Productivity loss by cancer stage in patients newly diagnosed with hepatocellular carcinoma: A claims database analysis.

BACKGROUND: New cancer diagnoses are associated with employment decrease, workplace absenteeism, and attributable costs to employers. OBJECTIVE: To estimate the workplace productivity loss in the year following a new diagnosis of early-, intermediate-, or advanced-stage hepatocellular carcinoma (HCC) in commercially insured US adults. METHODS: We conducted a retrospective cohort study using Merative MarketScan commercial claims to identify incident HCC diagnoses from 2010 to 2020. Patients were stratified into early-, intermediate-, or advanced-stage cohorts based on presence of secondary malignancy codes or first treatment received. Mean workdays lost and attributable cost in the year following a new diagnosis were calculated using the Kaplan-Meier sample averages to account for censoring. An exploratory analysis was conducted on subgroups in the early and advanced cohorts to assess productivity loss in patients with and without treatment. RESULTS: Mean workdays lost in the year following a new HCC diagnosis among the early, intermediate, and advanced cohorts was 22.6 days (95% CI = 16.0-29.8), 17.4 days (95% CI = 11.9-23.2), and 19.5 days (95% CI = 15.6-23.6), respectively. Corresponding indirect costs were $6,031(95% CI = $4,270-$7,953), $4,644 (95% CI = $3,176-$6,192), and $5,204 (95% CI = $4,163-$6,298). Early-stage patients without a liver transplant and advanced-stage patients who received systemic therapy had 19.7 (95% CI = 12.7-27.4) and 22.0 (95% CI = 16.6-27.7) mean workdays lost, respectively. CONCLUSIONS: Productivity loss varies by stage and appears to be higher in early-stage patients who receive more intensive treatments in the first year following a new HCC diagnosis.

Cost-Effectiveness of Reduced-Intensity Allogeneic Hematopoietic Cell Transplantation for Older Patients With High-Risk Myelodysplastic Syndrome: Analysis of BMT CTN 1102.

PURPOSE: BMT CTN 1102 was a phase III trial comparing reduced-intensity allogeneic hematopoietic cell transplantation (RIC alloHCT) to standard of care for persons with intermediate- or high-risk myelodysplastic syndrome (MDS). We report results of a cost-effectiveness analysis conducted alongside the clinical trial. METHODS: Three hundred eighty-four patients received HCT (n = 260) or standard of care (n = 124) according to availability of a human leukocyte antigen-matched donor. Cost-effectiveness was calculated from US commercial and Medicare perspectives over a 20-year time horizon. Health care utilization and costs were estimated using propensity score-matched cohorts of HCT recipients in the OptumLabs Data Warehouse (age 50-64 years) and Medicare (age 65 years and older). EuroQol 5 Dimension (EQ-5D) surveys of trial participants were used to derive health state utilities. RESULTS: Extrapolated 20-year overall survival for those age 50-64 years was 29% for HCT (n = 105) versus 13% for usual care (n = 44) and 31% for HCT (n = 155) versus 12% for non-HCT (n = 80) for those age 65 years and older. HCT was more effective (+2.36 quality-adjusted life-years [QALYs] for age 50-64 years and +2.92 QALYs for age 65 years and older) and more costly (+$452,242 in US dollars (USD) for age 50-64 years and +$233,214 USD for age 65 years and older) than usual care, with incremental cost-effectiveness ratios of $191,487 (USD)/QALY and $79,834 (USD)/QALY, respectively. For persons age 50-64 years, there was a 29% chance that HCT was cost-effective using a willingness-to-pay (WTP) threshold of $150K (USD)/QALY and 51% at a $200K (USD)/QALY. For persons age 65 years and older, the probability was 100% at a WTP >$150K (USD)/QALY. CONCLUSION: Among patients age 65 years and older with high-risk MDS, RIC HCT is a high-value strategy. For those age 50-64 years, HCT is a lower-value strategy but has similar cost-effectiveness to other therapies commonly used in oncology.

Using Machine Learning to Leverage Biomarker Change and Predict Colorectal Cancer Recurrence.

PURPOSE: The risk of colorectal cancer (CRC) recurrence after primary treatment varies across individuals and over time. Using patients' most up-to-date information, including carcinoembryonic antigen (CEA) biomarker profiles, to predict risk could improve personalized decision making. METHODS: We used electronic health record data from an integrated health system on a cohort of patients diagnosed with American Joint Committee on Cancer stage I-III CRC between 2008 and 2013 (N = 3,970) and monitored until recurrence or end of follow-up. We addressed missingness in recurrence outcomes and longitudinal CEA measures, and engineered CEA features using current and past biomarker values for inclusion in a risk prediction model. We used a discrete time Superlearner model to evaluate various algorithms for predicting recurrence. We evaluated the time-varying discrimination and calibration of the algorithms and assessed the role of individual predictors. RESULTS: Recurrence was documented in 448 (11.3%) patients. XGBoost with depth = 1 (XGB-D1) predicted recurrence substantially better than all other algorithms at all time points, with AUC ranging from 0.87 (95% CI, 0.86 to 0.88) at 6 months to 0.94 (95% CI, 0.92 to 0.96) at 54 months. The only variable used by XGB-D1 was 6-month change in log CEA. Predicted 1-year risk of recurrence was nearly zero for patients whose log CEA did not increase in the last 6 months, between 12.2% and 34.1% for patients whose log CEA increased between 0.10 and 0.40, and 43.6% for those with a log CEA increase >0.40. Compared with XGB, penalized regression approaches (lasso, ridge, and elastic net) performed poorly, with AUCs ranging from 0.58 to 0.69. CONCLUSION: A flexible, machine learning approach that incorporated longitudinal CEA information yielded a simple and high-performing model for predicting recurrence on the basis of 6-month change in log CEA.

