A comparison of different methods to handle missing data in the context of propensity score analysis

Type 2 diabetes, metabolic health, and the development of frozen shoulder: a cohort study in UK electronic health records

OBJECTIVE

Estimate the effect of type 2 diabetes on the development of frozen shoulder and investigate whether the effect is mediated by other metabolic factors.

METHODS

Primary care medical record-based cohort study containing 43,977 people newly diagnosed with type 2 diabetes and 43,977 without diabetes. Variables were identified using established Read codes. A weighting approach with Cox regression was used to decompose the total effect into the direct effect and indirect effect, mediated by metabolic health (which was defined as the number of metabolic factors developed during follow-up). Estimates were expressed as hazard ratios (HR). Confounders were identified using a DAG. Sensitivity to unmeasured confounding, extreme weights, and missing data were tested.

RESULTS

The total effect of type 2 diabetes on the development of frozen shoulder was HR = 4.38 (95% CI: 3.70-5.21), the natural indirect effect (mediated through metabolic health) was HR = 0.98 (95% CI: 0.93-1.03) and the natural direct effect was HR = 4.46 (95% CI: 3.68-5.41). Results were robust to unmeasured confounding, extreme weights, and missing data.

CONCLUSIONS

This study suggests that type 2 diabetes may be a cause of frozen shoulder but does not support the hypothesis that the effect is mediated by metabolic health. Clinicians should remain alert that shoulder pain in people with diabetes could be indicative of a frozen shoulder. This study should raise awareness that, despite often being overlooked, musculoskeletal conditions can be complications of diabetes and should be considered during clinical conversations with patients.

ISAC PROTOCOL REGISTRATION NUMBER

19_219R.

Comparative Effectiveness of Glucagon-Like Peptide-1 Receptor Agonists and Sodium/Glucose Cotransporter 2 Inhibitors in Preventing Chronic Kidney Failure and Mortality in Patients With Type 2 Diabetes and CKD

RATIONALE & OBJECTIVE

Both glucagon-like peptide-1 receptor agonists (GLP-1RAs) and sodium/glucose cotransporter 2 (SGLT2) inhibitors improve cardiovascular, kidney, and survival outcomes in patients with type 2 diabetes; however, the comparative effectiveness of these drugs in a real-world setting remains unclear.

STUDY DESIGN

Retrospective cohort study.

SETTING & PARTICIPANTS

79,047 patients with type 2 diabetes and an eGFR <60 mL/min/1.73 m2 between 2016 and 2021 from the Taiwan's national health database.

EXPOSURE

Treatment with GLP1RA or treatment with SGLT2i.

OUTCOME

Initiation of kidney replacement therapy (KRT) and all-cause mortality.

ANALYTIC APPROACH

Propensity score matching was performed to balance baseline characteristics between the groups. Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for each outcome, using an intention-to-treat approach.

RESULTS

14,182 (7,091 initiating GLP1RA and 7,091 initiating SGLT2i) from the original cohort of 79,047 individuals were included in the propensity score-matched analysis. With a median follow-up duration of 2.5 years, people initiating GLP1RA had a higher risk of requiring KRT compared to those initiating SGLT2i (HR: 1.39, 95% CI: 1.19-1.63). Although tests of interaction were not statistically significant, stratified analyses suggested possibly greater differences between the two drugs among patients with eGFR <45 mL/min/1.73 m2 or urine albumin-to-creatinine ratio >300 mg/g. Overall mortality did not differ between treatment groups.

LIMITATIONS

Nonrandomized treatment selection.

CONCLUSIONS

Patients receiving SGLT2i demonstrated lower rates of progression to KRT compared to those receiving GLP1RA. These findings may inform the choice of these therapies in the setting of chronic kidney disease and type 2 diabetes.

Outcomes for people experiencing homelessness with COVID-19 presenting to emergency departments in Canada, compared with housed patients

BACKGROUND

Whether people experiencing homelessness (PEH) have different COVID-19 outcomes than housed patients in Canada remains unclear. We sought to ascertain whether rates of in-hospital mortality, hospital admission, critical care admission, and mechanical ventilation differed between PEH and housed people with symptomatic SARS-CoV-2 infection.

METHODS

We conducted a propensity score-matched cohort study to compare the outcomes of PEH and housed patients presenting to emergency departments for acute symptomatic COVID-19. We used data from the Canadian COVID-19 Emergency Department Rapid Response Network (CCEDRRN) registry. Covariates in our propensity score model included age, sex, comorbidities, substance use, vaccination status, previous do-not-resuscitate documentation, hospital type, province and calendar quarter of presentation to the emergency department, symptom duration, and severity of illness on presentation.

RESULTS

We found no difference in mortality for PEH (3%) compared with a propensity score-matched cohort of housed patients (3%) (odds ratio [OR] 0.87, 95% confidence interval [CI] 0.43-1.74). We also found no difference in admission rates for PEH (44%) versus housed patients (45%). There was a reduced rate of critical care admission for PEH compared with housed patients (OR 0.66, 95% CI 0.44-1.00), and a trend toward decreased use of mechanical ventilation for PEH versus housed patients, which was not significant (OR 0.60, 95% CI 0.35-1.02).

INTERPRETATION

We found no difference in mortality for PEH with COVID-19 compared with those who were housed. A signal for reduced critical care admission among PEH may reflect differential treatment unrelated to clinical characteristics that we matched for. Future research on resource allocation during pandemics could shed light on potential inequities for vulnerable populations and how best to address them.

A comprehensive analysis of the prognostic value, expression characteristics and immune correlation of MKI67 in cancers

Background

nuclear-associated antigen Ki67 (Ki67) emerges as a clinically practical biomarker for proliferation assessment among many cancer types. However, the definite prognostic value of Ki67 against a specific cancer type has remained vague. This study aims to perform a comprehensive pan-cancer analysis of the prognosis value of Ki67 across various cancer types.

Methods

This study explored the expression, prognostic value, and tumor-infiltrating immune of MKI67 in the TCGA database by pan-cancer, and then performed immunohistochemical, correlation analysis and prognostic analysis using 10028 patients of the top 10 cancer patients in China we collected. The correlation between MKI67 expression and survival outcome, clinical features, MSI, TMB, and tumor-infiltrating immune cells by TCGA database, xCell, and TIMER algorithms.

Results

MKI67 expression was significantly upregulated across varied cancer types verified by datasets. We found MKI67 expression was significantly associated with poor prognosis in LUADLUSC, LIHC, and BRCA patients, but good prognosis in COADREAD and READ patients via Kaplan-Meier survival analysis using 10028 patients collected. These results of our validation were generally consistent with TCGA database except BRCA, COADREAD and READ. Meanwhile, upregulation of MKI67 elevates the degree of immune infiltration of several immune cell subtypes, such as functional T cells, CD4+ T cells, and CD8+ T cells, as well as, MKI67 was related to Cell cycle, Oocyte meiosis, p53 and other pathways.

Conclusion

Our comprehensive analysis may supply useful guidance on MKI67 applicability across various cancer types. These observed results contribute to the promise of MKI67 in a realistic clinical setting and improve the outcomes of cancer patients.

The relationship between estimated glucose disposal rate and cognitive function in older individuals

The estimated glucose disposal rate (eGDR) serves as a novel indicator of insulin resistance, which has been shown to correlate with cardiovascular disease risk; however, its relationship with cognitive function remains unclear.This article describes a cross-sectional study design based on data from the 2011-2014 National Health and Nutrition Examination Survey (NHANES). The Weighted logistic regression and the restricted cubic spline were employed to examine the relationship between eGDR and cognitive ability.The subjects were divided into two categories: the normal group and the cognitive function decline (CFD) group, based on their cognitive scores. There were significant differences in eGDR levels between the two groups(P = 0.001).After adjusting for relevant covariates, notable differences were found between eGDR and cognitive function when eGDR was expressed in both continuous and categorical data forms (P < 0.05). The stability of these findings was further confirmed through sensitivity analyses.This difference persisted in subgroups, including women, individuals with education beyond high school, moderate drinkers, and those who had not been diagnosed with stroke (P < 0.05). A restricted cubic spline revealed a non-linear relationship with an inflection point between the two (P-for-non-linear < 0.05, P-overall < 0.001). This study contributes to the understanding of the relationship between eGDR and cognitive performance by identifying a potential non-linear association.

Real-World Persistence and Effectiveness of Upadacitinib versus Other Janus Kinase Inhibitors and Tumor Necrosis Factor Inhibitors in Australian Patients with Rheumatoid Arthritis

INTRODUCTION

This study sought to describe treatment patterns, persistence, and effectiveness of upadacitinib (UPA) alone and compared to other Janus kinase inhibitors (JAKis) or tumor necrosis factor inhibitors (TNFis) in patients with rheumatoid arthritis (RA).

METHODS

This retrospective, non-interventional study used the OPAL dataset, derived from electronic medical records. Patients initiated UPA (N = 2624), other JAKis (baricitinib and tofacitinib [N = 925]), or TNFis (adalimumab, etanercept, certolizumab, golimumab, infliximab [N = 3540]) between May 2020 and March 2023. Median persistence (Kaplan-Meier) and effectiveness (Disease Activity Score 28-joint C-reactive protein, three variables [DAS28CRP{3}]) were evaluated for UPA-treated patients and in three propensity score-matched cohorts: UPA monotherapy versus combination therapy, UPA versus other JAKis, and UPA versus TNFis.

RESULTS

In patients prescribed UPA, 41.3% were ≥ 65 years old, 33.8% were prescribed as first-line advanced therapy, and 27.2% were prescribed monotherapy. Persistence on UPA was 26.6 months (95% confidence intervals: 24.4, 29.9) and longest in earlier lines of therapy. The DAS28CRP(3) remission rate was 73% at 3 months, with improvements observed across lines of therapy. UPA monotherapy and combination therapy had similar persistence (27.8 [23.5, 33.4] versus 30.4 months [22.1, 35.3], p = 0.84) and effectiveness. UPA showed longer persistence than other JAKis (28.8 [25.6, 32.4] versus 17.2 months [14.9, 19.8], p < 0.001) and TNFis (26.6 [24.9, 30.8] versus 13.3 months [11.5, 14.5], p < 0.001). DAS28CRP(3) remission rates were greater at 3 months for UPA than other JAKis (75.0% versus 61.5%) and TNFis (72.7% versus 59.5%). In unmatched subgroups, compared to cycling between TNFis, switching to UPA from other JAKis or TNFis resulted in longer persistence (JAKi-to-UPA: 25.3 [16.1, not reached]; TNFi-to-UPA: 27.8 [23.2, 35.4]; TNFi-to-TNFi: 9.6 [8.4, 10.7]) and greater DAS28CRP(3) remission rates over 9 months.

