Trends in Use of Next-Generation Sequencing in Patients With Solid Tumors by Race and Ethnicity After Implementation of the Medicare National Coverage Determination.

Importance: In March 2018, Medicare issued a national coverage determination (NCD) for next-generation sequencing (NGS) to facilitate access to NGS testing among Medicare beneficiaries. It is unknown whether the NCD affected health equity issues for Medicare beneficiaries and the overall population. Objective: To examine the association between the Medicare NCD and NGS use by insurance types and race and ethnicity. Design, Setting, and Participants: A retrospective cohort analysis was conducted using electronic health record data derived from a real-world database. Data originated from approximately 280 cancer clinics (approximately 800 sites of care) in the US. Patients with advanced non-small cell lung cancer (aNSCLC), metastatic colorectal cancer (mCRC), metastatic breast cancer (mBC), or advanced melanoma diagnosed from January 1, 2011, through March 31, 2020, were included. Exposure: Pre- vs post-NCD period. Main Outcomes and Measures: Patients were classified by insurance type and race and ethnicity to examine patterns in NGS testing less than or equal to 60 days after diagnosis. Difference-in-differences models examined changes in average NGS testing in the pre- and post-NCD periods by race and ethnicity, and interrupted time-series analysis examined whether trends over time varied by insurance type and race and ethnicity. Results: Among 92 687 patients with aNSCLC, mCRC, mBC, or advanced melanoma, mean (SD) age was 66.6 (11.2) years, 51 582 (55.7%) were women, and 63 864 (68.9%) were Medicare beneficiaries. The largest racial and ethnic categories according to the database used and further classification were Black or African American (8605 [9.3%]) and non-Hispanic White (59 806 [64.5%]). Compared with Medicare beneficiaries, changes in pre- to post-NCD NGS testing trends were similar in commercially insured patients (odds ratio [OR], 1.03; 95% CI, 0.98-1.08; P = .25). Pre- to post-NCD NGS testing trends increased at a slower rate among patients in assistance programs (OR, 0.93; 95% CI, 0.87-0.99; P = .03) compared with Medicare beneficiaries. The rate of increase for patients receiving Medicaid was not statistically significantly different compared with those receiving Medicare (OR, 0.92; 95% CI, 0.84-1.01; P = .07). The NCD was not associated with statistically significant changes in NGS use trends by racial and ethnic groups within Medicare beneficiaries alone or across all insurance types. Compared with non-Hispanic White individuals, increases in average NGS use from the pre-NCD to post-NCD period were 14% lower (OR, 0.86; 95% CI, 0.74-0.99; P = .04) among African American and 23% lower (OR, 0.77; 95% CI, 0.62-0.96; P = .02) among Hispanic/Latino individuals; increases among Asian individuals and those with other races and ethnicities were similar. Conclusions and Relevance: The findings of this study suggest that expansion of Medicare-covered benefits may not occur equally across insurance types, thereby further widening or maintaining disparities in NGS testing. Additional efforts beyond coverage policies are needed to ensure equitable access to the benefits of precision medicine.

Association Between Medicare's National Coverage Determination and Utilization of Next-Generation Sequencing.

PURPOSE: In 2018, Medicare issued a national coverage determination (NCD) providing reimbursement for next-generation sequencing (NGS) tests for beneficiaries with advanced or metastatic cancer and no previous NGS testing. We examined the association between NCD implementation and NGS utilization trends in Medicare beneficiaries versus commercially insured patients. METHODS: This was a retrospective study of patients with advanced non-small-cell lung cancer (aNSCLC), metastatic colorectal cancer (mCRC), metastatic breast cancer (mBC), or advanced melanoma with a de novo or recurrent advanced diagnosis from January 1, 2011, through December 30, 2019, using a nationwide US electronic health record-derived deidentified database. Patients were classified by insurance and by advanced diagnosis date. NGS testing was assessed by receipt of first NGS test result ≤ 60 days of advanced diagnosis. Interrupted time series analysis assessed NGS utilization pre- and post-NCD effective date by insurance type. RESULTS: The utilization and repeat NGS testing analysis included 70,290 and 4,295 patients, respectively. Use of NGS rose from < 1% in 2011 to > 45% in Q4 2019 in aNSCLC while remaining < 20% in mBC and advanced melanoma. Among patients with aNSCLC, mCRC, or mBC, NGS testing increased post-NCD versus pre-NCD (P < .05). There was no significant difference in trends pre- and post-NCD between Medicare beneficiaries and commercially insured patients in any tumor. Repeat NGS testing was similar before the NCD (Medicare v commercial: 24.8% v 28.5%). Post-NCD, fewer Medicare beneficiaries had repeat NGS testing (27.7% v 36.0%; P < .01). CONCLUSION: Trends in NGS utilization significantly changed post-NCD, although the magnitude of change was not significantly different by insurance type, indicating private insurers may also be incorporating NCD guidance. Implementation of the NCD may have limited use of repeat NGS testing in Medicare beneficiaries.