Intervention Nonadherence in the TrACER (S1415CD) Study: A Pragmatic Randomized Trial of a Standardized Order Entry for CSF Prescribing.

PURPOSE: We conducted a pragmatic, cluster-randomized trial to test whether a guideline-based standing order entry (SOE) improves use of primary prophylactic CSF (PP-CSF) prescribing for patients receiving myelosuppressive chemotherapy. We investigated variability in adherence to the intervention. METHODS: We conducted a cluster-randomized trial among 32 oncology clinics from the NCI Community Oncology Research Program. Clinics were randomized 3:1 to a guideline-based PP-CSF SOE or usual care. Among SOE sites, automated orders for PP-CSF were included for regimens at high risk for febrile neutropenia (FN) and an alert not to use PP-CSF for low FN risk. A secondary 1:1 randomization was done among intervention sites to either SOE to prescribe or an alert to not prescribe PP-CSF for patients receiving intermediate risk-regimens. Providers were allowed to override the SOE. RESULTS: Overall, PP-CSF use among patients receiving high FN risk treatment was high and not different between arms; however, rates of PP-CSF use varied widely by site, ranging from 48.6% to 100%. Among those receiving low FN risk regimens, PP-CSF use was low and not different between arms; however, PP-CSF use ranged from 0% to 19.4% across sites. In the intermediate-risk substudy, PP-CSF was five-fold higher among sites randomized to SOE; however, there was considerable variability in adherence to intervention assignment: PP-CSF use ranged from 0% to 75% among sites randomized to SOE. Despite an alert to not prescribe, PP-CSF prescribing ranged from 0% to 33%. CONCLUSION: In this randomized pragmatic trial aimed at improving PP-CSF prescribing, there was substantial variability in site adherence to the intervention assignment. Although the ability to opt out of the intervention is a feature of pragmatic trials, planning to estimate nonadherence is critical to ensure adequate power.

Impact of treatment experience on patient knowledge of colony-stimulating factors among patients receiving cancer chemotherapy: evidence from S1415CD-a large pragmatic trial.

PURPOSE: Primary prophylactic granulocyte colony-stimulating factors (PP-CSFs) are prescribed alongside chemotherapy regimens that carry a significant risk of febrile neutropenia (FN). As part of S1415CD, a prospective, pragmatic trial evaluating the impact of automated orders to improve PP-CSF prescribing, we evaluated patients' baseline knowledge of PP-CSF and whether that knowledge improved following the first cycle of chemotherapy. METHODS: Adult patients with breast, colorectal, or non-small-cell lung cancer initiating chemotherapy were enrolled in S1415CD between January 2016 and April 2020. Eight questions assessing knowledge of CSF indications, risks, benefits, and out-of-pocket costs were included in a baseline survey and in a follow-up survey at the end of the first cycle of chemotherapy. Responses were stratified by the trial arm and whether chemotherapy was low, intermediate, or high FN risk. RESULTS: Of the 3605 eligible patients, 3580 (99.3%) completed the baseline survey, and 3420 (95.5%) completed the follow-up survey. At baseline, 803 (22.4%) patients responded "Don't know" to all 8 questions, and all patients averaged 2.75 correct questions. At follow-up, knowledge increased by 0.34 in the high-FN-risk group (p < 0.001) but declined for the other FN-risk groups. In multivariate analysis, receiving a high-FN-risk regimen and younger age were significantly associated with knowledge improvement. CONCLUSION: Chemotherapy patients had poor knowledge of PP-CSF that improved only modestly among recipients of high-FN-risk chemotherapy. Further efforts to inform patients about the risks, benefits, and costs of PP-CSF may be warranted, particularly for those in whom prophylaxis is indicated. TRIAL REGISTRATION: NCT02728596, April 6, 2016.

Underutilization or appropriate care? Assessing adjuvant chemotherapy use and survival in 3 heterogenous subpopulations with stage II/III colorectal cancer within a large integrated health system.

