Association of household net worth with healthcare costs after radical cystectomy using real-world data.

BACKGROUND: Financial toxicity of bladder cancer care may influence how patients utilize healthcare resources, from emergency department (ED) encounters to office visits. We aim to examine whether greater household net worth (HHNW) confers differential access to healthcare resources after radical cystectomy (RC). METHODS: This population-based cohort study examined the association between HHNW and healthcare utilization costs in the 90 days post-RC in commercially insured patients with bladder cancer. Costs accrued from the index hospitalization to 90 days after including health plan costs (HPC) and out-of-pocket costs (OPC). Multivariable logistic regression models were generated by encounter (acute inpatient, ED, outpatient, and office visit). RESULTS: A total of 141,903 patients were identified with HHNW categories near evenly distributed. Acute inpatient encounters incurred the greatest HPC and OPC. Office visits conferred the lowest HPC while ED visits had the lowest OPC. Black patients harbored increased odds of an acute inpatient encounter (OR 1.22, 95% CI 1.16-1.29) and ED encounter (OR 1.20, 95% CI 1.14-1.27) while Asian (OR 0.76, 95% CI 0.69-0.85) and Hispanic (OR 0.74, 95% CI 0.69-0.78, p < 0.001) patients had lower odds of an outpatient encounter, compared to White counterpart. Increasing HHNW was associated with decreasing odds of acute inpatient or ED encounters and greater odds of office visits. CONCLUSIONS: Lower HHNW conferred greater risk of costly inpatient encounters while greater HHNW had greater odds of less costly office visits, illustrating how financial flexibility fosters differences in healthcare utilization and lower costs. HHNW may serve as a proxy for financial flexibility and risk of financial hardship than income alone.

Impact of an automated peri-procedural digital health intervention on rates of emergency department visits and readmissions.

BACKGROUND: Providing timely peri-procedural education, reminders, and check-ins can improve patient adherence and clinical outcomes. We sought to retrospectively evaluate the impact of a peri-procedural digital health tool on emergency department (ED) visits and readmissions. METHODS: A digital health tool for peri-procedural care engaged patients at scheduled intervals, resulting in an overall engagement score. Multivariate models determined predictors of tool engagement and post-procedural 30- and 90-day rehospitalizations and ED visits. RESULTS: 11,737 unique completed procedures were analyzed from 10,438 patients. Patients of Black and Latinx race/ethnicity (vs White), those with Medicare and Medicaid insurance (vs commercial), and those with non-activated patient portals (vs activated) were less likely to engage. After adjustment for confounders, higher engagement with the tool was associated with lower rates of 30-day hospitalizations (OR 0.64), 90-day hospitalizations (OR 0.65), and 90-day ED visits (OR 0.77). CONCLUSIONS: Highly engaged patients had fewer 30-day and 90-day ED visit and readmissions, even after adjustment for key confounders. Engagement, and thus the resulting benefits, were not equitably distributed.

Patient engagement in a mobile health intervention to improve preparedness for prostate biopsy.

OBJECTIVE: We designed and implemented a peri-procedural text message (SMS) program for patients undergoing transrectal prostate biopsy and aimed to evaluate predictors of patient enrollment and engagement with the SMS program. METHODS: We designed an SMS-based program with 8 messages containing web-based modules with educational content and reminders confirming MRI for fusion biopsy, antibiotic adherence, enema use, and anticoagulation cessation. Data on patient demographics, enrollment, and engagement with modules were collected from June 1, 2018 to February 28, 2021. Engagement was defined as a patient clicking a link delivered via SMS to access modules. We made multivariable models to identify predictors of patient enrollment and engagement. RESULTS: Of the 1,760 prostate biopsies between June 2018 and March 2021, 1,383 (78.6%) were enrolled in SMS, 182 (10.3%) in email, 106 (6.0%) in both, and 240 (13.6%) were not enrolled. Of 1418 patients enrolled, 1,270 (89.6%) engaged with at least one module. African American patients had 50% lower odds of being enrolled (OR = 0.50, 95% CI 0.28-0.96; P = 0.03), but once enrolled there were no differences in engagement. Patients for whom English was not listed as their primary language had 60% lower odds of engagement (OR = 0.40, 95% CI 0.17-1.00, P = .04) and patients who were single or divorced had a 40% lower odds of engagement (OR = 0.60, 95% CI 0.41-0.91, P = 0.01). CONCLUSIONS: A cohort of older men undergoing prostate biopsy were able to engage with a text message-based education and reminder program. Future efforts must address barriers to enrollment for Black or African American men and improve accessibility to non-English speaking patients.