CONCLUSIONS

Overall, the breadth and depth of data from this large real-world dataset continue to support a favorable clinical profile of UPA for the treatment of RA and may inform treatment choices in everyday clinical practice.

Prognostic Nutritional Index as a Potential Biomarker for the Risk of Lower Extremity Deep Venous Thrombosis: A Large Retrospective Study

BACKGROUND

Prognostic nutritional index (PNI) has recently been identified as a novel marker of nutritional status. However, existing evidences on the association between PNI and the risk of lower extremity deep venous thrombosis (LEDVT) are limited and conflicting.

OBJECTIVE

To determine the association between PNI and the risk of LEDVT, and further evaluate its diagnostic value.

METHODS

Over a 10-year period (2012-2022), a total of 12790 patients who underwent compression ultrasonography examinations were consecutively included, and 1519 (11.9%) LEDVT events occurred. Multivariate logistic regression analysis was used to investigate the association, and receiver operating characteristic (ROC) curve was constructed to evaluate its diagnostic performance.

RESULTS

After full adjustment, patients in third quartile (odds ratio [OR] = 1.486, 95% confidence interval [CI]: 1.205-1.832), second quartile (OR = 2.436, 95% CI: 1.993-2.978) and first quartile (OR = 3.422, 95% CI: 2.791-4.195) of PNI were at higher risk of LEDVT compared with those in fourth quartile, and the test for trend was significant. Consistently, each unit decrease in PNI was associated with a 6.0 % (95% CI: 1.052-1.069) increased risk of LEDVT. Moreover, adding PNI to a base model improved the area under the curve (AUC) from 0.721 (95% CI: 0.709-0.734) to 0.746 (95% CI: 0.734-0.758).

CONCLUSION

PNI is inversely associated with the risk of LEDVT, and provides significant incremental diagnostic value for the identification of LEDVT events. These findings suggest that PNI may be a potential biomarker to help clinicians identify patients at risk of thrombosis and make clinical decisions timely.

Estimating the effect of pre-exposure prophylaxis in Black men who have sex with men

BACKGROUND

Black men who have sex with men (MSM) are disproportionately burdened by the HIV epidemic in the USA. The effectiveness of pre-exposure prophylaxis (PrEP) in preventing HIV infection has been demonstrated through randomized placebo-controlled clinical trials in several populations. Importantly, no such trial has been conducted exclusively among Black MSM in the USA, and it would be unethical and infeasible to do so now.

METHODS

To estimate the causal effects of PrEP access, initiation, and adherence on HIV risk, we utilized causal inference methods to combine data from two non-randomized studies that exclusively enrolled Black MSM.

RESULTS

The estimated relative risks of HIV were: (i) 0.52 (95% confidence interval: 0.21, 1.22) for individuals with versus without PrEP access, (ii) 0.48 (0.12, 0.89) for individuals who initiated PrEP but were not adherent versus those who did not initiate, and (iii) 0.23 (0.02, 0.80) for individuals who were adherent to PrEP versus those who did not initiate.

CONCLUSION

Beyond addressing the knowledge gap around the effect of PrEP in Black MSM in the USA, which may have ramifications for public health, we have provided a framework to combine data from multiple non-randomized studies to estimate causal effects, which has broad utility.

Clinical Outcomes of 5000 IU Heparin Versus Activated Clotting Time-Guided Heparinization During Noncardiac Arterial Procedures: A Propensity Score Matched Analysis

PURPOSE

Previous studies have shown that activated clotting time (ACT)-guided heparinization leads to better anticoagulation levels during noncardiac arterial procedures (NCAP) than a standardized bolus of 5000 IU. Better anticoagulation should potentially result in lower incidence of thrombo-embolic complications (TEC). Comparative investigations on clinical outcomes of these heparinization strategies are scarce. This study investigated clinical outcomes of ACT-guided heparinization with a starting dose of 100 IU/kg in comparison with a single standardized bolus of 5000 IU heparin during NCAP.

MATERIALS AND METHODS

Analysis from a prospectively collected database of patients undergoing NCAP in 2 vascular centers was performed. Patients receiving ACT-guided heparinization were matched 1:1 with patients receiving 5000 IU heparin using propensity score matching (PSM). Primary outcomes were TEC, bleeding complications, and mortality within 30 days of procedure or during the same admission.

RESULTS

A total of 759 patients (5000 IU heparin: 213 patients, ACT-guided heparinization: 546 patients) were included. Propensity score matching resulted in 209 patients in each treatment group. After PSM, the groups were comparable, with the exception of a higher prevalence of peripheral arterial disease in the ACT-guided heparinization group (103 patients, 49% vs 82 patients, 39%, p=0.039). The target ACT (>200 seconds) was reached in 198 patients (95%) of the ACT-guided group versus 71 patients (34%) of the 5000 IU group (p<0.001), indicating successful execution of the ACT-guided protocol. Incidence of TEC (13 patients, 6.2% vs 10 patients, 4.8%, p=0.52), mortality (3 patients, 1.4% vs 0 patients, p=0.25), and bleeding complications (32 patients, 15% vs 25 patients, 12%, p=0.32) did not differ between patients receiving ACT-guided heparinization and 5000 IU heparin. Protamine was administered in 118 patients (57%) in the ACT group versus 11 patients (5.3%) in the 5000 IU group (p<0.001), but did not influence incidence of TEC (17 patients, 5.9% vs 6 patients, 4.7%, p=0.61) or bleeding complications (34 patients, 12% vs 22 patients, 17%, p=0.14).

CONCLUSION

No difference in TEC, bleeding complications, or mortality was found between ACT-guided heparinization and a single bolus of 5000 IU heparin during NCAP.

CLINICAL IMPACT

Previous studies have shown that activated clotting time (ACT)-guided heparinization leads to better anticoagulation levels during non-cardiac arterial procedures (NCAP) then a standardized bolus of 5000 IU. Comparative investigations on clinical outcomes are scarce. This study focussed on clinical outcomes of both protocols in NCAP in a propensity score matched cohort. Thrombo-embolic complications (TEC), bleeding complications and mortality within 30 days after NCAP or during the same admission were comparable between groups. Future studies should focus on optimizing ACT-guided protocols, specifically in patients with a high risk of TEC and bleeding complications.

Balance diagnostics in propensity score analysis following multiple imputation: A new method

The combination of propensity score analysis and multiple imputation has been prominent in epidemiological research in recent years. However, studies on the evaluation of balance in this combination are limited. In this paper, we propose a new method for assessing balance in propensity score analysis following multiple imputation. A simulation study was conducted to evaluate the performance of balance assessment methods (Leyrat's, Leite's, and new method). Simulated scenarios varied regarding the presence of missing data in the control or treatment and control group, and the imputation model with/without outcome. Leyrat's method was more biased in all the studied scenarios. Leite's method and the combine method yielded balanced results with lower mean absolute difference, regardless of whether the outcome was included in the imputation model or not. Leyrat's method had a higher false positive ratio and Leite's and combine method had higher specificity and accuracy, especially when the outcome was not included in the imputation model. According to simulation results, most of time, Leyrat's method and Leite's method contradict with each other on appraising the balance. This discrepancy can be solved using new combine method.

Association of oxaliplatin-containing adjuvant duration with post-treatment fall-related injury and fracture in patients with stage III colon cancer: a population-based retrospective cohort study

PURPOSE

Oxaliplatin-containing adjuvant chemotherapy yields a significant survival benefit in stage III colon cancer and is the standard of care. Simultaneously, it causes dose-dependent peripheral neuropathy that may increase the risk of fall-related injury (FRI) such as fracture and laceration. Because these events carry significant morbidity and the global burden of colon cancer is on the rise, we examined the association between treatment with a full versus shortened course of adjuvant chemotherapy and post-treatment FRI and fracture.

METHODS

In this overlap propensity score weighted, retrospective cohort study, we included patients aged ≥ 18 years with resected stage III colon cancer diagnosed 2007-2019 and treated with oxaliplatin-containing adjuvant chemotherapy (oxaliplatin plus a fluoropyrimidine; capecitabine [CAPOX] or 5-fluorouracil and leucovorin [FOLFOX]). Propensity score methods facilitate the separation of design from analysis and comparison of baseline characteristics across the weighted groups. Treatment groups were defined as 50% (4 cycles CAPOX/6 cycles FOLFOX) and > 85% (7-8 cycles CAPOX/11-12 cycles FOLFOX) of a maximal course of adjuvant chemotherapy to approximate the treatment durations received in the IDEA collaboration. The main outcomes were time to any FRI and time to fracture. We determined the subdistribution hazard ratios (sHR) estimating the association between FRI/fracture and treatment group, accounting for the competing risk of death.

RESULTS

We included 3,461 patients; 473 (13.7%) received 50% and 2,988 (86.3%) received > 85% of a maximal course of adjuvant therapy. For post-treatment FRI, median follow-up was 4.6 years and total follow-up was 17,968 person-years. There were 508 FRI, 301 fractures, and 692 deaths. Treatment with > 85% of a maximal course of therapy conferred a sHR of 0.84 (95% CI 0.62-1.13) for post-treatment FRI and a sHR of 0.72 (95% CI 0.49-1.06) for post-treatment fracture.

CONCLUSION

For patients with stage III colon cancer undergoing treatment with oxaliplatin-containing adjuvant chemotherapy, any potential neuropathy associated with longer durations of treatment was not found to result in greater rates of FRI and fracture. Within the limits of this retrospective study, our findings suggest concern about FRI, while mechanistically plausible, ought not to determine treatment duration.

Ten things to remember about propensity scores

Propensity score methods are popular to control for confounding in observational biomedical studies of risk factors or medical treatments. This paper focused on aspects of propensity score methods that often remain undiscussed, including unmeasured confounding, missing data, variable selection, statistical efficiency, estimands, the positivity assumption, and predictive performance of the propensity score model.