BACKGROUND: Clinical guidelines have recommended adjuvant chemotherapy (ACT) for patients with high-risk stage II colon cancer, although the survival benefit is unclear. ACT is also recommended for patients with stage III colon cancer to reduce the risk of recurrence and mortality. For stage II/III rectal cancer, however, the role of perioperative chemotherapy (PCT, adjuvant or neoadjuvant) remains controversial, resulting in substantial variation in its use in clinical practice. OBJECTIVES: To understand real-world use and predictors of ACT or PCT use and survival outcomes in 3 heterogeneous patient groups with colorectal cancer (CRC), and to inform the evidence gap between guideline-based care and clinical practice. METHODS: This retrospective cohort study included patients with an initial stage II/III CRC diagnosis between 2008 and 2013 identified from Kaiser Permanente Southern California electronic health record databases. Patients were eligible if they were aged 18-74 years at diagnosis and received primary curative surgery. We fitted mixed effects logistic regression models to evaluate predictors of ACT receipt and Cox proportional hazards models on propensity score-matched (PS-matched) samples to assess the association between ACT/PCT receipt and survival. RESULTS: We included 1,690 patients with colon cancer (stage II: 820 and stage III: 870) and 587 patients with rectal cancer (stage II: 241 and stage III: 346). We found that 65% of patients with high-risk stage II colon cancer, 15% of those with stage III colon, and 15% of those with stage II/III rectal cancer did not receive ACT/PCT. Patients with stage II colon cancer with T4 stage (odds ratio [OR] = 5.79, 95% CI = 3.33 - 10.06) and a lower comorbidity score were more likely to receive ACT (high vs low Charlson score: OR = 0.69, 95% CI = 0.55 - 0.87). Patients with stage III rectal cancer were more likely to receive PCT than those with stage II disease (OR = 7.85, 95% CI = 2.07 - 29.74). Patients with another cancer diagnosis prior to CRC diagnosis were less likely to receive PCT (OR = 0.37, 95% CI = 0.16 - 0.85). ACT/PCT use was associated with improved overall survival among patients with high-risk stage II colon cancer (PS-matched hazard ratio [HR] = 0.42, 95% CI = 0.25 - 0.70) and those with stage III CRC (stage III colon: PS-matched HR = 0.3, 95% CI = 0.25 - 0.36; stage III rectal: PS-matched HR = 0.2, 95% CI = 0.13 - 0.31). CONCLUSIONS: We found potential underuse of appropriate chemotherapy treatment in patients with high-risk stage II colon cancer and stage III CRC. Clinicians' and providers' decisions on ACT administration may not be fully guided by the risk of recurrence and 5-year survival benefits in stage II colon cancer. DISCLOSURES: This research was supported by the National Cancer Institute of the National Institutes of Health (NIH) (under R37-CA218413). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

Racial and Ethnic Bias in Risk Prediction Models for Colorectal Cancer Recurrence When Race and Ethnicity Are Omitted as Predictors.

Importance: Including race and ethnicity as a predictor in clinical risk prediction algorithms has received increased scrutiny, but there continues to be a lack of empirical studies addressing whether simply omitting race and ethnicity from the algorithms will ultimately affect decision-making for patients of minoritized racial and ethnic groups. Objective: To examine whether including race and ethnicity as a predictor in a colorectal cancer recurrence risk algorithm is associated with racial bias, defined as racial and ethnic differences in model accuracy that could potentially lead to unequal treatment. Design, Setting, and Participants: This retrospective prognostic study was conducted using data from a large integrated health care system in Southern California for patients with colorectal cancer who received primary treatment between 2008 and 2013 and follow-up until December 31, 2018. Data were analyzed from January 2021 to June 2022. Main Outcomes and Measures: Four Cox proportional hazards regression prediction models were fitted to predict time from surveillance start to cancer recurrence: (1) a race-neutral model that explicitly excluded race and ethnicity as a predictor, (2) a race-sensitive model that included race and ethnicity, (3) a model with 2-way interactions between clinical predictors and race and ethnicity, and (4) separate models by race and ethnicity. Algorithmic fairness was assessed using model calibration, discriminative ability, false-positive and false-negative rates, positive predictive value (PPV), and negative predictive value (NPV). Results: The study cohort included 4230 patients (mean [SD] age, 65.3 [12.5] years; 2034 [48.1%] female; 490 [11.6%] Asian, Hawaiian, or Pacific Islander; 554 [13.1%] Black or African American; 937 [22.1%] Hispanic; and 2249 [53.1%] non-Hispanic White). The race-neutral model had worse calibration, NPV, and false-negative rates among racial and ethnic minority subgroups than non-Hispanic White individuals (eg, false-negative rate for Hispanic patients: 12.0% [95% CI, 6.0%-18.6%]; for non-Hispanic White patients: 3.1% [95% CI, 0.8%-6.2%]). Adding race and ethnicity as a predictor improved algorithmic fairness in calibration slope, discriminative ability, PPV, and false-negative rates (eg, false-negative rate for Hispanic patients: 9.2% [95% CI, 3.9%-14.9%]; for non-Hispanic White patients: 7.9% [95% CI, 4.3%-11.9%]). Inclusion of race interaction terms or using race-stratified models did not improve model fairness, likely due to small sample sizes in subgroups. Conclusions and Relevance: In this prognostic study of the racial bias in a cancer recurrence risk algorithm, removing race and ethnicity as a predictor worsened algorithmic fairness in multiple measures, which could lead to inappropriate care recommendations for patients who belong to minoritized racial and ethnic groups. Clinical algorithm development should include evaluation of fairness criteria to understand the potential consequences of removing race and ethnicity for health inequities.