Improving natural language information extraction from cancer pathology reports using transfer learning and zero-shot string similarity.

OBJECTIVE: We develop natural language processing (NLP) methods capable of accurately classifying tumor attributes from pathology reports given minimal labeled examples. Our hierarchical cancer to cancer transfer (HCTC) and zero-shot string similarity (ZSS) methods are designed to exploit shared information between cancers and auxiliary class features, respectively, to boost performance using enriched annotations which give both location-based information and document level labels for each pathology report. MATERIALS AND METHODS: Our data consists of 250 pathology reports each for kidney, colon, and lung cancer from 2002 to 2019 from a single institution (UCSF). For each report, we classified 5 attributes: procedure, tumor location, histology, grade, and presence of lymphovascular invasion. We develop novel NLP techniques involving transfer learning and string similarity trained on enriched annotations. We compare HCTC and ZSS methods to the state-of-the-art including conventional machine learning methods as well as deep learning methods. RESULTS: For our HCTC method, we see an improvement of up to 0.1 micro-F1 score and 0.04 macro-F1 averaged across cancer and applicable attributes. For our ZSS method, we see an improvement of up to 0.26 micro-F1 and 0.23 macro-F1 averaged across cancer and applicable attributes. These comparisons are made after adjusting training data sizes to correct for the 20% increase in annotation time for enriched annotations compared to ordinary annotations. CONCLUSIONS: Methods based on transfer learning across cancers and augmenting information methods with string similarity priors can significantly reduce the amount of labeled data needed for accurate information extraction from pathology reports.

Interactive Data Visualization Tool for Patient-Centered Decision Making in Kidney Cancer.

PURPOSE: Patients and providers often lack clinical decision tools to enable effective shared decision making. This is especially true in the rapidly changing therapeutic landscape of metastatic kidney cancer. Using the International Metastatic Renal Cell Carcinoma Database Consortium (IMDC) criteria, a validated risk prediction tool for patients with metastatic renal cell carcinoma, we created and user-tested a novel interactive visualization for clinical use. METHODS: An interactive visualization depicting IMDC criteria was created, with the final version including data for more than 4,500 patients. Usability testing was performed with nonmedical lay-users and medical oncology fellow physicians. Subjects used the tool to calculate median survival times based on IMDC criteria. User confidence was surveyed. An iterative user feedback implementation cycle was completed and informed revision of the tool. RESULTS: The tool is available at CloViz-IMDC. Initially, 400 lay-users and 15 physicians completed clinical scenarios and surveys. Cumulative accuracy across scenarios was higher for physicians than lay-users (84% v 74%; P = .03). Eighty-three percent of lay-users and 87% of physicians thought the tool became intuitive with use. Sixty-eight percent of lay-users wanted to use the tool clinically compared with 87% of physicians. After revisions, the updated tool was user-tested with 100 lay-users and 15 physicians. Physicians, but not lay-users, showed significant improvement in accuracy in the updated version of the tool (90% v 67%; P = .008). Seventy-two percent of lay-users and 93% of physicians wanted to use the updated tool in a clinical setting. CONCLUSION: A graphical method of interacting with a validated nomogram provides prognosis results that can be used by nonmedical lay-users and physicians, and has the potential for expanded use across many clinical conditions.

Supervised line attention for tumor attribute classification from pathology reports: Higher performance with less data.