A Principled Approach to Characterize and Analyze Partially Observed Confounder Data from Electronic Health Records

Objective

Partially observed confounder data pose challenges to the statistical analysis of electronic health records (EHR) and systematic assessments of potentially underlying missingness mechanisms are lacking. We aimed to provide a principled approach to empirically characterize missing data processes and investigate performance of analytic methods.

Methods

Three empirical sub-cohorts of diabetic SGLT2 or DPP4-inhibitor initiators with complete information on HbA1c, BMI and smoking as confounders of interest (COI) formed the basis of data simulation under a plasmode framework. A true null treatment effect, including the COI in the outcome generation model, and four missingness mechanisms for the COI were simulated: completely at random (MCAR), at random (MAR), and two not at random (MNAR) mechanisms, where missingness was dependent on an unmeasured confounder and on the value of the COI itself. We evaluated the ability of three groups of diagnostics to differentiate between mechanisms: 1)-differences in characteristics between patients with or without the observed COI (using averaged standardized mean differences [ASMD]), 2)-predictive ability of the missingness indicator based on observed covariates, and 3)-association of the missingness indicator with the outcome. We then compared analytic methods including "complete case", inverse probability weighting, single and multiple imputation in their ability to recover true treatment effects.

Results

The diagnostics successfully identified characteristic patterns of simulated missingness mechanisms. For MAR, but not MCAR, the patient characteristics showed substantial differences (median ASMD 0.20 vs 0.05) and consequently, discrimination of the prediction models for missingness was also higher (0.59 vs 0.50). For MNAR, but not MAR or MCAR, missingness was significantly associated with the outcome even in models adjusting for other observed covariates. Comparing analytic methods, multiple imputation using a random forest algorithm resulted in the lowest root-mean-squared-error.

Conclusion

Principled diagnostics provided reliable insights into missingness mechanisms. When assumptions allow, multiple imputation with nonparametric models could help reduce bias.

Propensity Score Weighting with Missing Data on Covariates and Clustered Data Structure

Propensity score (PS) analyses are increasingly popular in behavioral sciences. Two issues often add complexities to PS analyses, including missing data in observed covariates and clustered data structure. In previous research, methods for conducting PS analyses with considering either issue alone were examined. In practice, the two issues often co-occur; but the performance of methods for PS analyses in the presence of both issues has not been evaluated previously. In this study, we consider PS weighting analysis when data are clustered and observed covariates have missing values. A simulation study is conducted to evaluate the performance of different missing data handling methods (complete-case, single-level imputation, or multilevel imputation) combined with different multilevel PS weighting methods (fixed- or random-effects PS models, inverse-propensity-weighting or the clustered weighting, weighted single-level or multilevel outcome models). The results suggest that the bias in average treatment effect estimation can be reduced, by better accounting for clustering in both the missing data handling stage (such as with the multilevel imputation) and the PS analysis stage (such as with the fixed-effects PS model, clustered weighting, and weighted multilevel outcome model). A real-data example is provided for illustration.

Influence of initial misdiagnosis on mortality in patients with bacteraemia: propensity score matching and propensity score weighting analyses

BACKGROUND

The diagnostic process is a key element of medicine but it is complex and prone to errors. Infectious diseases are one of the three categories of diseases in which diagnostic errors can be most harmful to patients. In this study we aimed to estimate the effect of initial misdiagnosis of the source of infection in patients with bacteraemia on 14 day mortality using propensity score methods to adjust for confounding.

METHODS

Data from a previously described longitudinal cohort of patients diagnosed with monobacterial bloodstream infection (BSI) at the Leiden University Medical Centre (LUMC) between 2013 and 2015 were used. Propensity score matching and inversed probability of treatment weighting (IPTW) were applied to correct for confounding. The average treatment effect on the treated (ATT), which in this study was the average effect of initial misdiagnosis on the misdiagnosed (AEMM), was estimated. Methodological issues that were encountered when applying propensity score methods were addressed by performing additional sensitivity analyses. Sensitivity analyses consisted of varying caliper in propensity score matching and using different truncated weights in inversed probability of treatment weighting.

RESULTS

Data of 887 patients were included in the study. Propensity scores ranged between 0.015 and 0.999 and 80 patients (9.9%) had a propensity score > 0.95. In the matched analyses, 35 of the 171 misdiagnosed patients died within 14 days (20.5%), versus 10 of the 171 correctly diagnosed patients (5.8%), yielding a difference of 14.6% (7.6%; 21.6%). In the total group of patients, the observed percentage of patients with an incorrect initial diagnosis that died within 14 days was 19.8% while propensity score reweighting estimated that their probability of dying would have been 6.5%, if they had been correctly diagnosed (difference 13.3% (95% CI 6.9%;19.6%)). After adjustment for all variables that showed disbalance in the propensity score a difference of 13.7% (7.4%; 19.9%) was estimated. Sensitivity analyses yielded similar results. However, performing weighted analyses without truncation yielded unstable results.

CONCLUSION

Thus, we observed a substantial increase of 14 day mortality in initially misdiagnosed patients. Furthermore, several patients received propensity scores extremely close to one and were almost sure to be initially misdiagnosed.

A propensity score matched analysis of liver transplantation outcomes in the setting of preservation solution shortage

The recent shortage of the University of Wisconsin (UW) solution prompted increased utilization of histidine-tryptophan-ketoglutarate (HTK) solution for liver graft preservation. This contemporary study analyzed deceased donor liver transplant outcomes following preservation with HTK vs UW. Patients receiving deceased donor liver transplantations between January 1, 2019, and June 30, 2022, were retrospectively identified utilizing the Organ Procurement and Transplant Network database, stratified by preservation with HTK vs UW, and a propensity score matching analysis was performed. Outcomes assessed included rates of primary nonfunction, graft survival, and patient survival. There were 4447 patients in each cohort. Primary nonfunction occurred in 60 (1.35%) patients in the HTK group vs 25 (0.54%) in the UW group (P < .001). HTK was associated with lower 90-day graft survival (94.39% vs 96.09%; P < .001) and 90-day patient survival (95.97% vs 97.38%; P = .001). Unmatched donation after cardiac death-specific analysis of HTK vs UW demonstrated respective rates of primary nonfunction of 1.63% vs 0.82% (P = .20), 90-day graft survival of 92.50% vs 95.29% (P = .069), and 90-day patient survival of 93.90% vs 96.35% (P = .077). These results suggest that HTK may not be an equivalent preservation solution for deceased donor liver transplantation.

Cost-effective prognostic evaluation of breast cancer: using a STAR nomogram model based on routine blood tests

Background

Breast cancer (BC) is the most common and prominent deadly disease among women. Predicting BC survival mainly relies on TNM staging, molecular profiling and imaging, hampered by subjectivity and expenses. This study aimed to establish an economical and reliable model using the most common preoperative routine blood tests (RT) data for survival and surveillance strategy management.

Methods

We examined 2863 BC patients, dividing them into training and validation cohorts (7:3). We collected demographic features, pathomics characteristics and preoperative 24-item RT data. BC risk factors were identified through Cox regression, and a predictive nomogram was established. Its performance was assessed using C-index, area under curves (AUC), calibration curve and decision curve analysis. Kaplan-Meier curves stratified patients into different risk groups. We further compared the STAR model (utilizing HE and RT methodologies) with alternative nomograms grounded in molecular profiling (employing second-generation short-read sequencing methodologies) and imaging (utilizing PET-CT methodologies).

Results

The STAR nomogram, incorporating subtype, TNM stage, age and preoperative RT data (LYM, LYM%, EOSO%, RDW-SD, P-LCR), achieved a C-index of 0.828 in the training cohort and impressive AUCs (0.847, 0.823 and 0.780) for 3-, 5- and 7-year OS rates, outperforming other nomograms. The validation cohort showed similar impressive results. The nomogram calculates a patient's total score by assigning values to each risk factor, higher scores indicating a poor prognosis. STAR promises potential cost savings by enabling less intensive surveillance in around 90% of BC patients. Compared to nomograms based on molecular profiling and imaging, STAR presents a more cost-effective, with potential savings of approximately $700-800 per breast cancer patient.

Conclusion

Combining appropriate RT parameters, STAR nomogram could help in the detection of patient anemia, coagulation function, inflammation and immune status. Practical implementation of the STAR nomogram in a clinical setting is feasible, and its potential clinical impact lies in its ability to provide an early, economical and reliable tool for survival prediction and surveillance strategy management. However, our model still has limitations and requires external data validation. In subsequent studies, we plan to mitigate the potential impact on model robustness by further updating and adjusting the data and model.

Estimating the Effect of PrEP in Black Men Who Have Sex with Men: A Framework to Utilize Data from Multiple Non-Randomized Studies to Estimate Causal Effects

Black men who have sex with men (MSM) are disproportionately burdened by the HIV epidemic in the US. The effectiveness of pre-exposure prophylaxis (PrEP) in preventing HIV infection has been demonstrated through randomized placebo-controlled clinical trials in several populations. Importantly, no such trial has been conducted exclusively among Black MSM in the US, and it would be unethical and infeasible to do so now. To estimate the causal effects of PrEP access, initiation, and adherence on HIV risk, we utilized causal inference methods to combine data from two non-randomized studies that exclusively enrolled Black MSM. The estimated relative risks of HIV were: (i) 0.52 (95% confidence interval: 0.21, 1.22) for individuals with versus without PrEP access, (ii) 0.48 (0.12, 0.89) for individuals who initiated PrEP but were not adherent versus those who did not initiate, and (iii) 0.23 (0.02, 0.80) for individuals who were adherent to PrEP versus those who did not initiate. Beyond addressing the knowledge gap around the effect of PrEP in Black MSM in the US, which may have ramifications for public health, we have provided a framework to combine data from multiple non-randomized studies to estimate causal effects, which has broad utility.

Mental Health Conditions in Partners and Adult Children of Stroke Survivors

Importance

Family caregiving after critical illness has been associated with several adverse health outcomes, including various aspects of mental health, but research focusing specifically on family members of stroke survivors is limited.