Racial Disparities in the Ascertainment of Cancer Recurrence in Electronic Health Records.

PURPOSE: There is growing interest in using computable phenotypes or proxies to identify important clinical outcomes, such as cancer recurrence, in rich electronic health records data. However, the race/ethnicity-specific accuracies of these proxies remain unclear. We examined whether the accuracy of a proxy for colorectal cancer (CRC) recurrence differed by race/ethnicity and the possible mechanisms that drove the differences. METHODS: Using data from a large integrated health care system, we identified a stratified random sample of 282 Black/African American (AA), Hispanic, and non-Hispanic White (NHW) patients with CRC who received primary treatment. Patient 5-year recurrence status was estimated using a utilization-based proxy and evaluated against the true recurrence status obtained using detailed chart review and by race/ethnicity. We used covariate-adjusted probit regression models to estimate the associations between race/ethnicity and misclassification. RESULTS: The recurrence proxy had excellent overall accuracy (positive predictive value [PPV] 89.4%; negative predictive value 96.5%; mean difference in timing 1.96 months); however, accuracy varied by race/ethnicity. Compared with NHW patients, PPV was 14.9% lower (95% CI, 2.53 to 28.6) among Hispanic patients and 4.3% lower (95% CI, -4.8 to 14.8) among Black/AA patients. The proxy disproportionately inflated the 5-year recurrence incidence for Hispanic patients by 10.6% (95% CI, 4.2 to 18.2). Compared with NHW patients, proxy recurrences for Hispanic patients were almost three times as likely to have been misclassified as positive (adjusted risk ratio 2.91 [95% CI, 1.21 to 8.31]). Higher false positives among racial/ethnic minorities may be related to higher prevalence of noncancerous lung-related problems and substantial delays in primary treatment because of insufficient patient-provider communication and abnormal treatment patterns. CONCLUSION: Using a proxy with worse accuracy among racial/ethnic minority patients to estimate population health may misdirect resources and support erroneous conclusions around treatment benefit for these patients.

Cost-Effectiveness of Unrelated Umbilical Cord Blood Transplantation versus HLA-Haploidentical Related Bone Marrow Transplantation: Evidence from BMT CTN 1101.

BMT CTN 1101 was a Phase III randomized controlled trial comparing reduced-intensity conditioning followed by double unrelated umbilical cord blood transplantation (UCBT) versus HLA-haploidentical related donor bone marrow transplantation (haplo-BMT) for patients with high-risk hematologic malignancies. Here we report the results of a parallel cost-effectiveness analysis of these 2 hematopoietic stem cell transplantation (HCT) techniques. In this study, 368 patients were randomized to unrelated UCBT (n = 186) or haplo-BMT (n = 182). We estimated healthcare utilization and costs using propensity score-matched haplo-BMT recipients from the OptumLabs Data Warehouse for trial participants age <65 years and Medicare claims for participants age ≥65 years. Weibull models were used to estimate 20-year survival. EQ-5D surveys by trial participants were used to estimate quality-adjusted life-years (QALYs). At a 5-year follow-up, survival was 42% for haplo-BMT recipients versus 36% for UCBT recipients (P = .06). Over a 20-year time horizon, haplo-BMT is expected to be more effective (+.63 QALY) and more costly (+$118,953) for persons age <65 years. For those age ≥65 years, haplo-BMT is expected to be more effective and less costly. In one-way uncertainty analyses, for persons age <65, the cost per QALY result was most sensitive to life-years and health state utilities, whereas for those age ≥65, life- years were more influential than costs and health state utilities. Compared to UCBT, haplo-BMT was moderately more cost-effective for patients age <65 years and less costly and more effective for persons age ≥65 years. Haplo-BMT is a fair value choice for commercially insured patients with high-risk leukemia and lymphoma who require HCT. For Medicare enrollees, haplo-BMT is a preferred choice when considering costs and outcomes.

A Pragmatic Cluster-Randomized Trial of a Standing Order Entry Intervention for Colony-Stimulating Factor Use Among Patients at Intermediate Risk for Febrile Neutropenia.

