Wei-Wen Hsu

Objective: To examine the intersection of Social Determinants of Health (SDoH) and healthcare needs documented in electronic health records (EHR) of women with breast cancer. Methods: Using the Kaiser Family Foundation (KFF) framework to categorize SDoH, we conducted a qualitative content analysis, combining deductive and inductive approaches of EHR notes for female breast cancer patients treated at a mid‐sized Midwestern academic cancer center from 2017 to 2022. Records were eligible if patients were ≥ 18 years, received anticancer treatment (medications/therapies, radiation, or surgery), were alive, and had ≥ 1 social work note. Exclusions included male sex, death, or end‐of‐life/hospice care. We randomly selected 200 patients, oversampling racial/ethnic minorities to ensure balanced representation. The final sample ( N = 200) was 65.5% White, 97.8% non‐Hispanic, 41.5% older adults, and 38% married. Results: Analysis demonstrated the interconnectedness of SDoH and their cumulative effects on outcomes. Economic strain, transportation barriers, and limited social support were critical themes shaping patient experiences and care navigation. Mental health needs were prominent and intersected with multiple SDoH, highlighting the value of holistic approaches that address medical and non‐medical factors in breast cancer management. Conclusion: Routine assessment of SDoH and mental health can help clinicians identify at‐risk patients and link them to resources. Integrating standardized SDoH evaluations into clinical workflows may strengthen treatment adherence, improve outcomes, and reduce disparities in breast cancer care at academic and community levels.

Background: The United States continues to experience a major drug epidemic driven by synthetic opioids, particularly fentanyl. Emerging evidence suggests that opioids alter immune responses and viral pathogenesis. This study evaluated inflammatory and apoptotic markers among individuals living with HIV and/or hepatitis C virus (HCV) infection in the context of opioid/fentanyl exposure. Methods: Whole blood was collected from adults with HIV, HCV, and/or opioid use disorder (OUD), along with healthy controls. Plasma toxicology screening for illicit substances was performed using LC-MS/MS. Inflammatory cytokines and chemokines (IL-6, IL-8, SDF-1α, MIP-1α, MIP-1β, RANTES, and TNF-α) were measured by Luminex, and M30 was measured by ELISA. Biomarkers were compared using the Kruskal–Wallis test, with multivariable regression used to identify independent associations. Results: Among 98 participants, 37.8% had OUD, 53.9% had a history of injection drug use, and 95.9% of toxicology-confirmed opioid users tested positive for fentanyl. Participants with HIV had higher RANTES and lower M30 levels, whereas participants with HCV had higher IL-8, MIP-1β, TNF-α, and M30 levels and lower RANTES. Multivariable analysis identified independent associations of IL-8 with HIV and HCV status, SDF-1α with injection drug use and recent fentanyl exposure, and MIP-1β with HCV status. The number of substances detected was not independently associated with inflammatory marker levels. Conclusions: HIV and HCV were associated with distinct inflammatory and apoptotic profiles in the setting of widespread fentanyl exposure, suggesting that chronic viral infection and opioid exposure contribute to immune dysregulation and may influence disease progression.

Background: Prostate cancer (PCa) is one of the most commonly diagnosed cancers. Treatments for PCa with less adverse effects than whole gland interventions, such as focal therapy (FT), often involve a higher risk of PCa recurrence. Combining FT with other treatments could increase the efficacy while maintaining a low side effect profile. This study aims to determine the proportion of residual/recurrent clinically significant PCa following the combination treatment of three months of androgen deprivation therapy (ADT) and FT of the prostate, as well as the safety of this treatment regimen. Methods: This study will be a single arm phase II trial with a recruitment goal of 57 patients with treatment naïve non-metastatic intermediate risk PCa. Patients will complete the I-PSS, SHIM, and EPIC-26 questionnaires at the screening visit. Patients will then comply with a three-month treatment course of ADT. Eight weeks after beginning ADT, patients will undergo FT of the prostate via high intensity focused ultrasound or cryoablation. Follow up visits will occur every three months for a year post FT to monitor for side effects, repeat questionnaires, perform a clinical assessment, and obtain PSA and testosterone values. Twelve months after FT, a surveillance multiparametric MRI and MRI-targeted biopsy will be performed to assess for treatment failure. Discussion: The primary endpoint of this trial is to determine the proportion of men with clinically significant PCa as evaluated by a surveillance mpMRI and MRI-TB at 12-months following FT. If successful, this treatment approach could offer a new option for treatment with fewer side effects than whole gland interventions and more efficacy than FT alone. Furthermore it could inform the need for further research into multimodal treatment options for PCa. Clinical trial registration: ClinicalTrials.gov, NCT05790213. Registered on March 30, 2023.