OBJECTIVE: We aim to build an accurate machine learning-based system for classifying tumor attributes from cancer pathology reports in the presence of a small amount of annotated data, motivated by the expensive and time-consuming nature of pathology report annotation. An enriched labeling scheme that includes the location of relevant information along with the final label is used along with a corresponding hierarchical method for classifying reports that leverages these enriched annotations. MATERIALS AND METHODS: Our data consists of 250 colon cancer and 250 kidney cancer pathology reports from 2002 to 2019 at the University of California, San Francisco. For each report, we classify attributes such as procedure performed, tumor grade, and tumor site. For each attribute and document, an annotator trained by an oncologist labeled both the value of that attribute as well as the specific lines in the document that indicated the value. We develop a model that uses these enriched annotations that first predicts the relevant lines of the document, then predicts the final value given the predicted lines. We compare our model to multiple state-of-the-art methods for classifying tumor attributes from pathology reports. RESULTS: Our results show that across colon and kidney cancers and varying training set sizes, our hierarchical method consistently outperforms state-of-the-art methods. Furthermore, performance comparable to these methods can be achieved with approximately half the amount of labeled data. CONCLUSION: Document annotations that are enriched with location information are shown to greatly increase the sample efficiency of machine learning methods for classifying attributes of pathology reports.

Residual Benign Prostate Glandular Tissue after Radical Prostatectomy is Not Associated with the Development of Detectable Postoperative Serum Prostate Specific Antigen.

PURPOSE: To determine if benign glandular tissue at the surgical margin (BGM) is associated with detectable prostate specific antigen (PSA) and/or biochemical recurrence (BCR) after radical prostatectomy (RP). MATERIALS AND METHODS: Participants underwent RP for localized prostate cancer between 2004 and 2018. Regression analysis was used to identify demographic, clinical and surgical factors associated with the likelihood of BGM presence on surgical pathology. Oncologic outcomes included detectable PSA (>0.03 ng/ml), BCR (≥0.2 ng/ml) and progression to BCR or salvage treatment after detectable PSA. Life tables and Cox proportional hazards regression models were used to determine the association of BGM and risk of oncologic outcomes. RESULTS: A total of 1,082 men underwent RP for localized prostate cancer with BGM reported on surgical pathology and an undetectable postoperative PSA. BGM was present on 249 (23%) specimens. Younger age, bilateral nerve sparing surgery and robotic approach were associated with presence of BGM while malignancy at the surgical margin (MSM) was not. At 7 years after RP, 29% experienced detectable PSA and 11% had BCR. In the subgroup of men who reached detectable PSA, 79% had progression within 7 years. On multivariate Cox proportional hazards regression, BGM status was not independently associated with detectable PSA, BCR and/or progression from detectable PSA to BCR or salvage treatment. CONCLUSIONS: The presence of BGM at RP was not associated with increased risk of MSM, detectable PSA, BCR or progression after detectable PSA.

Evaluating the impact of surgical supply cost variation during partial nephrectomy on patient outcomes.

BACKGROUND: Reducing surgical supply costs can help to lower hospital expenditures. We aimed to evaluate whether variation in supply costs between urologic surgeons performing both robotic or open partial nephrectomies is associated with differential patient outcomes. METHODS: In this retrospective cohort study, we reviewed 399 consecutive robotic (n=220) and open (n=179) partial nephrectomies performed at an academic center. Surgical supply costs were determined at the institution-negotiated rate. Through retrospective review, we identified factors related to case complexity, patient comorbidity, and perioperative outcomes. Two radiologists assigned nephrometry scores to grade tumor complexity. We created univariate and multivariable models for predictors of supply costs, length of stay, and change in serum creatinine. RESULTS: Median supply cost was $3,201 [interquartile range (IQR): $2,201-3,808] for robotic partial nephrectomy and $968 (IQR: $819-1,772) for open partial nephrectomy. Mean nephrometry score was 7.0 (SD =1.7) for robotic procedures and 8.2 (SD =1.6) for open procedures. In multivariable models, the surgeon was the primary significant predictor of variation in surgical supply costs for both procedure types. In multivariable mixed-effects analysis with surgeon as a random effect, supply cost was not a significant predictor of change in serum creatinine for robotic or open procedures. Supply cost was not a statistically significant predictor of length of stay for the open procedure. Supply cost was a significant predictor of longer length of stay for the robotic procedure, however it was not a clinically meaningful change in length of stay (0.02 days per $100 in supply costs). CONCLUSIONS: Higher supply spending did not predict significantly improved patient outcomes. Variability in surgeon supply preference is the likely source of variability in supply cost. These data suggest that efforts to promote cost-effective utilization and standardization of supplies in partial nephrectomy could help reduce costs without harming patients.