PURPOSE: Primary prophylactic colony-stimulating factors (PP-CSFs) are prescribed to reduce febrile neutropenia (FN) but their benefit for intermediate FN risk regimens is uncertain. Within a pragmatic, randomized trial of a standing order entry (SOE) PP-CSF intervention, we conducted a substudy to evaluate the effectiveness of SOE for patients receiving intermediate-risk regimens. METHODS: TrACER was a cluster randomized trial where practices were randomized to usual care or a guideline-based SOE intervention. In the primary study, sites were randomized 3:1 to SOE of automated PP-CSF orders for high FN risk regimens and alerts against PP-CSF use for low-risk regimens versus usual care. A secondary 1:1 randomization assigned 24 intervention sites to either SOE to prescribe or an alert to not prescribe PP-CSF for intermediate-risk regimens. Clinicians were allowed to over-ride the SOE. Patients with breast, colorectal, or non-small-cell lung cancer were enrolled. Mixed-effect logistic regression models were used to test differences between randomized sites. RESULTS: Between January 2016 and April 2020, 846 eligible patients receiving intermediate-risk regimens were registered to either SOE to prescribe (12 sites: n = 542) or an alert to not prescribe PP-CSF (12 sites: n = 304). Rates of PP-CSF use were higher among sites randomized to SOE (37.1% v 9.9%, odds ratio, 5.91; 95% CI, 1.77 to 19.70; P = .0038). Rates of FN were low and identical between arms (3.7% v 3.7%). CONCLUSION: Although implementation of a SOE intervention for PP-CSF significantly increased PP-CSF use among patients receiving first-line intermediate-risk regimens, FN rates were low and did not differ between arms. Although this guideline-informed SOE influenced prescribing, the results suggest that neither SOE nor PP-CSF provides sufficient benefit to justify their use for all patients receiving first-line intermediate-risk regimens.

Effects of a Guideline-Informed Clinical Decision Support System Intervention to Improve Colony-Stimulating Factor Prescribing: A Cluster Randomized Clinical Trial.

Importance: Colony-stimulating factors are prescribed to patients undergoing chemotherapy to reduce the risk of febrile neutropenia. Research suggests that 55% to 95% of colony-stimulating factor prescribing is inconsistent with national guidelines. Objective: To examine whether a guideline-based standing order for primary prophylactic colony-stimulating factors improves use and reduces the incidence of febrile neutropenia. Design, Setting, and Participants: This cluster randomized clinical trial, the Trial Assessing CSF Prescribing Effectiveness and Risk (TrACER), involved 32 community oncology clinics in the US. Participants were adult patients with breast, colorectal, or non-small cell lung cancer initiating cancer therapy and enrolled between January 2016 and April 2020. Data analysis was performed from July to October 2021. Interventions: Sites were randomized 3:1 to implementation of a guideline-based primary prophylactic colony-stimulating factor standing order system or usual care. Automated orders were added for high-risk regimens, and an alert not to prescribe was included for low-risk regimens. Risk was based on National Comprehensive Cancer Network guidelines. Main Outcomes and Measures: The primary outcome was to find an increase in colony-stimulating factor use among high-risk patients from 40% to 75%, a reduction in use among low-risk patients from 17% to 7%, and a 50% reduction in febrile neutropenia rates in the intervention group. Mixed model logistic regression adjusted for correlation of outcomes within a clinic. Results: A total of 2946 patients (median [IQR] age, 59.0 [50.0-67.0] years; 2233 women [77.0%]; 2292 White [79.1%]) were enrolled; 2287 were randomized to the intervention, and 659 were randomized to usual care. Colony-stimulating factor use for patients receiving high-risk regimens was high and not significantly different between groups (847 of 950 patients [89.2%] in the intervention group vs 296 of 309 patients [95.8%] in the usual care group). Among high-risk patients, febrile neutropenia rates for the intervention (58 of 947 patients [6.1%]) and usual care (13 of 308 patients [4.2%]) groups were not significantly different. The febrile neutropenia rate for patients receiving high-risk regimens not receiving colony-stimulating factors was 14.9% (17 of 114 patients). Among the 585 patients receiving low-risk regimens, colony-stimulating factor use was low and did not differ between groups (29 of 457 patients [6.3%] in the intervention group vs 7 of 128 patients [5.5%] in the usual care group). Febrile neutropenia rates did not differ between usual care (1 of 127 patients [0.8%]) and the intervention (7 of 452 patients [1.5%]) groups. Conclusions and Relevance: In this cluster randomized clinical trial, implementation of a guideline-informed standing order did not affect colony-stimulating factor use or febrile neutropenia rates in high-risk and low-risk patients. Overall, use was generally appropriate for the level of risk. Standing order interventions do not appear to be necessary or effective in the setting of prophylactic colony-stimulating factor prescribing. Trial Registration: ClinicalTrials.gov Identifier: NCT02728596.

Development and Internal Validation of a Prognostic Model of the Probability of Death or Lung Transplantation Within 2 Years for Patients With Cystic Fibrosis and FEV1 ≤ 50% Predicted.

