Clinical Epidemiology最新文献

Purpose: Self-measured blood pressure (SMBP) monitoring, or home blood pressure monitoring, is an evidence-based strategy for hypertension management. However, the extent to which SMBP readings are documented in the electronic health record (EHR)-a critical step in integrating SMBP monitoring into routine care-remains unclear. The objective of this study was to evaluate how race/ethnicity, insurance status, and preferred language are associated with documented SMBP monitoring adoption.

Patients and methods: This retrospective study included adults (aged ≥18 years) with a diagnosis of hypertension and at least one uncontrolled ambulatory blood pressure reading in 2023. All patients had ≥2 primary care visits in 2023. The primary outcome was the presence of at least one SMBP reading documented in the EHR using a structured patient-reported blood pressure readings field. Logistic regression was used to assess associations between patient characteristics and SMBP monitoring documentation.

Results: Among 156,444 eligible patients, only 5.0% had at least one SMBP reading documented in the EHR. SMBP readings were mostly recorded during office visits (62.8%). In fully adjusted analyses, Black (OR 0.68; 95% CI, 0.63-0.72) and Hispanic (OR 0.72; 95% CI, 0.65-0.80) patients compared to White patients, Medicaid-insured patients (OR 0.88; 95% CI, 0.79-0.98) compared to Commercial-insured patients, those preferring non-English languages (OR 0.77; 95% CI, 0.67-0.88) compared to those preferring English, had lower odds of SMBP monitoring documentation.

Conclusion: SMBP monitoring documentation in the EHR was rare and significantly lower among populations experiencing inequities in access, language, and outcomes. Although remote monitoring strategies show promise, poor EHR integration and scalability challenges limit their adoption. Targeted efforts are needed to improve SMBP monitoring documentation workflows, enhance EHR integration, and promote equitable access to hypertension self-management tools.

Purpose: To compare epidemiological estimates of 19 different cancer types in Clinical Practice Research Datalink (CPRD) Aurum and CPRD General Practice Online Database (GOLD) databases against linked secondary data sources in England to understand best use of these data sources for research.

Methods: The source population comprised patients in CPRD Aurum or GOLD (separately) who were eligible for linkage to Cancer Registry (CR), Hospital Episode Statistics (HES), and Office for National Statistics (ONS). We selected patients who had an incident cancer diagnosis recorded in ≥1 data sources (CPRD Aurum or GOLD, HES, CR) between January 1, 2011 and December 31, 2018. We estimated incidence rates (IR) and counts by cancer type and data source, and survival probability by cancer type among patients with an ONS death record recorded between January 1, 2011 and April 30, 2020.

Results: The highest incident case capture resulted from CPRD Aurum or GOLD linked to HES and CR. In the fully linked CPRD Aurum-HES-CR and CPRD GOLD-HES-CR datasets, cancers typically diagnosed and managed in primary care (eg, breast, prostate, and lung) had the highest IRs and more complete case capture compared with other data sources, whereas HES and CR had higher IRs for cancers diagnosed in secondary care settings (eg, gastric, renal, and bladder). Cancers with broad definitions (eg, head and neck) had wider variations in IRs across data sources than cancers with narrower definitions. Survival estimates were generally higher for cancer-related deaths versus all-cause deaths.

Conclusion: Findings highlight variation in cancer recording across different data sources. Researchers using CPRD data should assess the benefit of incorporating linked data on a study-by-study basis. For studies of breast, prostate, and lung cancers, CPRD Aurum or GOLD alone may be sufficient; however, linkage to HES and/or CR is recommended where a more complete case capture is required.

Purpose: To investigate the potential impact of changes in patient reported outcome measures (PROMs) on subsequent healthcare use following a self-management supportive program for low back pain.

Patients and methods: Clinical data from a cohort of 2803 participants enrolled in a self-management supportive intervention for LBP (2018 to 2022) was linked with Danish national registry data. PROMs (predictors) were collected at baseline and at 3-months follow-up (end of intervention), including pain intensity, disability, illness perceptions, self-efficacy, and health-related quality of life. Healthcare use (outcome) was measured as the total use in the year before and after the intervention for i) visits to physiotherapists or chiropractors and ii) analgesics use. Associations between standardized changes in PROMs and changes in healthcare use were analyzed using zero-inflated negative binomial regression models and reported as incidence rate ratios (IRRs) with 95% confidence intervals (CIs). Subgroup analyses were conducted for low, medium, and high baseline use of each outcome.