Integration of electronic pathology reporting with clinical trial matching for advanced prostate cancer.

INTRODUCTION: Racial/ethnic diversity in prostate cancer (CaP) clinical trials (CTs) is essential to address CaP disparities. California Cancer Registry mandated electronic reporting (e-path) of structured data elements from pathologists diagnosing cancer thereby creating an opportunity to identify and approach patients rapidly. This study tested the utility of an online CT matching tool (called Trial Library) used in combination with e-path to improve matching of underrepresented CaP patients into CTs at time of diagnosis. METHODS: This was a nonrandomized, single-arm feasibility study among patients with a new pathologic diagnosis of high-risk CaP (Gleason Score ≥8). Eligible patients were sent recruitment materials and enrolled patients were introduced to Trial Library. RESULTS: A total of 419 case listings were assessed. Patients were excluded due to physician contraindication, not meeting baseline eligibility, or unable to be reached. Final participants (N = 52) completed a baseline survey. Among study participants, 77% were White, 10% were Black/Hispanic/Missing, and 14% were Asian. The majority of the study participants were over 65 years of age (81%) and Medicare insured (62%). Additionally, 81% of participants reported using the Internet to learn about CaP. The majority (62%) of participants reported that Trial Library increased their interest in CT participation. CONCLUSIONS: The current study demonstrated that leveraging structured e-path data reporting to a population-based cancer registry to recruit men with high risk CaP to clinical research is feasible and acceptable. We observed that e-path may be linked with an online CT matching tool, Trial Library. Future studies will prioritize recruitment from reporting facilities that serve more racially/ethnically diverse patient populations.

Neoadjuvant chemotherapy in bladder cancer: Clinical benefit observed in prospective trials computed with restricted mean survival times.

PURPOSE: Neoadjuvant chemotherapy (NAC) is the standard of care for eligible patients with cT2-4a N0 M0 bladder cancer undergoing surgical resection. The extent to which (and if) NAC increases patient survival is not clear as clinical trials and meta-analyses have generated both negative and "borderline" positive results. The novel method of calculating restricted mean survival times (RMST) may provide a more meaningful way to quantify treatment efficacy due to inherent statistical limitations of conventional hazard ratios. In this study we analyzed the survival benefit attributable to NAC in bladder cancer by calculating RMST of previously published clinical trials. MATERIALS AND METHODS: All published randomized controlled clinical trials of bladder cancer with available survival data comparing NAC plus radical cystectomy with cystectomy alone were included. RMSTs were calculated for each cohort at the 5-year and total follow-up time periods, comparing the NAC and radical cystectomy groups. Fixed effect meta-analysis of the 5-year RMSTs was then performed to calculate the net impact of NAC on overall survival. RESULTS: For 2 among 7 included trails, RMST analysis changed the statistical significance. The SWOG 8,710 trial that had previously suggested a survival benefit associated with NAC (P = 0.06) was found to have a clearer beneficial association by 5-year RMST (6.5 month benefit; P = 0.01) and total follow-up RMST (13.6 month benefit over 168 months; P = 0.04). The International Collaboration of Trialists trial that had previously suggested a survival benefit with NAC (P = 0.04) was found to have a beneficial association by total follow-up RMST (6.7 months benefit over 120 months; P = 0.04) but not 5-year RMST (P = 0.10). The interpretation of other trials did not change.  Fixed effect meta-analysis suggested a clinically significant overall survival benefit associated with NAC (3.2 months benefit over 60 months; P < 0.01). CONCLUSIONS: Evaluation of published randomized controlled trials using RMSTs strengthens the association of neoadjuvant chemotherapy with survival benefit in bladder cancer. As RMST may enable improved detection of clinical benefit when compared to conventional statistical methods, consideration should be given to RMST-based endpoints in future clinical trial design.