BACKGROUND: Improved methods are needed to risk-stratify patients with cystic fibrosis (CF) and reduced FEV1. RESEARCH QUESTIONS: What are the predictors of death or lung transplantation (LTx) within 2 years among patients with CF whose FEV1 ≤ 50% predicted? Do these markers similarly predict outcomes among G551D patients taking ivacaftor since 2012? STUDY DESIGN AND METHODS: Patients with CF, age ≥ 6 years with FEV1 ≤ 50% predicted as of December 31, 2014, were identified in a data set that merged Cystic Fibrosis Foundation and United Network for Organ Sharing (UNOS) registries. The least absolute shrinkage and selection operator (LASSO) method was applied to a randomly selected training set to select important prognostic variables. Accuracy and association of the model with death or LTx with 2 years (2-year death or LTx) were validated via logistic regression on an independent test set. Sensitivity analyses explored predictors for patients with UNOS data. RESULTS: FEV1 percent predicted (OR, 1.51 for 5% decrease; 95% CI, 1.27-1.81), number of pulmonary exacerbations treated with IV antibiotics (OR, 1.35; 95% CI, 1.11-1.65), and continuous or nocturnal oxygen (OR, 3.71; 95% CI, 1.81-7.59) were significantly associated with 2-year death or LTx. Our model predicted outcomes with greater sensitivity (ratio of sensitivity, 1.26; 95% CI, 1.02-1.54), ratio of positive predictive value (1.25; 95% CI, 1.05-1.51), and ratio of negative predictive value (1.04; 95% CI, 1.01-1.07) than FEV1 < 30% predicted. Among those taking ivacaftor in 2014, only FEV1 remained associated with 2-year death or LTx. For patients with UNOS data, LASSO identified additional covariates of interest, including noninvasive ventilation use, low hemoglobin, pulmonary arterial systolic pressure, supplemental oxygen, mechanical ventilation, FEV1 percent predicted, and cardiac index. INTERPRETATION: Among individuals with CF and FEV1 ≤ 50% predicted, FEV1 percent predicted, oxygen therapy, and number of pulmonary exacerbations predicted 2-year death or LTx. Although limited by small sample size, only FEV1 remained predictive in patients receiving highly effective modulator therapy. Additional physiologic variables could improve prognostication in CF.

Assessing surveillance utilization and value in commercially insured patients with colorectal cancer.

OBJECTIVES: Real-world patterns of surveillance testing in colorectal cancer (CRC) and the effects on health and cost outcomes are largely unknown. Our objectives were to (1) assess trends in carcinoembryonic antigen (CEA) testing, CT scans, and colonoscopy utilization and (2) examine the value of CEA testing intensity by characterizing receipt of curative treatment for recurrence and measuring direct medical costs. STUDY DESIGN: Prospective cohort study. METHODS: We used an IBM MarketScan database to identify patients with a diagnosis of and treatment for CRC between 2008 and 2015. We used a negative binomial model to assess utilization of CEA testing and logistic models to assess utilization of CT scans and colonoscopies. We used a Cox proportional hazards model to assess surveillance intensity and time to curative treatment. We estimated direct medical costs using the Kaplan-Meier sample average estimator to account for censored costs. RESULTS: We identified 3197 eligible patients. The mean numbers of CEA tests, CT scans, and colonoscopies remained relatively constant in the study period, but adherence to guidelines varied by surveillance. When categorizing individuals by their CEA utilization adherence to guidelines (perfect utilizers and overutilizers), overutilizers had an HR for curative treatment of 2.11 (95% CI, 1.46-3.05) relative to perfect utilizers. Although overutilizers underwent potentially curative procedures for recurrence at higher rates compared with perfect utilizers, direct medical costs were much higher in the overutilizer group. CONCLUSIONS: Higher intensity of surveillance, beyond what is recommended by guidelines, may lead to earlier recurrence detection and subsequent treatment, but this is associated with significantly higher direct medical costs.

Successes and challenges of implementing a cancer care delivery intervention in community oncology practices: lessons learned from SWOG S1415CD.

BACKGROUND: Cancer Care Delivery (CCD) research studies often require practice-level interventions that pose challenges in the clinical trial setting. The SWOG Cancer Research Network (SWOG) conducted S1415CD, one of the first pragmatic cluster-randomized CCD trials to be implemented through the National Cancer Institute (NCI) Community Oncology Program (NCORP), to compare outcomes of primary prophylactic colony stimulating factor (PP-CSF) use for an intervention of automated PP-CSF standing orders to usual care. The introduction of new methods for study implementation created challenges and opportunities for learning that can inform the design and approach of future CCD interventions. METHODS: The order entry system intervention was administered at the site level; sites were affiliated NCORP practices that shared the same chemotherapy order system. 32 sites without existing guideline-based PP-CSF standing orders were randomized to the intervention (n = 24) or to usual care (n = 8). Sites assigned to the intervention participated in tailored training, phone calls and onboarding activities administered by research team staff and were provided with additional funding and external IT support to help them make protocol required changes to their order entry systems. RESULTS: The average length of time for intervention sites to complete reconfiguration of their order sets following randomization was 7.2 months. 14 of 24 of intervention sites met their individual patient recruitment target of 99 patients enrolled per site. CONCLUSIONS: In this paper we share seven recommendations based on lessons learned from implementation of the S1415CD intervention at NCORP community oncology practices representing diverse geographies and patient populations across the U. S. It is our hope these recommendations can be used to guide future implementation of CCD interventions in both research and community settings. TRIAL REGISTRATION: NCT02728596 , registered April 5, 2016.