Objectives

To examine the associations of stroke in a partner or parent with the risk of depression, substance use disorders, anxiety disorders, and self-harm or suicide.

Design, Setting, and Participants

This nationwide, population-based cohort study used data from Danish nationwide administrative and clinical registries (2004-2021). Participants included partners and adult children of survivors of stroke. Data analysis was performed from March to December 2023.

Exposure

Having a partner or parent who survived stroke.

Main Outcomes and Measures

The Aalen-Johansen estimator was used to compute propensity score-weighted 3-year absolute risks, risk differences, and risk ratios for depression, substance use disorders, anxiety disorders, and self-harm or suicide among partners or children of survivors of stroke compared with partners or children of survivors of myocardial infarction (MI) and matched individuals from the general population.

Results

The study included a total of 1 923 732 individuals: 70 917 partners of stroke survivors (median [IQR] age, 68 [59-76] years; 46 369 women [65%]), 70 664 partners of MI survivors (median [IQR] age, 65 [55-73] years; 51 849 women [73%]), 354 570 partners of individuals from the general population (median [IQR] age, 68 [59-76] years; 231 833 women [65%]), 207 386 adult children of stroke survivors (median [IQR] age, 45 [36-52] years; 99 382 women [48%]), 183 309 adult children of MI survivors (median [IQR] age, 42 [33-49] years; 88 078 women [48%]), and 1 036 886 adult children of individuals from the general population (median [IQR] age, 45 [36-52] years; 496 875 women [48%]). Baseline characteristics were well balanced across cohorts after propensity score weighting. Among partners of stroke survivors, the 3-year absolute risk was 1.0% for depression, 0.7% for substance use disorders, 0.3% for anxiety disorders, and 0.04% for self-harm or suicide. Risk ratio point estimates for the assessed outcomes ranged from 1.14 to 1.42 compared with the general population and from 1.04 to 1.09 compared with partners of MI survivors. The elevated risk of depression in partners of stroke survivors was more pronounced after severe or moderate stroke than after mild stroke. Among adult children of stroke survivors, the 3-year absolute risk was 0.6% for depression, 0.6% for substance use disorders, 0.2% for anxiety disorders, and 0.05% for self-harm or suicide. Both absolute risks and risk ratios for adult children of stroke survivors were smaller than those reported in the partner analyses.

Conclusions and Relevance

In this cohort study of partners and adult children of stroke survivors, risks of several mental health conditions and self-harm or suicide were moderately higher compared with the general population and, to a lesser extent, partners and adult children of MI survivors. These findings highlight the potential consequences of stroke among family members, particularly partners, and its findings may possibly serve as a quantitative foundation for the development of future stroke rehabilitation services.

Effectiveness of residential treatment services for veterans with substance use disorders: A propensity score matching evaluation

BACKGROUND

Prior reviews of substance use disorder (SUD) treatment have found mixed support for residential level of care but are limited by methodology problems and the ethical concerns of randomizing patients with severe SUD to lower levels of care.

METHODS

The present study is the first to use a large archival SUD residential sample with a matched comparison group and one-year follow-up period to examine the benefits of residential treatment provided to adults clinically assessed as warranting SUD residential care. We used propensity score matching in our sample (N = 6177) of veterans with a SUD who were screened and accepted for Veterans Affairs (VA) SUD residential treatment between January 1st, 2019 and June 30th, 2019.

RESULTS

We found evidence that VA SUD residential treatment saves veteran lives with an average 66% all-cause mortality risk reduction during the study period (b = -1.09, exp(b) = 0.34, p <0.001). Medium-to-large residential pre- to post-treatment self-reported mental health and SUD symptom improvements (|SMDrobust| = 0.54-0.93) were sustained by one-year post-screening. These residential treatment improvements were significantly larger than estimated counterfactual outcomes across self-reported SUD and stress disorder symptoms at one-year post-screening (ps <0.001). We found mixed behavioral, service utilization, and other self-reported mental health outcomes.

CONCLUSIONS

We conclude that VA SUD residential treatment is an effective level of care for veterans warranting residential care particularly for SUD symptom improvements and reductions in mortality risk.

Contribution of socio-demographic and clinical characteristics to predict initial referrals to psychosocial interventions in patients with serious mental illness

AIMS

Psychosocial rehabilitation (PSR) is at the core of psychiatric recovery. There is a paucity of evidence regarding how the needs and characteristics of patients guide clinical decisions to refer to PSR interventions. Here, we used explainable machine learning methods to determine how socio-demographic and clinical characteristics contribute to initial referrals to PSR interventions in patients with serious mental illness.

METHODS

Data were extracted from the French network of rehabilitation centres, REHABase, collected between years 2016 and 2022 and analysed between February and September 2022. Participants presented with serious mental illnesses, including schizophrenia spectrum disorders, bipolar disorders, autism spectrum disorders, depressive disorders, anxiety disorders and personality disorders. Information from 37 socio-demographic and clinical variables was extracted at baseline and used as potential predictors. Several machine learning models were tested to predict initial referrals to four PSR interventions: cognitive behavioural therapy (CBT), cognitive remediation (CR), psychoeducation (PE) and vocational training (VT). Explanatory power of predictors was determined using the artificial intelligence-based SHAP (SHapley Additive exPlanations) method from the best performing algorithm.

RESULTS

Data from a total of 1146 patients were included (mean age, 33.2 years [range, 16-72 years]; 366 [39.2%] women). A random forest algorithm demonstrated the best predictive performance, with a moderate or average predictive accuracy [micro-averaged area under the receiver operating curve from 'external' cross-validation: 0.672]. SHAP dependence plots demonstrated insightful associations between socio-demographic and clinical predictors and referrals to PSR programmes. For instance, patients with psychotic disorders were more likely to be referred to PE and CR, while those with non-psychotic disorders were more likely to be referred to CBT and VT. Likewise, patients with social dysfunctions and lack of educational attainment were more likely to be referred to CR and VT, while those with better functioning and education were more likely to be referred to CBT and PE.

CONCLUSIONS

A combination of socio-demographic and clinical features was not sufficient to accurately predict initial referrals to four PSR programmes among a French network of rehabilitation centres. Referrals to PSR interventions may also involve service- and clinician-level factors. Considering socio-demographic and clinical predictors revealed disparities in referrals with respect to diagnoses, current clinical and psychological issues, functioning and education.

Trends and Prescriber Variation in the Duration of Oxaliplatin-Containing Adjuvant Chemotherapy for Stage III Colon Cancer From 2007 to 2019: A Population-Based Retrospective Cohort Study

INTRODUCTION

The International Duration Evaluation of Adjuvant Therapy (IDEA) collaboration in 2017 established 3 months of adjuvant therapy as an alternative to 6 months of therapy for stage III colon cancer. We determined the association between the IDEA publication, changes in clinical practice, and prescriber variation.

PATIENTS AND METHODS

Using linked databases, we identified Ontarians aged ≥18 years at diagnosis of stage III colon cancer between 2007 and 2019 who received oxaliplatin-containing adjuvant therapy. The outcome was duration of therapy, categorized as ≤25%, >25% to ≤50%, >50% to ≤75%, and >75% of a 6-month course of therapy to approximate treatment durations in the IDEA collaboration. We examined trends in duration over time using an interrupted time series regression model. We analyzed treatment duration after accounting for patient and prescriber characteristics, using multivariable mixed effects logistic regression models to quantify between-prescriber variation.

RESULTS

We included 4695 patients with stage III colon cancer who received oxaliplatin-containing adjuvant chemotherapy, of whom 77.5% initiated treatment pre-IDEA and 22.5% initiated treatment post-IDEA. Post-IDEA, there was a 16.4% (95% CI, 12.5%-20.3%) absolute increase in the proportion of patients treated with ≤50% of a maximal course of therapy. This trend was greatest among patients with low-risk tumors. Prescriber variation increased pre-IDEA to 15.6% post-IDEA (variance partition coefficient 5.4% pre-IDEA and 15.6% post-IDEA).

CONCLUSION

The publication of IDEA was associated with increases in short duration adjuvant therapy and prescriber-level practice variation for stage III colon cancer. Clinicians should be better supported to make consistent recommendations about adjuvant duration under conditions of uncertainty and trade-offs.

Making the Improbable Possible: Generalizing Models Designed for a Syndrome-Based, Heterogeneous Patient Landscape

Syndromic conditions, such as sepsis, are commonly encountered in the intensive care unit. Although these conditions are easy for clinicians to grasp, these conditions may limit the performance of machine-learning algorithms. Individual hospital practice patterns may limit external generalizability. Data missingness is another barrier to optimal algorithm performance and various strategies exist to mitigate this. Recent advances in data science, such as transfer learning, conformal prediction, and continual learning, may improve generalizability of machine-learning algorithms in critically ill patients. Randomized trials with these approaches are indicated to demonstrate improvements in patient-centered outcomes at this point.

The Association of Oxaliplatin-Containing Adjuvant Chemotherapy Duration with Overall and Cancer-Specific Mortality in Individuals with Stage III Colon Cancer: A Population-Based Retrospective Cohort Study

PURPOSE

Few studies have examined the relationship between duration of oxaliplatin-containing adjuvant chemotherapy for stage III colon cancer and mortality in routine practice. We examined the association between treatment with 50% versus >85% of a maximal course of adjuvant therapy (eight cycles of CAPOX, twelve cycles of FOLFOX) and mortality in stage III colon cancer.

METHODS

Using linked databases, we identified Ontarians aged ≥18 years at diagnosis of stage III colon cancer between 2007 and 2019. In the primary comparison, we compared patients who received 50% or >85% of a maximal course of adjuvant therapy; in a secondary comparison, we evaluated a dose effect across patients who received FOLFOX in one-cycle increments from six to ten cycles against >85% (more than ten cycles) of a maximal course of FOLFOX. The main outcomes were overall and cancer-specific mortality. Follow-up began 270 days after adjuvant treatment initiation and terminated at the first of the outcome of interest, loss of eligibility for Ontario's Health Insurance Program, or study end. Overlap propensity score weights accounted for baseline between-group differences. We determined the hazard ratio, estimating the association between mortality and treatment. Non-inferiority was concluded in the primary comparison for either outcome if the upper limit of the two-sided 95% CI was ≤1.11, which is the margin used in the International Duration Evaluation of Adjuvant Chemotherapy Collaboration.