Results: Across PROMs, improvements from baseline to 3-months follow-up were associated with reductions in subsequent healthcare use. For example, one standard deviation improvement in back pain, disability, and self-efficacy were associated with 6-8% reductions in visits (adjusted IRRs [95% CI], back pain: 0.92 [0.88, 0.97]; disability: 0.94 [0.90, 0.99]; self-efficacy: 0.92 [0.87, 0.96]). For analgesic use, improvements in back pain and self-efficacy were related to a 4% reduction (adjusted IRR [95% CI], back pain: 0.96 [0.92, 1.01], self-efficacy 0.96 [0.91, 1.00]). Stronger associations were observed in subgroups with medium visit use and medium or low analgesic use at baseline.

Conclusion: On a group-level, improvements across PROMs were associated with reductions in subsequent healthcare use following participation in a self-management supportive intervention. The strength of the associations varied across subgroups and additional factors not measured by the PROMs are likely to also influence changes in healthcare use.

Background: Antibody-drug conjugates (ADCs) represent a transformative class of therapeutics for advanced bladder cancer. However, their real-world safety profiles are not yet fully characterized.

Methods: This retrospective pharmacovigilance study analyzed data from the FDA Adverse Event Reporting System (FAERS) from the first quarter of 2004 to the third quarter of 2024. Disproportionality analyses, including the reporting odds ratio (ROR), proportional reporting ratio (PRR), and Bayesian confidence propagation neural network (BCPNN), were used to detect significant adverse drug event (ADE) signals for four ADCs in bladder cancer treatment: enfortumab vedotin (EV), sacituzumab govitecan (SG), trastuzumab deruxtecan (DS-8201), and trastuzumab emtansine (T-DM1).

Results: Among 494 analyzed reports, EV constituted the majority (91.7%). Distinct safety signals were identified for each ADC: EV was strongly associated with skin disorders and metabolic disturbances; SG was primarily linked to gastrointestinal events, with emerging signals of renal abnormalities; DS-8201 was associated with systemic administration-related issues; and T-DM1 showed signals for respiratory and bleeding events. Notably, oral candidiasis related to EV was not explicitly highlighted in the current prescribing information.

Conclusion: This study delineates the safety profiles of ADC therapies for bladder cancer, confirming known risks and identifying potential new signals. The findings highlight the need for ADC-specific monitoring strategies and proactive management protocols to mitigate toxicities, thereby providing essential evidence for clinical decision-making.

Background: The Charlson Comorbidity Index (CCI) is widely used to control confounding factors in epidemiological studies. There is a need to characterize the performance of CCI in a contemporary general population. This study assessed the discriminative ability of CCI for long-term mortality in relation to age, sex, data source, length of look-back period, and calendar time.

Methods: Open cohort study of all adults residing in Sweden for at least one year between 2003 and 2022. Multiple versions of the CCI were calculated based on 1-10 years of lookback in the Swedish National Patient Register. Discrimination was determined using the concordance index (C-index). Kaplan-Meier curves were used to describe 10-year survival.

Results: About 10395689 unique individuals had a median follow-up for mortality of 10 years. The C-index of CCI increased with the length of the look-back period, from 0.694 (95% confidence interval [CI], 0.694-0.694) at 1 year, to 0.784 (95% CI, 0.783-0.784) at 5 years, and to 0.808 (95% CI, 0.807-0.808) at 10 years. Discrimination was highest in subjects aged 60 years, for whom the discrimination of CCI_10years was higher in women (C-index, 0.794 [95% CI, 0.788-0.799]) than in men (C-index, 0.730 [95% CI, 0.725-0.735]). Within age strata above 70 years, the risk of death was much lower in subjects with CCI = 0 vs 1, 2 vs 3, and 3 vs ≥4, but not in those with CCI = 1 vs 2.

Conclusion: The CCI discriminated risk of death best using a look-back period of at least 5 years and performed better in women than in men in most age groups. Our findings offer insights into both the practical utility of the CCI and the interpretation of studies in which it has been applied.