A Framework for Using Real-World Data and Health Outcomes Modeling to Evaluate Machine Learning-Based Risk Prediction Models.

OBJECTIVES: We propose a framework of health outcomes modeling with dynamic decision making and real-world data (RWD) to evaluate the potential utility of novel risk prediction models in clinical practice. Lung transplant (LTx) referral decisions in cystic fibrosis offer a complex case study. METHODS: We used longitudinal RWD for a cohort of adults (n = 4247) from the Cystic Fibrosis Foundation Patient Registry to compare outcomes of an LTx referral policy based on machine learning (ML) mortality risk predictions to referral based on (1) forced expiratory volume in 1 second (FEV1) alone and (2) heterogenous usual care (UC). We then developed a patient-level simulation model to project number of patients referred for LTx and 5-year survival, accounting for transplant availability, organ allocation policy, and heterogenous treatment effects. RESULTS: Only 12% of patients (95% confidence interval 11%-13%) were referred for LTx over 5 years under UC, compared with 19% (18%-20%) under FEV1 and 20% (19%-22%) under ML. Of 309 patients who died before LTx referral under UC, 31% (27%-36%) would have been referred under FEV1 and 40% (35%-45%) would have been referred under ML. Given a fixed supply of organs, differences in referral time did not lead to significant differences in transplants, pretransplant or post-transplant deaths, or overall survival in 5 years. CONCLUSIONS: Health outcomes modeling with RWD may help to identify novel ML risk prediction models with high potential real-world clinical utility and rule out further investment in models that are unlikely to offer meaningful real-world benefits.

Racial Disparities in Access to Prostate Cancer Clinical Trials: A County-Level Analysis.

Background: African American men have a higher burden of prostate cancer compared with other populations. We sought to determine if they experience disparities in access to prostate cancer clinical trials. Methods: We created a database of all US counties by linking prostate cancer clinical trial data with county-level socioeconomic, demographic, and health-care facility data derived from several external data sources. Using this data linkage, we examined 2 potential access barriers. We investigated the relationship between the proportion of African Americans and access to cancer facilities, adjusting for county population size and other characteristics. Additionally, among counties with cancer facilities, we investigated the relationship between the proportion of African Americans and number of available prostate cancer trials per capita per year. We addressed these questions using logistic and negative binomial regression, respectively. Results: Between 2008 and 2015, 613 prostate cancer trial sites were found among 3145 US counties. Counties with a higher proportion of African Americans were less likely to have cancer facilities (adjusted odds ratio = 0.85, 95% confidence interval = 0.78 to 0.92). Among counties with cancer facilities, those with a higher proportion of African Americans had statistically significantly fewer prostate cancer trials per capita per year (rate ratio per 10% increase in African Americans = 0.90, 95% confidence interval = 0.83 to 0.96). Conclusions: Counties with higher proportions of African Americans seem less likely to have access to cancer facilities. Among counties with cancer facilities, those with higher proportions of African Americans appear to have fewer prostate cancer trials available per capita per year. Clinical trials in prostate cancer therapy should ensure adequate availability of enrollment sites in regions with high concentrations of African Americans.

A Value-of-Information Framework for Personalizing the Timing of Surveillance Testing.

BACKGROUND: Patient surveillance using repeated biomarker measurements presents an opportunity to detect and treat disease progression early. Frequent surveillance testing using biomarkers is recommended and routinely conducted in several diseases, including cancer and diabetes. However, frequent testing involves tradeoffs. Although surveillance tests provide information about current disease status, the complications and costs of frequent tests may not be justified for patients who are at low risk of progression. Predictions based on patients' earlier biomarker values may be used to inform decision making; however, predictions are uncertain, leading to decision uncertainty. METHODS: We propose the Personalized Risk-Adaptive Surveillance (PRAISE) framework, a novel method for embedding predictions into a value-of-information (VOI) framework to account for the cost of uncertainty over time and determine the time point at which collection of biomarker data would be most valuable. The proposed sequential decision-making framework is innovative in that it leverages the patient's longitudinal history, considers individual benefits and harms, and allows for dynamic tailoring of surveillance intervals by considering the uncertainty in current information and estimating the probability that new information may change treatment decisions, as well as the impact of this change on patient outcomes. RESULTS: When applied to data from cystic fibrosis patients, PRAISE lowers costs by allowing some patients to skip a visit, compared to an "always test" strategy. It does so without compromising expected survival, by recommending less frequent testing among those who are unlikely to be treated at the skipped time point. CONCLUSIONS: A VOI-based approach to patient monitoring is feasible and could be applied to several diseases to develop more cost-effective and personalized strategies for ongoing patient care. HIGHLIGHTS: In many patient-monitoring settings, the complications and costs of frequent tests are not justified for patients who are at low risk of disease progression. Predictions based on patient history may be used to individualize the timing of patient visits based on evolving risk.We propose Personalized Risk-Adaptive Surveillance (PRAISE), a novel method for personalizing the timing of surveillance testing, where prediction modeling projects the disease trajectory and a value-of-information (VOI)-based pragmatic decision-theoretic framework quantifies patient- and time-specific benefit-harm tradeoffs.A VOI-based approach to patient monitoring could be applied to several diseases to develop more personalized and cost-effective strategies for ongoing patient care.