Estimated Impact of Public and Private Sector COVID-19 Diagnostics and Treatments on US Healthcare Resource Utilization.

INTRODUCTION: Coronavirus disease 2019 (COVID-19) has imposed a considerable burden on the United States (US) health system, with particular concern over healthcare capacity constraints. METHODS: We modeled the impact of public and private sector contributions to developing diagnostic testing and treatments on COVID-19-related healthcare resource use. RESULTS: We estimated that public sector contributions led to at least 30% reductions in COVID-19-related healthcare resource utilization. Private sector contributions to expanded diagnostic testing and treatments led to further reductions in mortality (- 44%), intensive care unit (ICU) and non-ICU hospital beds (- 30% and - 28%, respectively), and ventilator use (- 29%). The combination of lower diagnostic test sensitivity and proportions of patients self-isolating may exacerbate case numbers, and policies that encourage self-isolating should be considered. CONCLUSION: While mechanisms exist to facilitate research, development, and patient access to diagnostic testing, future policies should focus on ensuring equitable patient access to both diagnostic testing and treatments that, in turn, will alleviate COVID-19-related resource constraints.

Are National Comprehensive Cancer Network Evidence Block Affordability Ratings Representative of Real-World Costs? An Evaluation of Advanced Non-Small-Cell Lung Cancer.

PURPOSE: The National Comprehensive Cancer Network (NCCN) developed the Evidence Blocks framework to assess the value of oncology regimens. This study characterizes the relationship between real-world costs and NCCN affordability ratings (ARs) for advanced non-small-cell lung cancer (aNSCLC) treatments. METHODS: Using the MarketScan and PharMetrics Plus databases, we identified patients treated between 2012 and 2017 with an aNSCLC regimen evaluated by the NCCN Evidence Blocks. We estimated adjusted mean total per-patient-per-month (PPPM) costs and drug costs for each regimen using a log-linked gamma generalized linear model. Weighted regression was used to examine the correlation between adjusted mean PPPM costs per regimen and NCCN AR. RESULTS: A total of 25,162 patients with aNSCLC (mean age, 63 years [standard deviation, 10 years]; 52% male) had identifiable regimens. Mean total PPPM cost by therapeutic class ranged from $16,824 for epidermal growth factor receptors to $41,815 for immunotherapy-based treatment. Epidermal growth factor receptor and anaplastic lymphoma kinase inhibitor treatment had lower ARs compared with generic chemotherapy. No therapy was listed as AR group 5 (least expensive). In pairwise comparisons, AR group 1 had significantly higher PPPM total costs compared with AR groups 2 and 4. There were no significant differences in PPPM total cost among AR groups 2, 3, and 4. CONCLUSION: Real-world aNSCLC treatment costs are often inconsistent with the NCCN ARs. Given that NCCN Evidence Blocks are intended to inform provider-patient discussions and other decision support resources, such as the NCCN Categories of Preference, our results suggest that the NCCN ARs require further refinement and validation.

Comparison of the performance of particle filter algorithms applied to tracking of a disease epidemic.

We present general methodology for sequential inference in nonlinear stochastic state-space models to simultaneously estimate dynamic states and fixed parameters. We show that basic particle filters may fail due to degeneracy in fixed parameter estimation and suggest the use of a kernel density approximation to the filtered distribution of the fixed parameters to allow the fixed parameters to regenerate. In addition, we show that "seemingly" uninformative uniform priors on fixed parameters can affect posterior inferences and suggest the use of priors bounded only by the support of the parameter. We show the negative impact of using multinomial resampling and suggest the use of either stratified or residual resampling within the particle filter. As a motivating example, we use a model for tracking and prediction of a disease outbreak via a syndromic surveillance system. Finally, we use this improved particle filtering methodology to relax prior assumptions on model parameters yet still provide reasonable estimates for model parameters and disease states.