Statewide Price Variation for Generic Benign Prostatic Hyperplasia Medications.

OBJECTIVE: To examine the geographic and pharmacy-type variation in costs for generic benign prostatic hyperplasia (BPH) medications in order to improve drug price transparency and reduce health disparities. Medical therapy for BPH can be expensive, having significant implications for uninsured and underinsured patients. METHODS: We generated a 20% random sample of all pharmacies in Pennsylvania and queried each for the uninsured cash price of a 30-day prescription of tamsulosin 0.4mg daily, finasteride 5mg daily, oxybutynin immediate release 5mg TID and oxybutynin XL 10mg daily. Our primary objectives were to identify price variation based on pharmacy type (i.e., big chain and independent) and between geographic regions (predetermined by the Pennsylvania Health Care Cost Containment Council Database). We fit multivariable quantile regression models to test for an association between drug price and region after controlling for pharmacy type. RESULTS: Among 575 retail pharmacies contacted, 473 responded (82% response rate). The median cash price was significantly higher for big chain pharmacies than for independent pharmacies for tamsulosin ($66 vs. $15), finasteride ($68 vs. $15), oxybutynin immediate release ($49 vs. $35), and oxybutynin XL ($79 vs. $31) (all p < 0.05). When controlling for region, the median and 75th percentile price of all drugs was significantly higher for big chain pharmacies. When controlling for pharmacy type, regional variation was noted in all four drugs at the 75th percentile price and was greater for independent pharmacies. CONCLUSION: Compared to independent pharmacies, big chain pharmacies charged significantly more for generic BPH medications to uninsured patients. However, independent pharmacies demonstrated more regional variation in their pricing.

Evaluating determinants of receipt of molecular imaging in biochemical recurrent prostate cancer.

BACKGROUND: Molecular imaging with novel radiotracers is changing the treatment landscape in prostate cancer (PCa). Currently, standard of care includes either conventional and molecular imaging at time of biochemical recurrence (BCR). This study evaluated the determinants of and cost associated with utilization of molecular imaging for BCR PCa. METHODS: This is a retrospective observational cohort study among men with BCR PCa from June 2018 to May 2019. Multivariate logistic regression models were employed to analyze the primary outcome: receipt of molecular imaging (e.g. Fluciclovine PET and Prostate Specific Membrane Antigen PET) as part of diagnostic work-up for BCR PCa. Multivariate linear regression models were used to analyze the secondary outcome: overall healthcare cost within a 1-year time frame. RESULTS: The study sample included 234 patients; 79.1% White, 2.1% Black, 8.5% Asian/Pacific Islander, and 10.3% Other. The majority were 55 years or older (97.9%) and publicly insured (74.8%). Analysis indicated a one-unit reduction in PSA is associated with 1.3 times higher likelihood of receiving molecular imaging (p < 0.01). Analysis found that privately insured patients were associated with approximately $500,000 more in hospital reimbursement (p < 0.01) as compared to the publicly insured. Additionally, a one-unit increase in PSA is associated with $6254 increase in hospital reimbursement or an increase in total payments by 2.1% (p < 0.05). CONCLUSIONS: Higher PSA was associated with lower likelihood for molecular imaging and higher cost in a one-year time frame. Higher cost was also associated with private insurance, but there was no clear relationship between insurance type and imaging type.