RESULTS

We included 3546 patients in the analysis of overall mortality; 486 (13.7%) received 50% and 3060 (86.3%) received >85% of a maximal course of therapy. Median follow-up was 5.4 years, and total follow-up was 20,510 person-years. There were 833 deaths. Treatment with 50% of a maximal course of adjuvant therapy was associated with a hazard ratio of 1.13 (95% CI 0.88 to 1.47) for overall mortality and a subdistribution hazard ratio of 1.31 (95% CI 0.91 to 1.87) for cancer-specific mortality versus >85% of a maximal course of therapy. In the secondary comparison, there was a trend toward higher overall mortality in patients treated with shorter durations of therapy, though confidence intervals overlapped considerably.

CONCLUSION

We could not conclude that treatment with 50% of a maximal course is non-inferior to >85% of a maximal course of adjuvant therapy for mortality in stage III colon cancer. Clinicians and patients engaging in decision-making around treatment duration in this context should carefully consider the trade-off between treatment effectiveness and adverse effects of treatment.

Inverse Probability of Treatment Weighting and Confounder Missingness in Electronic Health Record-based Analyses: A Comparison of Approaches Using Plasmode Simulation

BACKGROUND

Electronic health record (EHR) data represent a critical resource for comparative effectiveness research, allowing investigators to study intervention effects in real-world settings with large patient samples. However, high levels of missingness in confounder variables is common, challenging the perceived validity of EHR-based investigations.

METHODS

We investigated performance of multiple imputation and propensity score (PS) calibration when conducting inverse probability of treatment weights (IPTW)-based comparative effectiveness research using EHR data with missingness in confounder variables and outcome misclassification. Our motivating example compared effectiveness of immunotherapy versus chemotherapy treatment of advanced bladder cancer with missingness in a key prognostic variable. We captured complexity in EHR data structures using a plasmode simulation approach to spike investigator-defined effects into resamples of a cohort of 4361 patients from a nationwide deidentified EHR-derived database. We characterized statistical properties of IPTW hazard ratio estimates when using multiple imputation or PS calibration missingness approaches.

RESULTS

Multiple imputation and PS calibration performed similarly, maintaining ≤0.05 absolute bias in the marginal hazard ratio even when ≥50% of subjects had missing at random or missing not at random confounder data. Multiple imputation required greater computational resources, taking nearly 40 times as long as PS calibration to complete. Outcome misclassification minimally increased bias of both methods.

CONCLUSION

Our results support multiple imputation and PS calibration approaches to missingness in missing completely at random or missing at random confounder variables in EHR-based IPTW comparative effectiveness analyses, even with missingness ≥50%. PS calibration represents a computationally efficient alternative to multiple imputation.

Association between systemic immune-inflammation index and risk of lower extremity deep venous thrombosis in hospitalized patients: a 10-year retrospective analysis

Background

The systemic immune-inflammation index (SII), as a novel inflammatory biomarker, has recently attracted attention in cardiovascular disease research. However, the relationship between SII and risk of lower extremity deep venous thrombosis (LEDVT) remains unclear to date. Thus, this study aimed to explore the association in a large sample over a 10-year period (2012-2022).

Methods

All hospitalized patients undergoing lower extremity compression ultrasonography (CUS) examination were consecutively screened by searching our hospital information system database. The receiver operating characteristic (ROC) curve analysis was used to identify the optimal cut-off value for high and low SII group. Multivariate logistic regression analyses were performed to investigate the relationship between SII and LEDVT risk. Propensity score matching (PSM), subgroup and sensitivity analyses were also conducted. Moreover, restricted cubic spline (RCS) regression and two-piecewise linear regression models were used to assess the dose-response relationship between natural log transformed SII [ln(SII)] and risk of LEDVT.

Results

A total of 16,725 consecutive hospitalized patients were included, and 1,962 LEDVT events occurred. After adjusting for confounding factors, patients in the high SII group (≥ 574.2 × 10⁹/L) showed a 1.740-fold risk of LEDVT (95% CI: 1.546-1.959, P < 0.001), and elevated ln(SII) was associated with a 36.1% increased risk of LEDVT (95% CI: 1.278-1.449, P < 0.001). PSM, subgroup and sensitivity analyses confirmed the robustness of the association. A non-linear relationship was observed (P _non-linear< 0.001), with a threshold value of 5.6 × 10⁹/L for all LEDVT events. Above the threshold, each unit increase in ln(SII) had a 1.369-fold higher risk of LEDVT (95% CI: 1.271-1.475, P < 0.001). The association also existed in both distal and proximal LEDVT.

Conclusion

Elevated SII is significantly associated with an increased risk of LEDVT in hospitalized patients. Additionally, the association is non-linear and exhibit a threshold effect.

Robust estimation of dementia prevalence from two-phase surveys with non-responders via propensity score stratification

BACKGROUND

Missing diagnoses are common in cross-sectional studies of dementia, and this missingness is usually related to whether the respondent has dementia or not. Failure to properly address this issue can lead to underestimation of prevalence. To obtain accurate prevalence estimates, we propose different estimation methods within the framework of propensity score stratification (PSS), which can significantly reduce the negative impact of non-response on prevalence estimates.

METHODS

To obtain accurate estimates of dementia prevalence, we calculated the propensity score (PS) of each participant to be a non-responder using logistic regression with demographic information, cognitive tests and physical function variables as covariates. We then divided all participants into five equal-sized strata based on their PS. The stratum-specific prevalence of dementia was estimated using simple estimation (SE), regression estimation (RE), and regression estimation with multiple imputation (REMI). These stratum-specific estimates were integrated to obtain an overall estimate of dementia prevalence.

RESULTS

The estimated prevalence of dementia using SE, RE, and REMI with PSS was 12.24%, 12.28%, and 12.20%, respectively. These estimates showed higher consistency than the estimates obtained without PSS, which were 11.64%, 12.33%, and 11.98%, respectively. Furthermore, considering only the observed diagnoses, the prevalence in the same group was found to be 9.95%, which is significantly lower than the prevalence estimated by our proposed method. This suggested that prevalence estimates obtained without properly accounting for missing data might underestimate the true prevalence.

CONCLUSION

Estimating the prevalence of dementia using the PSS provides a more robust and less biased estimate.

Target Trial Emulation and Bias Through Missing Eligibility Data: An Application to a Study of Palivizumab for the Prevention of Hospitalization Due to Infant Respiratory Illness

Target trial emulation (TTE) applies the principles of randomized controlled trials to the causal analysis of observational data sets. One challenge that is rarely considered in TTE is the sources of bias that may arise if the variables involved in the definition of eligibility for the trial are missing. We highlight patterns of bias that might arise when estimating the causal effect of a point exposure when restricting the target trial to individuals with complete eligibility data. Simulations consider realistic scenarios where the variables affecting eligibility modify the causal effect of the exposure and are missing at random or missing not at random. We discuss means to address these patterns of bias, namely: 1) controlling for the collider bias induced by the missing data on eligibility, and 2) imputing the missing values of the eligibility variables prior to selection into the target trial. Results are compared with the results when TTE is performed ignoring the impact of missing eligibility. A study of palivizumab, a monoclonal antibody recommended for the prevention of respiratory hospital admissions due to respiratory syncytial virus in high-risk infants, is used for illustration.

Tocilizumab Outcomes in Critically Ill COVID-19 Patients Admitted to the ICU and the Role of Non-Tocilizumab COVID-19-Specific Medical Therapeutics

Background: Tocilizumab is a monoclonal antibody proposed to manage cytokine release syndrome (CRS) associated with severe COVID-19. Previously published reports have shown that tocilizumab may improve the clinical outcomes of critically ill patients admitted to the ICU. However, no precise data about the role of other medical therapeutics concurrently used for COVID-19 on this outcome have been published. Objectives: We aimed to compare the overall outcome of critically ill COVID-19 patients admitted to the ICU who received tocilizumab with the outcome of matched patients who did not receive tocilizumab while controlling for other confounders, including medical therapeutics for critically ill patients admitted to ICUs. Methods: A prospective, observational, multicenter cohort study was conducted among critically ill COVID-19 patients admitted to the ICU of 14 hospitals in Saudi Arabia between 1 March 2020, and October 31, 2020. Propensity-score matching was utilized to compare patients who received tocilizumab to patients who did not. In addition, the log-rank test was used to compare the 28 day hospital survival of patients who received tocilizumab with those who did not. Then, a multivariate logistic regression analysis of the matched groups was performed to evaluate the impact of the remaining concurrent medical therapeutics that could not be excluded via matching 28 day hospital survival rates. The primary outcome measure was patients’ overall 28 day hospital survival, and the secondary outcomes were ICU length of stay and ICU survival to hospital discharge. Results: A total of 1470 unmatched patients were included, of whom 426 received tocilizumab. The total number of propensity-matched patients was 1278. Overall, 28 day hospital survival revealed a significant difference between the unmatched non-tocilizumab group (586; 56.1%) and the tocilizumab group (269; 63.1%) (p-value = 0.016), and this difference increased even more in the propensity-matched analysis between the non-tocilizumab group (466.7; 54.6%) and the tocilizumab group (269; 63.1%) (p-value = 0.005). The matching model successfully matched the two groups’ common medical therapeutics used to treat COVID-19. Two medical therapeutics remained significantly different, favoring the tocilizumab group. A multivariate logistic regression was performed for the 28 day hospital survival in the propensity-matched patients. It showed that neither steroids (OR: 1.07 (95% CI: 0.75–1.53)) (p = 0.697) nor favipiravir (OR: 1.08 (95% CI: 0.61–1.9)) (p = 0.799) remained as a predictor for an increase in 28 day survival. Conclusion: The tocilizumab treatment in critically ill COVID-19 patients admitted to the ICU improved the overall 28 day hospital survival, which might not be influenced by the concurrent use of other COVID-19 medical therapeutics, although further research is needed to confirm this.

Associations of Preexisting Vascular Risk Factors With Outcomes After Traumatic Brain Injury: A TRACK-TBI Study

OBJECTIVE

To evaluate associations of preinjury vascular risk factors with traumatic brain injury (TBI) outcomes.

