19 June 2023 in Expert Review of Pharmacoeconomics & Outcomes Research
The objective of this systematic review is to summarize the use of machine learning (ML) in predicting overall survival (OS) in patients with bladder cancer. Search terms for bladder cancer, ML algorithms, and mortality were used to identify studies in PubMed and Web of Science as of February 2022. Notable inclusion/exclusion criteria contained the inclusion of studies that utilized patient-level datasets and exclusion of primarily gene expression-related dataset studies. Study quality and bias were assessed using the International Journal of Medical Informatics (IJMEDI) checklist. Of the 14 included studies, the most common algorithms were artificial neural networks (n = 8) and logistic regression (n = 4). Nine articles described missing data handling, with five articles removing patients with missing data entirely. With respect to feature selection, the most common sociodemographic variables were age (n = 9), gender (n = 9), and smoking status (n = 3), with clinical variables most commonly including tumor stage (n = 8), grade (n = 7), and lymph node involvement (n = 6). Most studies (n = 10) were of medium IJMEDI quality, with common areas of improvement being the descriptions of data preparation and deployment. ML holds promise for optimizing bladder cancer care through accurate OS predictions, but challenges related to data processing, feature selection, and data source quality must be resolved to develop robust models. While this review is limited by its inability to compare models across studies, this systematic review will inform decision making by various stakeholders to improve understanding of ML-based OS prediction in bladder cancer and foster interpretability of future models.
16 March 2022 in Alzheimer Disease & Associated Disorders
Objective: This study examined the extent to which chronic comorbidities contribute to excess health care expenditures between older adults with dementia and propensity score (PS)-matched nondementia controls. Methods: This was a retrospective, cross-sectional, PS-matched case (dementia): control (nondementia) study of older adults (65 y or above) using alternative years data from pooled 2005 to 2015 Medical Expenditure Panel Surveys (MEPS). Chronic comorbidities were identified based on Clinical Classifications System or ICD-9-CM codes. Ordinary least squares regression was utilized to quantify the impact of chronic comorbidities on the excess expenditures with logarithmic transformation. Expenditures were expressed as 2019 US dollars. All analyses accounted for the complex survey design of MEPS. Results: The mean yearly home health care expenditures were particularly higher among older adults with dementia and co-occurring anemia, eye disorders, hyperlipidemia, and hypertension compared with PS-matched controls. Ordinary least squares regression models revealed that home health care expenditures were 131% higher (β=0.837, P<0.001) among older adults with dementia compared with matched nondementia controls before adjusting for chronic comorbidities. When additionally adjusting for chronic comorbidities, the percentage increase, while still significant (P<0.001) decreased from 131% to 102%. Conclusions: The excess home health care expenditures were partially explained by chronic comorbidities among community-dwelling older adults with dementia.
21 April 2025 in Therapeutic Advances in Drug Safety
Background: Novel antidiabetic medications (SGLT-2 inhibitors, DPP-4 inhibitors, and GLP-1 agonists) are commonly used worldwide; however, the available research lacks definitive conclusions on their protective effects or potential risks on cancer. Objectives: Compared to other antidiabetics, our systematic review and network meta-analysis (NMA) aims to use real-world studies to assess the potential cancer risks or protective effects of these novel antidiabetics. Methods: We comprehensively searched PubMed, CINAHL, and Web of Science from their inception until November 30, 2023. We included observational studies examining at least one novel antidiabetics in the systematic review. The novel antidiabetics include sodium-glucose cotransporter-2 inhibitors (SGLT-2i), dipeptidyl peptidase-4 inhibitors (DPP-4i), and glucagon-like peptide-1 agonists (GLP-1a). Design: We focused on cohort studies that provided data on cancer incidence and sample size in the NMA. Using NetMetaXL ® , the random effects model with informative priors was used in the NMA to estimate the pooled odds ratio (OR) with 95% credible intervals (CrI). Results: The systematic review included 62 studies, of which 22 met the inclusion criteria for the NMA. SGLT-2i users had lower overall cancer risk compared to sulfonylureas (OR: 0.54; 95% CrI: 0.40–0.74, low certainty), GLP-1a (OR: 0.70; 95% CrI: 0.53–0.92, low certainty), and DPP-4i users (OR: 0.72; 95% CrI: 0.57–0.92, very low certainty). DPP-4i users also had a lower cancer risk than sulfonylureas users (OR: 0.76; 95% CrI: 0.60–0.96, low certainty). No other statistically significant ORs were found in other direct comparisons. Conclusion: SGLT-2i users have a lower risk of developing cancers than sulfonylureas, GLP-1a, and DPP-4i users. These results may improve patient safety by guiding future clinical practice and medication choices. Future studies should investigate the mechanisms behind these observed associations. Trial registration: This NMA was registered in PROSPERO (CRD42023469941).