Artificial intelligence in breast cancer screening: primary care provider preferences.

BACKGROUND: Artificial intelligence (AI) is increasingly being proposed for use in medicine, including breast cancer screening (BCS). Little is known, however, about referring primary care providers' (PCPs') preferences for this technology. METHODS: We identified the most important attributes of AI BCS for ordering PCPs using qualitative interviews: sensitivity, specificity, radiologist involvement, understandability of AI decision-making, supporting evidence, and diversity of training data. We invited US-based PCPs to participate in an internet-based experiment designed to force participants to trade off among the attributes of hypothetical AI BCS products. Responses were analyzed with random parameters logit and latent class models to assess how different attributes affect the choice to recommend AI-enhanced screening. RESULTS: Ninety-one PCPs participated. Sensitivity was most important, and most PCPs viewed radiologist participation in mammography interpretation as important. Other important attributes were specificity, understandability of AI decision-making, and diversity of data. We identified 3 classes of respondents: "Sensitivity First" (41%) found sensitivity to be more than twice as important as other attributes; "Against AI Autonomy" (24%) wanted radiologists to confirm every image; "Uncertain Trade-Offs" (35%) viewed most attributes as having similar importance. A majority (76%) accepted the use of AI in a "triage" role that would allow it to filter out likely negatives without radiologist confirmation. CONCLUSIONS AND RELEVANCE: Sensitivity was the most important attribute overall, but other key attributes should be addressed to produce clinically acceptable products. We also found that most PCPs accept the use of AI to make determinations about likely negative mammograms without radiologist confirmation.

Predictors of tyrosine kinase inhibitor adherence trajectories in patients with newly diagnosed chronic myeloid leukemia.

INTRODUCTION: Although consistent use of tyrosine kinase inhibitors (TKIs) confers significant improvements in long-term survival for individuals with chronic myeloid leukemia (CML), only 70% of CML patients are adherent to TKIs. Understanding the factors that contribute to non-adherence and establishing dynamic adherence patterns in this population are essential aspects of targeted drug monitoring and intervention strategies. METHODS: Newly diagnosed CML patients were identified in the MarketScan database and relevant covariate values extracted. Proportion of days covered (PDC) per 30-day interval was used to calculate adherence over a 12-month follow-up period. We conducted a latent profile analysis (LPA) on these PDC estimates to identify distinct, dynamic patterns of TKI adherence. Identified trajectories were grouped into four clinically relevant categories and predictors of membership in these categories were determined via multinomial logistic regression. RESULTS: Four broad adherence categories were identified from the LPA: never adherent, initially non-adherent becoming adherent, initially adherent becoming non-adherent, and stable adherent. Results from the subsequent multinomial logistic regression indicated that younger age, female sex, greater monthly financial burden, fewer comorbidities, fewer concomitant medications, year of diagnosis, higher starting dose, TKI type, and a longer duration from diagnosis to treatment were significantly associated with membership in at least one of the three non-stable adherent groups. CONCLUSION: Select sociodemographic and clinical characteristics were found to predict membership in clinically meaningful groups of longitudinal TKI adherence. These findings could have major implications for informing personalized monitoring and intervention strategies for individuals who are likely to be non-adherent.

A comparison of general, genitourinary, bowel, and sexual quality of life among long term survivors of prostate, bladder, colorectal, and lung cancer.

OBJECTIVES: Studies of local stage prostate cancer survivors suggest that treatments carry risk of persistent impotence, incontinence, and bowel dysfunction. To examine impacts of cancer type and side effects on health-related quality of life (HRQoL) in long-term cancer survivorship, we evaluated 5-year follow-up of patients with prostate cancer and compared results with a matched group of male long-term survivors of other local-stage cancers. MATERIALS AND METHODS: We examined genitourinary, bowel and sexual symptoms, and general quality of life. Matched survivors of colorectal, lung, and bladder cancers were recruited via registries in 3 different regions in the United States. Patients were surveyed 3-5 years after diagnosis with the SF-12 and EPIC to evaluate general mental and physical health-related quality of life (HRQoL) and patient function and bother. RESULTS: We analyzed responses from long-term prostate (n = 77) and bladder, colorectal, and lung cancer (n = 124) patients. In multivariate analysis, long-term local stage prostate cancer survivors had significantly higher SF-12 physical component scores but did not differ from long-term survivors of other cancers in terms of their SF-12 mental summary scores. Prostate survivors had similar mental, urinary, bowel, and sexual HRQoL compared to long-term survivors of other local stage cancers. CONCLUSION: Long-term general and prostate-specific HRQoL was similar between local stage prostate and bladder, colorectal, and lung patients with cancer. Future research focusing on factors other than initial treatment and the cancer type per se may provide more meaningful information regarding factors that predict disparities on HRQoL among longer-term survivors of early stage male cancers.