Background: Taiwan's National Health Insurance Research Database (NHIRD) has evolved into a cornerstone of real-world evidence generation. As Taiwan's National Health Insurance program reaches its 30th anniversary, a comprehensive reassessment of the NHIRD's development, challenges and future directions is warranted.

Objective: To provide an updated review of the NHIRD.

Methods: We conducted a narrative review of Taiwan's NHIRD, synthesizing published studies, government reports, and policy documents from 2019 through 2024. We summarized developments related to database infrastructure, data collection, validation studies, linkage strategies, governance reforms and interoperability initiatives, with a particular focus on their implications for real-world evidence generation and AI-driven research.

Results: The NHIRD has additionally incorporated structured laboratory results and medical imaging data, significantly broadening its research capabilities. Validation studies have demonstrated the reliability of International Classification of Diseases, 10th Revision, Clinical Modification codes across various conditions, reinforcing the database's applicability for epidemiological research. Integration efforts with national registries, surveys and electronic medical records have further enhanced the depth and accuracy of clinical outcome measurements. Nonetheless, critical challenges persist, including data standardization inconsistencies, cybersecurity vulnerabilities, and heightened scrutiny following constitutional court rulings on data governance and privacy rights. In response, Taiwan's Ministry of Health and Welfare has launched initiatives to address these concerns, notably through the development of a Fast Healthcare Interoperability Resources (FHIR)-based data infrastructure aimed at improving interoperability, data security, and artificial intelligence (AI)-readiness to balance ethical governance with scientific innovation.

Conclusion: The NHIRD's transformation over three decades underscores the importance of continuous investment in data quality, privacy protection, and interoperability. With sustained reforms, the NHIRD will be poised to remain a leading resource for real-world evidence generation and to contribute meaningfully to global health and digital medicine.

Purpose: We examined the association between socioeconomic position (SEP) and risk of any infection after surgery for hip fracture, and whether markers of poor health modify this.

Methods: Individual-level data on SEP markers (education, liquid assets, marital status, and cohabitation) were obtained from Danish registries for hip fracture patients undergoing surgery (2010-2018). We computed cumulative incidences of any hospital-treated infection within one month after surgery. Using Cox regression we estimated adjusted hazard ratios (aHRs) with 95% confidence intervals. Analyses were stratified by comorbidity clusters based on latent class analysis, body mass index (BMI), pre-fracture mobility, and residence type.

Results: The incidences of infection were: 17% for low vs 16% for high education (aHR 1.10, 1.02-1.18), 19% for low vs 16% for high liquid assets (aHR 1.21, 1.15-1.28), 18% for divorced vs 16% for married (aHR 1.24, 1.15-1.32), and 18% for living alone vs 15% for cohabiting (aHR 1.16, 1.06-1.28). The incidence of infection was highest among patients with diabetic-renal comorbidity, underweight, poor mobility, or nursing home residency. The magnitude and direction of associations were modified by comorbidity clusters, BMI, mobility, and residence type.

Conclusion: We observed socioeconomic inequalities in 30-day risk of infection after hip fracture surgery. Health modified the observed inequalities but could not fully explain them.

Objective: Comorbidity indices are often used to adjust for confounding in epidemiological studies. However, the performance of comorbidity indices may vary depending on the clinical context. In the present study, we aimed to assess the incremental value of different comorbidity indices in predicting mortality in Danish pancreatic cancer patients.

Methods: We conducted a nationwide cohort study of Danish patients diagnosed with pancreatic cancer from 2004 to 2022. Using national healthcare registries, we assessed comorbidities through five indices: Charlson, Elixhauser, van Walraven, Gagne, and Nordic Multimorbidity. We evaluated the added prognostic value of these indices using different lookback periods for predicting one-year mortality using logistic regression models with and without comorbidity scores to a basis model consisting of demographic characteristics, year of diagnosis, and tumour stage. Model performance was assessed by area under the receiver operating characteristic curve (AUC). We also conducted a sensitivity analysis restricting to patients undergoing surgery.