SETTING

The level 1 trauma center-based T ransforming R esearch a nd C linical K nowledge in TBI (TRACK-TBI) Study.

PARTICIPANTS

A total of 2361 acute TBI patients 18 years or older who presented to the emergency department within 24 hours of head trauma warranting clinical evaluation with a noncontrast head CT between February 26, 2014, and August 8, 2018.

DESIGN

A multicenter prospective cohort study.

MAIN MEASURES

Vascular risk factors (hypertension, diabetes, hyperlipidemia, and smoking) were assessed at baseline by self- or proxy-report and chart review. The primary outcome was the 6-month Glasgow Outcome Scale-Extended TBI version (GOSE-TBI). Secondary 6-month outcomes included the Rivermead Post-Concussion Symptoms Questionnaire (RPQ), the Satisfaction with Life Scale (SWLS), and the 18-item Brief Symptom Inventory Global Severity Index (BSI-18-GSI).

RESULTS

Mean age of participants was 42 years, 31% were women, and 16% were Black. Current smoking was the most common vascular risk factor (29%), followed by hypertension (17%), diabetes (8%), and hyperlipidemia (6%). Smoking was the only risk factor associated with worse scores on all 4 outcome indices. Hypertension and diabetes were associated with worse RPQ scores, and hypertension was associated with worse BSI-18-GSI scores (all P < .05). Compared with individuals with no vascular risk factors, individuals with 1 but not 2 or more vascular risk factors had significantly worse GOSE-TBI and SWLS scores, while a higher burden of vascular risk factors was significantly associated with worse RPQ and BSI-18-GSI scores.

CONCLUSION

Our study found that preinjury vascular risk factors, especially smoking, are associated with worse outcomes after TBI. Aggressive postinjury treatment of vascular risk factors may be a promising strategy to improve TBI outcomes.

Should multiple imputation be stratified by exposure group when estimating causal effects via outcome regression in observational studies?

BACKGROUND

Despite recent advances in causal inference methods, outcome regression remains the most widely used approach for estimating causal effects in epidemiological studies with a single-point exposure and outcome. Missing data are common in these studies, and complete-case analysis (CCA) and multiple imputation (MI) are two frequently used methods for handling them. In randomised controlled trials (RCTs), it has been shown that MI should be conducted separately by treatment group. In observational studies, causal inference is now understood as the task of emulating an RCT, which raises the question of whether MI should be conducted by exposure group in such studies.

METHODS

We addressed this question by evaluating the performance of seven methods for handling missing data when estimating causal effects with outcome regression. We conducted an extensive simulation study based on an illustrative case study from the Victorian Adolescent Health Cohort Study, assessing a range of scenarios, including seven outcome generation models with exposure-confounder interactions of differing strength.

RESULTS

The simulation results showed that MI by exposure group led to the least bias when the size of the smallest exposure group was relatively large, followed by MI approaches that included the exposure-confounder interactions.

CONCLUSIONS

The findings from our simulation study, which was designed based on a real case study, suggest that current practice for the conduct of MI in causal inference may need to shift to stratifying by exposure group where feasible, or otherwise including exposure-confounder interactions in the imputation model.

Propensity score analysis with missing data using a multi-task neural network

BACKGROUND

Propensity score analysis is increasingly used to control for confounding factors in observational studies. Unfortunately, unavoidable missing values make estimating propensity scores extremely challenging. We propose a new method for estimating propensity scores in data with missing values.

MATERIALS AND METHODS

Both simulated and real-world datasets are used in our experiments. The simulated datasets were constructed under 2 scenarios, the presence (T = 1) and the absence (T = 0) of the true effect. The real-world dataset comes from LaLonde's employment training program. We construct missing data with varying degrees of missing rates under three missing mechanisms: MAR, MCAR, and MNAR. Then we compare MTNN with 2 other traditional methods in different scenarios. The experiments in each scenario were repeated 20,000 times. Our code is publicly available at https://github.com/ljwa2323/MTNN .

RESULTS

Under the three missing mechanisms of MAR, MCAR and MNAR, the RMSE between the effect and the true effect estimated by our proposed method is the smallest in simulations and in real-world data. Furthermore, the standard deviation of the effect estimated by our method is the smallest. In situations where the missing rate is low, the estimation of our method is more accurate.

CONCLUSIONS

MTNN can perform propensity score estimation and missing value filling at the same time through shared hidden layers and joint learning, which solves the dilemma of traditional methods and is very suitable for estimating true effects in samples with missing values. The method is expected to be broadly generalized and applied to real-world observational studies.

Protocol for the effectiveness evaluation of an antenatal, universally offered, and remotely delivered parenting programme 'Baby Steps' on maternal outcomes: a Born in Bradford's Better Start (BiBBS) study

BACKGROUND

Poor perinatal mental health and maternal sensitivity towards a child in the early years can carry a long-term cost to individuals and to society, and result in negative child outcomes such as poor mental health and social emotional issues. Despite the recognition of early intervention and prevention, there is mixed evidence regarding antenatal parenting interventions that aim to enhance perinatal mental health and maternal sensitivity to prevent negative child outcomes. 'Baby Steps' is a relationship-based antenatal and postnatal parenting programme. The service evaluated in this study is delivered in a low-income and ethnically diverse community via Better Start Bradford. This study aims to assess whether the universally, and remotely delivered Baby Steps programme is effective in improving postnatal maternal sensitivity (primary outcome) and postnatal maternal mental health (secondary outcome) when compared to services as usual 6-10 weeks post-birth. It will also assess differences in birth outcomes, and differences in the prevalence of poor perinatal mental ill health through routine data. The feasibility of collecting cost and health related resource use data for a future economic evaluation will be explored.

METHODS

The study is a quasi-experimental evaluation in a single centre. All participants are drawn from Born in Bradford's Better Start (BiBBS) interventional family cohort study. Intervention participants will be matched to a demographically comparable control group using propensity score matching. The required minimum sample is n = 130 (ratio 1:1) to detect a medium effect (± 2.35, d = .50) on the primary outcome-maternal-child sensitivity, using the Mothers Object Relations Scale Short Form (MORS-SF). Secondary outcomes include the Patient Health Questionnaire (PHQ-8), Generalised Anxiety Disorder assessment 7 (GAD-7), identification of poor perinatal mental health through routine data, and birth outcomes (delivery method, gestation period, low birth weight). Service delivery costs and health resource use will be gathered from routine data.

DISCUSSION

This study will evaluate the effectiveness of Baby Steps for enhancing maternal-child sensitivity and maternal mental health when delivered universally and remotely. The findings regarding programme effectiveness, process, and costs will be relevant for researchers, service commissioners, and service staff.

TRIAL REGISTRATION

This study was prospectively registered with ISRCTN (22/04/2022, ISRCTN12196131).

Labour market participation and retirement after stroke in Denmark: registry based cohort study

OBJECTIVE

To examine labour market participation and retirement among patients with stroke and matched people in the general population according to stroke subtype.

DESIGN

Nationwide, population based, matched cohort study.

SETTING

Danish Stroke Registry, covering all Danish hospitals, and other nationwide registries (2005-18).

PARTICIPANTS

Patients (aged 18-60 years and active in the labour market) with a first time diagnosis of ischaemic stroke (n=16 577), intracerebral haemorrhage (n=2025), or subarachnoid haemorrhage (n=4305), and individuals from the general population, matched on age, sex, and calendar year (n=134 428). The median Scandinavian stroke scale score was 55.

MAIN OUTCOME MEASURES

Unweighted prevalences of labour market participation, receipt of sick leave benefits, receipt of disability pension, voluntary early retirement, state pension, and death were computed for each week and up to five years after stroke diagnosis. A log-linear Poisson model was used to obtain exact prevalence estimates as well as propensity score weighted prevalence differences and prevalence ratios at six months, one year, two years, and five years after stroke diagnosis.

RESULTS

Most patients (62% of those with ischaemic stroke, 69% of those with intracerebral haemorrhage, and 52% of those with subarachnoid haemorrhage) went on sick leave within three weeks of diagnosis. Prevalence of labour market participation among patients with ischaemic stroke compared with matched individuals from the general population was 56.6% versus 96.6% at six months, and 63.9% versus 91.6% at two years. Prevalence of sick leave was 39.8% versus 2.6% at six months, and 15.8% versus 3.8% at two years. Prevalence of receipt of a disability pension was 0.9% versus 0.2% at six months, and 12.2% versus 0.6% at two years. Adjusting for socioeconomic and comorbidity differences between patients and matched individuals from the general population using propensity score weighting methods had little impact on contrasts. Patients with intracerebral haemorrhage had higher prevalences of sick leave and receipt of a disability pension and thus a lower prevalence of labour market participation, while prevalences for patients with subarachnoid haemorrhage were similar in magnitude to those for patients with ischaemic stroke.

CONCLUSIONS

In a highly resourced country, about two thirds of working age adults with ischaemic stroke of primarily mild severity participated in the labour market two years after diagnosis. Sick leave and receipt of a disability pension were the most common reasons for non-participation. Patients with intracerebral haemorrhage were less likely to return to the labour market than patients with ischaemic stroke and subarachnoid haemorrhage.

Anatomic distribution of lower extremity deep venous thrombosis is associated with an increased risk of pulmonary embolism: A 10-year retrospective analysis

Aims

To investigate the potential relationship between anatomic distribution of lower extremity deep venous thrombosis (LEDVT) and pulmonary embolism (PE).

Methods

A retrospective case-control study was performed in patients diagnosed with LEDVT, which were confirmed by bilateral lower extremity compression ultrasonography (CUS) examination. According to the ultrasound reports, thrombus sidedness was categorized as unilateral and bilateral lower extremity, thrombus location was classified into distal and proximal LEDVT. Anatomic distributions of LEDVT were further subdivided depending on the combination of thrombus sidedness and location. Patients with PE were identified using the International Classification of Diseases-10 (ICD-10) codes (I26.0 and I26.9), and divided into PE group and Non-PE group. Univariate and multivariate logistic regression analyses were used to assess the association between anatomic distribution of LEDVT and PE. Sensitivity analyses were also conducted.