Unplanned cancer readmissions present a significant burden on patients and hospitals. Current predictive models often overlook socioeconomic factors such as social determinants of health (SDoH), which have the potential to improve prediction performance, as measured by the Area Under the Receiver Operating Characteristic (AUROC) and the Precision-Recall Curve (AUPRC). To investigate this, the present study developed predictive models using cancer readmission data from a large health system in Hamilton County, OH. The models incorporated geocoding-based SDoH along with clustering techniques and compared machine learning (ML) and deep learning (DL) algorithms. Overall, models, regardless of algorithm type, not using SDoH variables had higher AUROC and AUPRCs. The best-performing ML and DL models are comparable (AUROC = 0.7605 for ML; AUROC = 0.7585 for DL). However, when top-performing models were evaluated across certain organ and system cancers, using SDoH and clustering techniques significantly improved model performance. This was most notable for cancers of the skin, subcutaneous tissue, and breast with improvements of 8.20% in AUROC and 11.04% in AUPRC. For all cancer patient cases, utilizing individualized SDoH information extracted from clinical notes was recommended for future studies.

The impact of residential proximity (RP) to EPA-designated Superfund sites on personal well-being remains understudied. This study evaluated the well-being of residents living near a former uranium processing facility in Fernald, Ohio, now a Superfund site, which emitted significant amounts of radioactive hazardous waste. Baseline data were collected from 7,957 participants who voluntarily enrolled in the Fernald Community Cohort. Participants completed the Health Risk Appraisal (HRA), which assessed four well-being indicators: Life Satisfaction, Life in Control, Life Perspective, and Overall Physical Health. Separate multivariable ordinal logistic regression models were used to evaluate the cross-sectional relationship between RP and well-being outcomes. The results indicated that residents living within a 1-mile radius of the Superfund site had 38% greater odds of reporting lower Life Satisfaction compared to those residing 4–5 miles away [OR 1.38, 95% CI: 1.17, 1.64]. In the fully adjusted models, other significant risk factors emerged. Participants who reported two or more recent misfortunes in the past year were more likely to report a negative Life Perspective [OR 2.79, 95% CI: 2.36, 3.32]. Additionally, those with a weakened social ties had significantly higher odds of reporting diminished Life Satisfaction [OR 5.16, 95% CI: 4.33, 6.16]. This study identified RP to a Superfund site and psychosocial characteristics as significant predictors of well-being. Given the global prevalence of contaminated sites, particularly in socioeconomically disadvantaged regions, further investigation into their impact on local communities is essential.

The impact of living near environmental contamination on the health-related quality of life (HRQoL) is not well understood. This study examined the impacts of the residential proximity (RP) and time spent near a former Department of Energy uranium processing facility (located in Fernald, Ohio) on the surrounding community’s HRQoL outcomes. A cross-sectional analysis was conducted using the data collected from participants using the Short Form-36 survey at the time of the enrollment in the Fernald Medical Monitoring Program (n = 7957). Mental and physical component summary scores (MCS and PCS, respectively) were computed for each participant. The scores were compared among the cohort participants, stratified by the RP to the facility and to the U.S. general population. Multivariable linear regression analyses were performed to identify associations between the RP from the facility, duration at residence, participant characteristics, and HRQoL. The adults and minors at enrollment (MAEs) living within two miles of the facility reported slightly lower MCS scores compared to those of residents who lived >2 miles from the facility, after controlling for confounding variables (adults: marginal effect (ME): −0.553, p-value: 0.002, MAEs: ME: −1.594, p-value: 0.040). The RP had a significant association with PCS scores among adults but not with the MAEs (adults: ME: −0.456, p-value: 0.010). No significant relationship was observed between the duration at residence and HRQoL. Considering the association between the RP and HRQoL in the Fernald cohort, integrating a health perception screening tool into community healthcare practices would benefit residents near environmentally contaminated sites to capture the variable nature of perceived health over time.