Disparities in precision medicine-Examining germline genetic counseling and testing patterns among men with prostate cancer.

INTRODUCTION: This study sought to examine whether germline genetic counseling and testing were employed differentially among men with prostate cancer by race and/or ethnicity and other social factors. METHODS: In this retrospective analysis, all patients with prostate cancer listed as a visit diagnosis during the study period (April 2011 to August 2020) were identified from electronic health records. Patient characteristics were collected along with genetic counselor visits and germline genetic testing results in electronic health records. Multivariable analyses were performed with the primary outcome defined as the receipt of a genetic counseling visit and receipt of genetic testing. RESULTS: A total of 14,610 patients with a prostate cancer diagnosis code were identified. The majority of patients were White (72%), aged >=65 years (62.7%), English-speaking (95%), married (71.4%), and publicly insured (58.7%). A total of 667 patients completed an appointment with a genetic counselor. A total of 439 patients received germline genetic test result, of whom 403 (91.8%) had also completed an appointment with a genetic counselor. Patients that were 65 years or older (adjusted odds ratio 0.53, 95%CI 0.44-0.65) and non-English proficient (adjusted odds ratio 0.71, 95%CI 0.42-1.21) were less likely to receive genetic counseling. Receiving genetic counseling was the strongest independent predictor of receipt of genetic testing. CONCLUSIONS: The results of the current study highlight that the role of social factors in contributing to disparities in genetic counseling and testing among men with prostate cancer. These results underscore the importance of developing novel strategies to tackle contributors of observed disparities including language, age, and insurance status.

Natural language processing systems for pathology parsing in limited data environments with uncertainty estimation.

OBJECTIVE: Cancer is a leading cause of death, but much of the diagnostic information is stored as unstructured data in pathology reports. We aim to improve uncertainty estimates of machine learning-based pathology parsers and evaluate performance in low data settings. MATERIALS AND METHODS: Our data comes from the Urologic Outcomes Database at UCSF which includes 3232 annotated prostate cancer pathology reports from 2001 to 2018. We approach 17 separate information extraction tasks, involving a wide range of pathologic features. To handle the diverse range of fields, we required 2 statistical models, a document classification method for pathologic features with a small set of possible values and a token extraction method for pathologic features with a large set of values. For each model, we used isotonic calibration to improve the model's estimates of its likelihood of being correct. RESULTS: Our best document classifier method, a convolutional neural network, achieves a weighted F1 score of 0.97 averaged over 12 fields and our best extraction method achieves an accuracy of 0.93 averaged over 5 fields. The performance saturates as a function of dataset size with as few as 128 data points. Furthermore, while our document classifier methods have reliable uncertainty estimates, our extraction-based methods do not, but after isotonic calibration, expected calibration error drops to below 0.03 for all extraction fields. CONCLUSIONS: We find that when applying machine learning to pathology parsing, large datasets may not always be needed, and that calibration methods can improve the reliability of uncertainty estimates.

"Can you hear me now?": postoperative patient-initiated communication with providers.

OBJECTIVES: Surgical patients often leave the hospital with many questions and concerns after their surgery and will contact their providers to get answers. The growth of patient-provider communication (PPC) technologies allows for many new opportunities to study postoperative patient-initiated communication. We aimed to characterize a growing body of literature on postoperative patient-initiated communication. STUDY DESIGN: Review. METHODS: A scoping review methodology was used to identify 17 studies analyzing patient-initiated communication in the postoperative period and to characterize key results and areas of investigation in the literature. Patient-initiated communication in the postoperative period was defined as any communication initiated by the patient after discharge. RESULTS: The majority of studies were published between 2014 and 2018 (82.4%). Telephone calls were the most common type of medium investigated (11 studies; 64.7%), followed by secure messaging (2 studies; 11.8%). Patients most commonly initiated contact regarding study results, medications, and wounds. Common areas of investigation included communication timing and sociodemographic associations. CONCLUSIONS: As health systems adopt new technologies for PPC, understanding how and why patients initiate contact with providers postoperatively can inform efforts to strengthen PPC broadly. Moreover, research on sociodemographic variation in communication patterns after surgery can help address communication gaps that patient groups may experience. Future research can build upon this work to improve patient outcomes and increase clinic efficiency.