Results

A total of 2,363 consecutive patients with LEDVT were included, of whom 66.10% and 33.90% were unilateral and bilateral LEDVT, as well as 71.39% and 28.61% were isolated distal and proximal LEDVT, respectively. After the diagnosis of LEDVT, 185 patients (7.83%) developed PE. The proportions of PE ranged between the lowest (4.07%) in unilateral-distal LEDVT and highest (14.55%) in bilateral-proximal LEDVT. Multivariate logistic regression analysis showed that bilateral LEDVT (odds ratios [OR] = 2.455, 95% confidence interval [CI]: 1.803-3.344, P < 0.001) and proximal LEDVT (OR = 1.530, 95% CI: 1.105-2.118, P = 0.010) were risk factors for developing PE. Moreover, unilateral-proximal (OR = 2.129, 95% CI: 1.365-3.320, P = 0.00), bilateral-distal (OR = 3.193, 95% CI: 2.146-4.752, P < 0.001) and bilateral-proximal LEDVT(OR = 3.425, 95% CI: 2.093-5.603, P < 0.001) were significantly associated with an increased risk of PE. Sensitivity analyses also confirmed the robustness of these associations.

Conclusion

Patients with unilateral-proximal, bilateral-distal or bilateral-proximal are more likely to suffer from PE than those with unilateral-distal LEDVT.

High systemic inflammation response index level is associated with an increased risk of lower extremity deep venous thrombosis: a large retrospective study

BACKGROUND

The association between inflammation and venous thromboembolism (VTE) has attracted increasing research interest. Recently, the systemic inflammation response index (SIRI) has been proposed as a novel inflammatory biomarker, but its potential association with lower extremity deep venous thrombosis (LEDVT) has not been investigated. Thus, this study aimed to explore the association between SIRI and LEDVT risk in a large sample over a 10-year period (2012-2022).

METHODS

All hospitalized patients who underwent lower extremity compression ultrasonography (CUS) examinations were consecutively identified from our hospital information system database. Multivariate logistic regression analysis was used to investigate the association between SIRI and LEDVT risk. Sensitivity, restricted cubic spline and subgroup analyses were also performed.

RESULTS

In total, 12643 patients were included, and 1346 (10.6%) LEDVT events occurred. After full adjustment, a higher SIRI level was significantly associated with an increased risk of LEDVT (odds ratio [OR] = 1.098, 95% confidence interval [CI]: 1.068-1.128, p < 0.001), and patients in quartile 4 had a 2.563-fold higher risk of LEDVT than those in quartile 1 (95% CI: 2.064-3.182, p < 0.001). A nonlinear relationship was observed (P for nonlinearity < 0.001), with an inflection point of 4.17. Below this point, each unit increase in SIRI corresponded to a 35.3% increase in LEDVT risk (95% CI: 1.255-1.458, p < 0.001). No significant difference was found above the inflection point (OR = 1.015, 95% CI: 0.963-1.069, p = 0.582). Sensitivity and subgroup analyses confirmed the robustness of the association. This association also existed in both distal and proximal LEDVT.

CONCLUSION

A High SIRI is significantly associated with an increased risk of LEDVT in hospitalized patients. Given that the SIRI is a readily available biomarker in clinical settings, its potential clinical use deserves further exploration.

Application of multilevel models for predicting pain following root canal treatment

OBJECTIVES

This study developed predictive models for one-week acute and six-month persistent pain following root canal treatment (RCT). An additional aim was to study the gain in predictive efficacy of models containing clinical factors only, over models containing sociodemographic characteristics.

METHODS

A secondary data analysis of 708 patients who received RCTs was conducted. Three sets of predictors were used: (1) combined set, containing all predictors in the data set; (2) clinical set and (3) sociodemographic set. Missing data were handled by multiple imputation using the missing indicator method. The multilevel least absolute selection and shrinkage operator (LASSO) regression was used to select predictors into the final multilevel logistic models. Three measures, the area under the receiver operating characteristic curve (AUROC) and precision-recall curve (AUPRC) and calibration curves, were used to assess the predictive performance of the models.

RESULTS

The selected-in factors in the final models, using LASSO regression, are related to pre- and intra-treatment clinical symptoms and pain experience. Predictive performance of the models remained the same with the inclusion (exclusion) of the socio-demographic factors. For predicting one-week outcome, the model built with combined set of predictors yielded the highest AUROC and AUPRC of 0.85 and 0.72, followed by the models built with clinical factors (AUROC = 0.82, AUPRC = 0.66). The lowest predictive ability was found in models with only sociodemographic characteristics (AUROC = 0.68, AUPRC = 0.40). Similar patterns were observed in predicting six-month outcome, where the AUROC for models with combined, clinical and sociodemographic sets of predictors were 0.85, 0.89 and 0.66, respectively, and the AUPRC were 0.48, 0.53 and 0.22, respectively.

CONCLUSIONS

Clinical factors such as the severity and experience of pre-operative and intra-operative pain were discovered important to the subsequent development of pain following RCTs. Adding sociodemographic characteristics to the models with clinical factors did not change the models' predictive performance or the proportion of explained variance.

Time-of-day dependent effects of midazolam administration on myocardial injury in non-cardiac surgery

Background

Animal studies have shown that midazolam can increase vulnerability to cardiac ischemia, potentially via circadian-mediated mechanisms. We hypothesized that perioperative midazolam administration is associated with an increased incidence of myocardial injury in patients undergoing non-cardiac surgery (MINS) and that circadian biology may underlie this relationship.

Methods

We analyzed intraoperative data from the Multicenter Perioperative Outcomes Group for the occurrence of MINS across 50 institutions from 2014 to 2019. The primary outcome was the occurrence of MINS. MINS was defined as having at least one troponin-I lab value ≥0.03 ng/ml from anesthesia start to 72 h after anesthesia end. To account for bias, propensity scores and inverse probability of treatment weighting were applied.

Results

A total of 1,773,118 cases were available for analysis. Of these subjects, 951,345 (53.7%) received midazolam perioperatively, and 16,404 (0.93%) met criteria for perioperative MINS. There was no association between perioperative midazolam administration and risk of MINS in the study population as a whole (odds ratio (OR) 0.98, confidence interval (CI) [0.94, 1.01]). However, we found a strong association between midazolam administration and risk of MINS when surgery occurred overnight (OR 3.52, CI [3.10, 4.00]) or when surgery occurred in ASA 1 or 2 patients (OR 1.25, CI [1.13, 1.39]).

Conclusion

Perioperative midazolam administration may not pose a significant risk for MINS occurrence. However, midazolam administration at night and in healthier patients could increase MINS, which warrants further clinical investigation with an emphasis on circadian biology.

Initial Assessment of a Brief Health, Fitness, and Spirituality Survey for Epidemiological Research: A Pilot Study

Background

Not many population-based health studies include items to assess both fitness and spirituality concepts. Therefore, the purpose of this study was to examine initial data of a brief health, fitness, and spirituality survey for epidemiological research.

Methods

This first phase pilot study used data from N = 56 adults 18+ years of age via electronic questionnaire. Measures of general health, fitness, physical activity (PA), body mass index (BMI), religiosity, and happiness were assessed. Reliability analyses were employed for PA, religiosity, and happiness scales. Validity coefficients were computed to evaluate convergence between scale scores and related measures. Finally, difference in health was examined between different levels of fitness to evaluate known groups discrimination.

Results

Respondents were middle-aged (Mean = 50.5 yr, SD = 14.3), majority white (69.5%, SD = 6.2), with relatively low BMI (Mean = 25.3, SD = 5.3). All three scales showed internal consistency reliability of α = 0.93, α = 0.89, and KR-20 = 0.56 for religiosity, happiness, and PA, respectively. Furthermore, scores converged (ps < 0.05) between fitness and PA (r = 0.43), health (r = 0.66), BMI (r = -0.28), and happiness (r = 0.25). Finally, health scores were significantly greater for high fitness versus low fitness in both male (p < 0.001) and female (p = 0.015) populations.

Conclusion

Results from this study indicate that a brief health, fitness, and spirituality survey can reliably measure its intended traits. A single-item of self-assessed fitness in particular has promise for large scale epidemiological research.

Risk of major adverse limb events in patients with type 2 diabetes mellitus receiving sodium glucose cotransporter 2 inhibitors and glucagon-like peptide-1 receptor agonists: A population-based retrospective cohort study

Background: Both sodium glucose cotransporter 2 inhibitors (SGLT-2i) and glucagon-like peptide-1 receptor agonists (GLP-1RA) have cardiovascular protective effects in patients with type 2 diabetes mellitus. However, the comparative risk of GLP-1RA versus SGLT-2i for major adverse limb events remains unknown.

Materials and methods: We studied a nationwide cohort involving 123,048 diabetes patients 20–100 years of age who initiated a SGLT-2i or GLP-1RA during 2012 and 2017. The patients in the two groups were matched by propensity score (PS), and incidence rates for hospitalization for major adverse limb events, critical limb ischemia (CLI) and lower extremity amputation (LEA), were assessed. Cox proportional hazards regression was applied to estimate hazard ratios (HRs) between patients receiving SGLT-2i as compared with GLP-1RA. The modification effects of age, a history of established cardiovascular disease, and chronic kidney disease were examined. In addition, use of dipeptidyl peptidase-4 inhibitor (DPP-4i) was chosen as a second active comparator.

Results: After PS-matching, a total of 13,378 SGLT-2i and 13,378 GLP-1RA initiators were identified. Use of SGLT-2i was not associated with an increased risk for hospitalization for CLI and LEA, either compared with GLP-1RA (HR, 1.13; 95% CI, 0.77–1.65 and 1.27; 95% CI, 0.63–2.55, respectively) or compared with DPP-4i use (HR, 1.06; 95% CI, 0.75–1.50 and HR, 0.80; 95% CI, 0.42–1.53, respectively). Although the study was underpowered to explore potential effect modification, a trend of higher risks for LEA was noted among SGLT-2i users with cardiovascular disease as compared with either GLP-1RA or DPP-4i.

Conclusion: Use of SGLT-2i was not associated with higher risks for hospitalization for CLI and LEA as compared with reference drugs. Further large-scale studies are needed for a precise risk estimation.