Focal Therapy (FT) is an emerging treatment modality for prostate cancer (PCa). Due to its novelty, the research exploring how patients should be followed-up after treatment is limited. There is currently no established role for non-prostate-specific-antigen (PSA) biomarkers and PSMA PET. However, a combination of PSA testing, multiparametric magnetic resonance imaging (mpMRI), and systematic and targeted biopsies should routinely be used for surveillance after FT. PSA values that rise 1.0 ng/mL over the nadir after twelve months or rise 1.5 ng/mL over nadir after twenty-four to thirty-six months should raise suspicion for recurrence. The standard imaging technique is mpMRI, but it can often be difficult to interpret after FT, so using a scoring system such as prostate imaging after focal ablation (PI-FAB) or the transatlantic recommendations for prostate gland evaluation with magnetic resonance imaging after focal therapy (TARGET) allows for greater consistency between readers. This review seeks to summarize the current literature regarding surveillance after FT as it relates to biomarkers, imaging, biopsies, and consensus statements.

Background: Prostate cancer (PCa) diagnosis remains a complex field of study. Multiparametric magnetic resonance imaging (mpMRI) technology presents opportunities to enhance diagnostic precision. While recent advances in imaging and biopsy techniques show promise, the oncological implications of prebiopsy magnetic resonance imaging (MRI) and combination biopsy (ComBx) are not fully understood. This retrospective study evaluates the potential clinical impact of prebiopsy MRI and ComBx on PCa treatment outcomes. Methods: We conducted a comprehensive review of treatment-naïve patients undergoing prostate biopsy and subsequent radiation therapy (RT) or radical prostatectomy at the University of Cincinnati Health Center (2014–2020). Patients were categorized into two cohorts: those with prebiopsy mpMRI and ComBx versus those with systematic biopsy (SBx) alone. Patients with prostate-specific antigen (PSA) > 20 ng/mL were excluded. Biochemical recurrence (BCR) was defined as PSA ≥ 0.2 ng/mL post-prostatectomy or ≥2 ng/mL above nadir post-RT. Results: This study included 518 patients (189 SBx, 329 ComBx) with a median follow-up of 19.1 months. Median patient ages were 65.9 years (SBx) and 64.6 years (ComBx). The overall BCR rate was 10% with significantly lower rates in the ComBx group compared to SBx (6.4% vs. 16.4%, p < 0.001). Multivariable Cox regression analysis showed patients undergoing prebiopsy mpMRI with ComBx were 63% less likely to experience BCR (HR: 0.37, 95%CI 0.20–0.70, p = 0.002). Conclusions: Prebiopsy MRI followed by ComBx demonstrated lower BCR rates, suggesting improved PCa diagnosis and risk stratification. These findings highlight the potential of advanced imaging and biopsy techniques to benefit the management of PCa. Further longitudinal studies are needed to confirm the long-term clinical benefits of this approach.

Background: Actinic keratosis (AK) is a pre-cancerous skin condition caused by sun exposure. Number bias, a phenomenon that occurs when meaning other than numerical value is associated with numbers, may influence the reporting of AK removal. The present study aims to determine if number bias is affecting healthcare providers’ documentation of patient-provider encounters. Methods: A single-center retrospective chart review of 1415 patients’ charts was conducted at the University of Cincinnati Medical Center. To determine if there was a significant difference between even and odd-numbered AK removals reported, an exact binomial test was used. The frequency of removals per encounter was fitted to a zero-truncated negative binomial distribution to predict the number of removals expected. All data were analyzed with RStudio. Results: There were 741 odd and 549 even encounters. Odd removals were reported at a significantly greater frequency than even p < 0.001. Age may be contributing to the observed number bias (p < 0.001). One, two, and eight were reportedly removed more frequently, while nine, 13, and 14 were reportedly removed less frequently than expected, respectively. Conclusion: Number bias may be affecting clinicians’ documentation of AK removal and should be investigated in other clinical settings.

In many applications of hierarchical models, there is often interest in evaluating the inherent heterogeneity in view of observed data. When the underlying hypothesis involves parameters resting on the boundary of their support space such as variances and mixture proportions, it is a usual practice to entertain testing procedures that rely on common heterogeneity assumptions. Such procedures, albeit omnibus for general alternatives, may entail a substantial loss of power for specific alternatives such as heterogeneity varying with covariates. We introduce a novel and flexible approach that uses covariate information to improve the power to detect heterogeneity, without imposing unnecessary restrictions. With continuous covariates, the approach does not impose a regression model relating heterogeneity parameters to covariates or rely on arbitrary discretizations. Instead, a scanning approach requiring continuous dichotomizations of the covariates is proposed. Empirical processes resulting from these dichotomizations are then used to construct the test statistics, with limiting null distributions shown to be functionals of tight random processes. We illustrate our proposals and results on a popular class of two-component mixture models, followed by simulation studies and applications to two real datasets in cancer and caries research.