needed to find strategies to increase overall adherence to AET in early BC.

BACKGROUND AND PURPOSE: We have recently demonstrated that screen-detected invasive breast cancers had more favourable tumour characteristics than non-screen-detected. The objective of the study was to analyse differences in breast cancer treatment between screen-detected and non-screen-detected cases by age at diagnosis, with and without adjustment for tumour (T) and nodal (N) status, within a nationwide, population-based mammography screening programme utilising register data.

MATERIAL AND METHODS: Data spanning 2008-2017 were collected from the National Quality Register for Breast Cancer. Multivariable logistic regression analysis was used to estimate odds ratios and 95% confidence intervals for treatment disparities between screen-detected and non-screen-detected breast cancer.

RESULTS: Among 46,481 women diagnosed with invasive breast cancer aged 40-74 and invited for mammography screening, significant differences in treatment were observed. Screen-detected cases showed higher likelihoods of partial mastectomy compared to mastectomy, endocrine therapy, and radiotherapy, whereas chemotherapy and antibody therapy were less likely compared to non-screen-detected cases. However, when adjusting for surgery type, screen-detected cases showed lower likelihoods of radiotherapy. Age at diagnosis significantly influenced treatment odds ratios, with interactions observed for all treatments except radiotherapy adjusted for surgery. Differences increased with age, except for endocrine therapy. Radiotherapy adjusted for surgery type showed no age-related interaction. Adjusting for T and N did not alter these patterns.

INTERPRETATION: In general, screen-detected cases received less aggressive treatment, such as mastectomy, chemotherapy, and antibody therapy, compared to non-screen-detected cases. Disparities increased with age, except for endocrine therapy and radiotherapy adjusted for surgery. Differences persisted after adjusting for T and N, suggesting that these factors cannot solely explain the results.

Objective:  To analyze differences between screen-detected and non-screen-detected invasive breast cancers by tumour characteristics and age at diagnosis in the nationwide population-based mammography screening program in Sweden.

Methods:  Data were retrieved from the National Quality Register for Breast Cancer for 2008-2017. Logistic regression analysis was used to estimate the likelihood for a tumour to be screen-detected by tumour characteristics and age group at diagnosis.

Results:  In total there were 51,429 invasive breast cancers in the target age group for mammography screening of 40-74 years. Likelihood of screen detection decreased with larger tumour size, lymph node metastases, higher histological grade and distant metastasis. Odds ratios (ORs) for negative oestrogen (ER) and progesterone (PgR) were 0.41 and 0.57; for positive HER2, 0.62; for Ki-67 high versus low, 0.49. Molecular sub-types had OR of 0.56, 0.40 and 0.28, respectively, for luminal B-like, HER2-positive and triple negative versus luminal A-like. Adjusting for tumour size (T), lymph node status (N), age, year and county at diagnosis slightly elevated the ORs. Statistically significant interactions between tumour characteristics and age were found (p < 0.05) except for ER and PgR. The age group 40-49 deviated most from the other age groups.

Conclusions:  Our study demonstrates that screen-detected invasive breast cancers had more favourable tumour characteristics than non-screen-detected after adjusting for age, year and county of diagnosis, and even after adjusting for T and N. The trend towards favourable tumour characteristics was less pronounced in the 40-49 age group compared to the other age groups, except for ER and PgR.

Introduction: We implemented two cross-sectional reproductive age mortality surveys in 2007–2008 and 2018–2019 to assess changes in the MMR and causes of death in Zimbabwe. We collected data from health institutions, civil registration and vital statistics, the community, and surveillance. This paper analyses missingness and misclassification of deaths in the two surveys.

Methods: We compared proportions of missed and misclassified deaths in the surveys using Chi-square or Fisher’s exact tests. Using log-linear regression models, we calculated and compared risk ratios of missed deaths in the data sources. We assessed the effect on MMRs of misclassifying deaths and analysed the sensitivity and specificity of identifying deaths in the surveys using the six-box method and risk ratios calculated through Binomial exact tests.

Results: All data sources missed and misclassified the deaths. The community survey was seven times [RR 7.1 (5.1–9.7)] and CRVS three times [RR 3.4 (2.4–4.7)] more likely to identify maternal deaths than health records in 2007–08. In 2018–19, CRVS [RR 0.8 (0.7–0.9)] and surveillance [RR 0.7 (0.6–0.9)] were less likely to identify maternal deaths than health records. Misclassification of causes of death significantly reduced MMRs in health records [RR 1.4 (1.2–1.5)]; CRVS [RR 1.3 (1.1–1.5)] and the community survey/surveillance [RR 1.4 (1.2–1.6)].

Conclusion: Incompleteness and misclassification of maternal deaths are still high in Zimbabwe. Maternal mortality studies must triangulate data sources to improve the completeness of data while efforts to reduce misclassification of deaths continue.

PURPOSE: Education and feedback on hypertension management has been associated with improved hypertension control. This study aimed to assess the effectiveness of such interventions to reduce the risk of stroke and cardiovascular events. MATERIALS AND METHODS: Individuals ≥18 years with a blood pressure (BP) recording in Västerbotten or Södermanland County during the study period 2001 to 2009 were included in 108 serial cohort studies, each with 24 months follow-up. The primary outcome was risk of first-ever stroke in Västerbotten County (intervention) compared with Södermanland County (control). Secondary outcomes were first-ever major adverse cardiovascular event (MACE), myocardial infarction, and heart failure, as well as all-cause and cardiovascular mortality. All outcomes were analysed using time-to-event data included in a Cox proportional hazards model adjusted for age, sex, hypertension, diabetes, coronary artery disease, atrial fibrillation, systolic BP at inclusion, marital status, and disposable income. RESULTS: A total of 121 365 individuals (mean [SD] age at inclusion 61.7 [16.3] years; 59.9% female; mean inclusion BP 142.3/82.6 mmHg) in the intervention county were compared to 131 924 individuals (63.6 [16.2] years; 61.2% female; 144.1/81.1 mmHg) in the control county. A first-ever stroke occurred in 2 823 (2.3%) individuals in the intervention county, and 3 584 (2.7%) individuals in the control county (adjusted hazard ratio 0.96, 95% CI 0.90 to 1.03). No differences were observed for MACE, myocardial infarction or heart failure, whereas all-cause mortality (HR 0.91, 95% CI 0.87 to 0.95) and cardiovascular mortality (HR 0.91, 95% CI 0.85 to 0.98) were lower in the intervention county. CONCLUSIONS: This study does not support an association between education and feedback on hypertension management to primary care physicians and the risk for stroke or cardiovascular outcomes. The observed differences for mortality outcomes should be interpreted with caution.