Results: We included 10,413 patients diagnosed with pancreatic cancer during the study period. Tumour stage was the strongest predictor of mortality, increasing the AUC from 0.64 to 0.82. Adding any comorbidity index provided no meaningful improvement (AUC remained 0.82-0.83). Results were consistent across different lookback periods and in the analysis restricted to patients undergoing surgery.

Conclusion: Comorbidity indices offer minimal additional prognostic value for mortality in pancreatic cancer beyond tumour stage and basic demographic factors.

Introduction: Esophageal and gastric cancers account for nearly 1.5 million new cases and 1.1 million deaths annually worldwide. In western countries, the incidence of esophageal cancer is rising while that of gastric cancer has decreased, although the pattern varies between the morphological types and subsites. We aim to describe the burden of esophageal and gastric cancers in Norway by providing national trends in incidence and mortality, separately for esophageal squamous cell carcinoma (SCC) and adenocarcinoma (AC), and for gastric ACs by gastric subsites.

Methods: We extracted information about all esophageal (ICD10 C15) and gastric cancer (ICD10 C16) patients diagnosed 1993‒2022 from the Cancer Registry of Norway. Age-standardized (European standard population) rates and performed joinpoint regression analyses were calculated to examine trends in incidence and mortality over time, for esophageal cancer SCC and AC and by subsite for gastric AC (cardia: ICD10 C16.0 and non-cardia: ICD10 C16.1-9). We used annual percent change (APC) and weighted average APC (AAPC), stratified by sex, age group, and stage at diagnosis.

Results: During 1993-2022, 6,433 esophageal cancers (2,616 SCC, 3,817 AC) and 14,453 gastric AC were diagnosed, and 4,683 esophageal and 10,421 gastric AC deaths occurred. The incidence and mortality of esophageal ACs increased whereas the rates for esophageal SCC declined in men and were stable in women. The highest AC incidence and mortality increases were seen in men (incidence AAPC = 2.8) and ages ≥70 years (incidence AAPC = 5.9). In contrast, the incidence and mortality of gastric cancer decreased over time, most pronounced for non-cardia gastric AC (incidence AAPC men =-5.3, women =-3.9).

Conclusion: The incidence and mortality of esophageal AC has increased in Norway during the last decades, most pronounced in men, ages ≥70 years. The rates of SCCs decreased, although trends differed between sex and age groups. The incidence and mortality of gastric AC decreased in all age-groups for both sexes, especially for non-cardia gastric cancer.

Purpose: We applied a Bayesian approach to further investigate the association of sodium-glucose cotransporter-2 inhibitors (SGLT2i) with the composite outcome of Major Adverse Cardiovascular Event and Heart Failure hospitalization (MACE+HF) and its individual components leveraging the ability of a Bayesian approach to incorporate prior clinical information and to make probability statements about the parameters.

Methods: We use a Bayesian time-to-event model, where the covariates are directly modeled in the hazard function. Following propensity score matching, we fit three Bayesian models; one with a relatively flat, normal prior on the SGLT2i coefficient (Uninformative) and 2 with informative priors from a meta-analysis (based on a cohort with no history of cardiovascular disease [No CVD] and cohorts with a history of CVD [CVD]). We estimate the posterior distribution for the hazard ratio (HR) using a Hamiltonian Monte Carlo algorithm. It allows us to estimate the probability of a meaningful protective association (HR < 0.90) in addition to point and interval estimates.

Results: The posterior means and 95% credible intervals for the HR suggested a protective association for SGLT2i versus dipeptidyl peptidase 4 inhibitors (DPP4i) for the MACE+HF outcome: No CVD: 0.82 (0.68, 0.96), CVD: 0.82 (0.71, 0.94), and Uninformative: 0.79 (0.65, 0.94). The probability of a meaningful protective association for the No CVD, CVD, and Uninformative priors were 88%, 92%, and 93%, respectively. The probability of a meaningful protective association for the HF hospitalization, CVD hospitalization and CVD death components of MACE+HF were 95%, 67%, and 93%, respectively.

Conclusion: The Bayesian analysis allowed for the incorporation of prior information via an informative prior and further investigation of the association between SGLT2 and the components of the MACE+HF composite outcome. It allowed for the calculation of an easily interpretable summary measure, the probability of a meaningful protective association.