Diagnostic stewardship for blood cultures in the emergency department: A multicenter validation and prospective evaluation of a machine learning prediction tool

Background

Overuse of blood cultures (BCs) in emergency departments (EDs) leads to low yields and high numbers of contaminated cultures, accompanied by increased diagnostics, antibiotic usage, prolonged hospitalization, and mortality. We aimed to simplify and validate a recently developed machine learning model to help safely withhold BC testing in low-risk patients.

Methods

We extracted data from the electronic health records (EHR) for 44.123 unique ED visits with BC sampling in the Amsterdam UMC (locations VUMC and AMC; the Netherlands), Zaans Medical Center (ZMC; the Netherlands), and Beth Israel Deaconess Medical Center (BIDMC; United States) in periods between 2011 and 2021. We trained a machine learning model on the VUMC data to predict blood culture outcomes and validated it in the AMC, ZMC, and BIDMC with subsequent real-time prospective evaluation in the VUMC.

Findings

The model had an Area Under the Receiver Operating Characteristics curve (AUROC) of 0.81 (95%-CI = 0.78–0.83) in the VUMC test set. The most important predictors were temperature, creatinine, and C-reactive protein. The AUROCs in the validation cohorts were 0.80 (AMC; 0.78–0.82), 0.76 (ZMC; 0.74–0.78), and 0.75 (BIDMC; 0.74–0.76). During real-time prospective evaluation in the EHR of the VUMC, it reached an AUROC of 0.76 (0.71–0.81) among 590 patients with BC draws in the ED. The prospective evaluation showed that the model can be used to safely withhold blood culture analyses in at least 30% of patients in the ED.

Interpretation

We developed a machine learning model to predict blood culture outcomes in the ED, which retained its performance during external validation and real-time prospective evaluation. Our model can identify patients at low risk of having a positive blood culture. Using the model in practice can significantly reduce the number of blood culture analyses and thus avoid the hidden costs of false-positive culture results.

Funding

This research project was funded by the Amsterdam Public Health – Quality of Care program and the Dutch “Doen of Laten” project (project number: 839205002).

Association of early sedation level with patient outcomes in moderate-to-severe acute respiratory distress syndrome: Propensity-score matched analysis

Purpose

Studies of early depth of sedation in mixed critically ill populations have suggested benefit to light sedation; however, the relationship of early depth of sedation with outcomes in patients with acute respiratory distress syndrome (ARDS) is unknown.

Materials and methods

We performed a propensity-score matched analysis of early light sedation (Richmond Agitation Sedation Scale Score, RASS 0 to −1 or equivalent) versus deep sedation (RASS −2 or lower) in patients enrolled in the non-intervention group of The Reevaluation of Systemic Early Neuromuscular Blockade trial. Primary outcome was 90 day mortality. Secondary outcomes included days free of mechanical ventilation, days not in ICU, days not in hospital at day 28.

Results

137 of 486 participants (28.2%) received early light sedation. Vasopressor usage and Apache III scores significantly differed between groups. Prior to matching, 90-day mortality was higher in the early deep sedation (45.3%) compared to light sedation (34.2%) group. In the propensity score matched cohort, there was no difference in 90-day mortality (Odds Ratio (OR) 0.72, 95% CI 0.41, 1.27, p = 0.26) or secondary outcomes between the groups.

Conclusions

We did not find an association between early depth of sedation and clinical outcomes in this cohort of patients with moderate-to-severe ARDS.

Dealing with Missing Responses in Cognitive Diagnostic Modeling

Missing data are a common problem in educational assessment settings. In the implementation of cognitive diagnostic models (CDMs), the presence and/or inappropriate treatment of missingness may yield biased parameter estimates and diagnostic information. Using simulated data, this study evaluates ten approaches for handling missing data in a commonly applied CDM (the deterministic inputs, noisy “and” gate (DINA) model): treating missing data as incorrect (IN), person mean (PM) imputation, item mean (IM) imputation, two-way (TW) imputation, response function (RF) imputation, logistic regression (LR), expectation-maximization (EM) imputation, full information maximum likelihood (FIML) estimation, predictive mean matching (PMM), and random imputation (RI). Specifically, the current study investigates how the estimation accuracy of item parameters and examinees’ attribute profiles from DINA are impacted by the presence of missing data and the selection of missing data methods across conditions. While no single method was found to be superior to other methods across all conditions, the results suggest the use of FIML, PMM, LR, and EM in recovering item parameters. The selected methods, except for PM, performed similarly across conditions regarding attribute classification accuracy. Recommendations for the treatment of missing responses for CDMs are provided. Limitations and future directions are discussed.

The effect of national antenatal care guidelines and provider training on obstetric danger sign counselling: a propensity score matching analysis of the 2014 Ethiopia service provision assessment plus survey

Background

Most pregnant women in low and lower-middle-income countries do not receive all components of antenatal care (ANC), including counselling on obstetric danger signs. Facility-level ANC guidelines and provider in-service training are major factors influencing ANC counselling. In Ethiopia, little is known about the extent to which guidelines and provider in-service training can increase the quality of ANC counselling.

Methods

We examined the effect of national ANC guidelines and ANC provider in-service training on obstetric danger sign counselling for pregnant women receiving ANC using the 2014 Ethiopian service provision assessment plus (ESPA +) survey data. We created two analysis samples by applying a propensity score matching method. The first sample consisted of women who received ANC at health facilities with guidelines matched with those who received ANC at health facilities without guidelines. The second sample consisted of women who received ANC from the providers who had undertaken in-service training in the last 24 months matched with women who received ANC from untrained providers. The outcome variable was the number of obstetric danger signs described during ANC counselling, ranging from zero to eight. The covariates included women’s socio-demographic characteristics, obstetric history, health facility characteristics, and ANC provider characteristics.

Results

We found that counselling women about obstetric danger signs during their ANC session varied according to the availability of ANC guidelines (61% to 70%) and provider training (62% to 68%). After matching the study participants by the measured covariates, the availability of ANC guidelines at the facility level significantly increased the average number of obstetric danger signs women received during counselling by 24% (95% CI: 12–35%). Similarly, providing refresher training for ANC providers increased the average number of obstetric danger signs described during counselling by 37% (95% CI: 26–48%).

Conclusion

The findings suggest that the quality of ANC counselling in Ethiopia needs strengthening by ensuring that ANC guidelines are available at every health facility and that the providers receive regular ANC related in-service training.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12978-022-01442-6.

Maternal death from preventable pregnancy-related complications remains a global health challenge. In 2017, there were 295,000 maternal deaths worldwide, and about two-thirds of these deaths were from Sub-Saharan Africa. Ethiopia is a Sub-Saharan African country with 401 maternal deaths per 100,000 live births in 2017, and this rate is higher than the target indicated in sustainable development goals. Most maternal deaths are due to obstetric complications and could have been averted through early detection and treatment. Providing antenatal care counselling about obstetric danger signs enhances women’s awareness of obstetric complications and encourages women to seek treatment from a skilled care provider. However, most women from low-income settings, including Ethiopia, do not receive counselling about obstetric danger signs. Facility-level antenatal care guidelines and provider in-service training improve antenatal care counselling. In Ethiopia, little is known to what extent antenatal care guidelines and provider training increase counselling on obstetric danger signs. The present study used the 2014 Ethiopian service provision assessment data and estimated the effect of antenatal care guidelines and provider training on counselling about obstetric danger signs. The analysis involved a propensity score matching method and included 1725 pregnant women. The study found that antenatal care guidelines at health facilities and antenatal care provider in-service training significantly increase counselling on obstetric danger signs by 24% and 37%, respectively. The finding suggests improving the quality of antenatal care counselling in Ethiopia needs antenatal care guidelines at each antenatal care clinic and refresher training for the providers.

Right-wing ideological constraint and vaccine refusal: The case of the COVID-19 vaccine in Norway

Research has shown a correlation between votes for populist parties and the belief that vaccines are not important or effective. More recent investigations in the United States and France have similarly shown that attitudes toward the COVID-19 vaccine have been politicized. In this article, we show a similar pattern analyzing survey data from Norway, a country relatively mildly hit by the pandemic and characterized by high trust and a consensual political culture. We find that refusal to vaccinate is associated with right-wing ideological constraint, even when considering a wide array of control variables (e.g., lack of confidence, complacency), and sociodemographic characteristics. The results imply that vaccine refusal latch onto established political cleavages, particularly among the most ideologically consistent. Thus, polarization in the form of increasing ideological constraint may represent a mounting challenge for vaccine uptake, suggesting that vaccine communication should go beyond "explaining the science" and factor in ideology.

Surgical and Oncologic Outcomes of Robotic and Conventional Nipple-Sparing Mastectomy with Immediate Reconstruction: International Multicenter Pooled Data Analysis

BACKGROUND

Robotic nipple-sparing mastectomy (RNSM) has been developed to reduce conspicuous scar and increase the quality of life in women. This study aimed to evaluate the surgical and oncologic outcomes of RNSM with immediate breast reconstruction (IBR) compared with conventional nipple-sparing mastectomy (CNSM).

PATIENTS AND METHODS

This international multicenter, pooled analysis of individual patient-level data enrolled a total of 755 procedures in 659 women (609 had breast cancer and 50 underwent risk-reducing mastectomy) who underwent nipple-sparing mastectomy with IBR. Surgical and oncologic outcomes, including 30-days postoperative (POD 30d) complication rate, nipple necrosis rate, grade of Clavien-Dindo classification, disease-free survival, and overall survival, were evaluated. Propensity score-matched analyses were performed to adjust for confounding factors.

RESULTS

The median age of both the RNSM and CNSM groups was 45 years. The RNSM group had lower body mass index (BMI) and a higher proportion of benign disease compared with the CNSM group. POD 30d complications and postoperative complication grade III rates were lower in the RNSM group than in the CNSM group (p < 0.05). The nipple necrosis rate was 2.2% and 7.8% for RNSM and CNSM, respectively (p = 0.002). After propensity score matching, significantly lower rates of POD 30d complications, nipple necrosis, and postoperative complication grade III occurred in the RNSM group than in the CNSM group (all p < 0.05). Oncologic outcomes were not significantly different between the two groups.

CONCLUSION

RNSM can provide better cosmetic results with favorable surgical and oncologic outcomes for women with early breast cancer or BRCA mutation.