The Impact of the COVID-19 Pandemic on Genitourinary Cancer Care: Re-envisioning the Future.

CONTEXT: The coronavirus disease 2019 (COVID-19) pandemic necessitated rapid changes in medical practice. Many of these changes may add value to care, creating opportunities going forward. OBJECTIVE: To provide an evidence-informed, expert-derived review of genitourinary cancer care moving forward following the initial COVID-19 pandemic. EVIDENCE ACQUISITION: A collaborative narrative review was conducted using literature published through May 2020 (PubMed), which comprised three main topics: reduced in-person interactions arguing for increasing virtual and image-based care, optimisation of the delivery of care, and the effect of COVID-19 in health care facilities on decision-making by patients and their families. EVIDENCE SYNTHESIS: Patterns of care will evolve following the COVID-19 pandemic. Telemedicine, virtual care, and telemonitoring will increase and could offer broader access to multidisciplinary expertise without increasing costs. Comprehensive and integrative telehealth solutions will be necessary, and should consider patients' mental health and access differences due to socioeconomic status. Investigations and treatments will need to maximise efficiency and minimise health care interactions. Solutions such as one stop clinics, day case surgery, hypofractionated radiotherapy, and oral or less frequent drug dosing will be preferred. The pandemic necessitated a triage of those patients whose treatment should be expedited, delayed, or avoided, and may persist with severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) in circulation. Patients whose demographic characteristics are at the highest risk of complications from COVID-19 may re-evaluate the benefit of intervention for less aggressive cancers. Clinical research will need to accommodate virtual care and trial participation. Research dissemination and medical education will increasingly utilise virtual platforms, limiting in-person professional engagement; ensure data dissemination; and aim to enhance patient engagement. CONCLUSIONS: The COVID-19 pandemic will have lasting effects on the delivery of health care. These changes offer opportunities to improve access, delivery, and the value of care for patients with genitourinary cancers but raise concerns that physicians and health administrators must consider in order to ensure equitable access to care. PATIENT SUMMARY: The coronavirus disease 2019 (COVID-19) pandemic has dramatically changed the care provided to many patients with genitourinary cancers. This has necessitated a transition to telemedicine, changes in threshold or delays in many treatments, and an opportunity to reimagine patient care to maintain safety and improve value moving forward.

Rapid Utilization of Telehealth in a Comprehensive Cancer Center as a Response to COVID-19: Cross-Sectional Analysis.

BACKGROUND: The emergence of the coronavirus disease (COVID-19) pandemic in March 2020 created unprecedented challenges in the provision of scheduled ambulatory cancer care. As a result, there has been a renewed focus on video-based telehealth consultations as a means to continue ambulatory care. OBJECTIVE: The aim of this study is to analyze the change in video visit volume at the University of California, San Francisco (UCSF) Comprehensive Cancer Center in response to COVID-19 and compare patient demographics and appointment data from January 1, 2020, and in the 11 weeks after the transition to video visits. METHODS: Patient demographics and appointment data (dates, visit types, and departments) were extracted from the electronic health record reporting database. Video visits were performed using a HIPAA (Health Insurance Portability and Accountability Act)-compliant video conferencing platform with a pre-existing workflow. RESULTS: In 17 departments and divisions at the UCSF Cancer Center, 2284 video visits were performed in the 11 weeks before COVID-19 changes were implemented (mean 208, SD 75 per week) and 12,946 video visits were performed in the 11-week post-COVID-19 period (mean 1177, SD 120 per week). The proportion of video visits increased from 7%-18% to 54%-72%, between the pre- and post-COVID-19 periods without any disparity based on race/ethnicity, primary language, or payor. CONCLUSIONS: In a remarkably brief period of time, we rapidly scaled the utilization of telehealth in response to COVID-19 and maintained access to complex oncologic care at a time of social distancing.

A Nationally Representative Study of Nonindex Hospital Readmissions following Radical Prostatectomy: Implications for Bundled Payment Models.

PURPOSE: Implementing episode based payment models requires a detailed understanding of health care utilization throughout the 90-day postoperative episode. This includes nonindex hospital readmissions, which currently do not exist for patients treated with radical prostatectomy. We compared the causes, costs and predictors of index vs nonindex hospital readmissions after radical prostatectomy. MATERIALS AND METHODS: We identified patients with prostate cancer who underwent radical prostatectomy from 2010 to 2014 in the Nationwide Readmissions Database. Sociodemographic factors, hospital costs and causes of 90-day readmissions were compared between index and nonindex hospital readmissions. Multivariable regression models were used to determine whether nonindex readmissions were more costly than index readmission for several causes of readmission and also to identify predictors of nonindex readmissions. RESULTS: Of the 214,473 patients treated with radical prostatectomy 12,316 (5.7%) experienced a 90-day readmission and 4,283 (30.6%) had a nonindex readmission. Nonindex readmissions were more likely for complications which were cardiovascular specific (16.6% vs 10.3%) and nonradical prostatectomy specific (49.4% vs 32.8%, each p <0.01). On multivariable modeling readmission costs were significantly higher for nonindex vs index readmissions ($10,751 vs $10,113, p <0.01). Cardiovascular and electrolyte related nonindex readmissions ($12,995 vs $10,108, p <0.001, and $4,962 vs $3,179, p=0.01, respectively) were more expensive. Nonindex hospital readmission predictors included minimally invasive radical prostatectomy (OR 1.28, 95% CI 1.03-1.58), radical prostatectomy done at a high volume institution (OR 2.02, 95% CI 1.41-2.89) and residence in a more rural location (less than 50,000 population OR 1.68, 95% CI 1.21-2.35). CONCLUSIONS: In this nationally representative study nonindex hospital readmissions were associated with higher readmission costs, which were driven by differences in a small subset of readmissions. The benefits of undergoing radical prostatectomy at a high volume center should be carefully balanced with the increased odds of nonindex hospital readmissions and higher costs associated with such centers as regionalization continues.

Automating the Capture of Structured Pathology Data for Prostate Cancer Clinical Care and Research.

PURPOSE: Cancer pathology findings are critical for many aspects of care but are often locked away as unstructured free text. Our objective was to develop a natural language processing (NLP) system to extract prostate pathology details from postoperative pathology reports and a parallel structured data entry process for use by urologists during routine documentation care and compare accuracy when compared with manual abstraction and concordance between NLP and clinician-entered approaches. MATERIALS AND METHODS: From February 2016, clinicians used note templates with custom structured data elements (SDEs) during routine clinical care for men with prostate cancer. We also developed an NLP algorithm to parse radical prostatectomy pathology reports and extract structured data. We compared accuracy of clinician-entered SDEs and NLP-parsed data to manual abstraction as a gold standard and compared concordance (Cohen's κ) between approaches assuming no gold standard. RESULTS: There were 523 patients with NLP-extracted data, 319 with SDE data, and 555 with manually abstracted data. For Gleason scores, NLP and clinician SDE accuracy was 95.6% and 95.8%, respectively, compared with manual abstraction, with concordance of 0.93 (95% CI, 0.89 to 0.98). For margin status, extracapsular extension, and seminal vesicle invasion, stage, and lymph node status, NLP accuracy was 94.8% to 100%, SDE accuracy was 87.7% to 100%, and concordance between NLP and SDE ranged from 0.92 to 1.0. CONCLUSION: We show that a real-world deployment of an NLP algorithm to extract pathology data and structured data entry by clinicians during routine clinical care in a busy clinical practice can generate accurate data when compared with manual abstraction for some, but not all, components of a prostate pathology report.