Digitalcommons@DMU (Des Moines University)
Not a member yet
3309 research outputs found
Sort by
Late-Stage Oral Cancer Detection After California and Illinois Restored Medicaid Dental Benefits
No full textObjective: Previous research found an association between California\u27s Medicaid dental coverage and oral cancer detection. However, this relationship has yet to be explored in other states or by subgroup populations. Study Design: In addition to controlling for sociodemographic and tumor characteristics, this study implemented a traditional difference-in-differences design to compare distant-stage diagnosis trends in states restoring Medicaid dental benefits (California [CA] and Illinois [IL]) with trends in states with constant Medicaid dental benefits. Setting: This retrospective, observational study analyzed oral cavity and pharynx cancer case data from The Surveillance, Epidemiology, and End Results program (2004-2017). Methods: The outcome was a binary variable indicating whether a patient was diagnosed at a distant stage. Subgroup analyses were conducted by state, race/ethnic group, sex, age, and county-level household income. Results: The sample included 109,997 adults diagnosed with cancer of the oral cavity and pharynx. Restoring Medicaid dental benefits was associated with a statistically significant 2.7%-point decline in the probability of a distant-stage oral cancer diagnosis. This estimate represented a 14% relative change from baseline rates. Results were consistent for CA and IL and by county-level median income. Estimates were significantly larger for adults under age 65, males, and adults identifying as Hispanic; non-Hispanic Black; American Indian; or Asian American or Pacific Islander. Conclusion: Restoring Medicaid dental coverage improved early detection in both CA and IL, with the greatest reductions in distant-stage diagnoses occurring in younger adults, males, and minoritized racial/ethnic groups. Future research should investigate whether earlier detection reduces oral cancer mortality disparities
Evaluating machine learning model bias and racial disparities in non-small cell lung cancer using SEER registry data
No full textBackground
Despite decades of pursuing health equity, racial and ethnic disparities persist in healthcare in America. For cancer specifically, one of the leading observed disparities is worse mortality among non-Hispanic Black patients compared to non-Hispanic White patients across the cancer care continuum. These real-world disparities are reflected in the data used to inform the decisions made to alleviate such inequities. Failing to account for inherently biased data underlying these observations could intensify racial cancer disparities and lead to misguided efforts that fail to appropriately address the real causes of health inequity. Objective
Estimate the racial/ethnic bias of machine learning models in predicting two-year survival and surgery treatment recommendation for non-small cell lung cancer (NSCLC) patients. Methods
A Cox survival model, and a LOGIT model as well as three other machine learning models for predicting surgery recommendation were trained using SEER data from NSCLC patients diagnosed from 2000-2018. Models were trained with a 70/30 train/test split (both including and excluding race/ethnicity) and evaluated using performance and fairness metrics. The effects of oversampling the training data were also evaluated. Results
The survival models show disparate impact towards non-Hispanic Black patients regardless of whether race/ethnicity is used as a predictor. The models including race/ethnicity amplified the disparities observed in the data. The exclusion of race/ethnicity as a predictor in the survival and surgery recommendation models improved fairness metrics without degrading model performance. Stratified oversampling strategies reduced disparate impact while reducing the accuracy of the model. Conclusion
NSCLC disparities are complex and multifaceted. Yet, even when accounting for age and stage at diagnosis, non-Hispanic Black patients with NSCLC are less often recommended to have surgery than non-Hispanic White patients. Machine learning models amplified the racial/ethnic disparities across the cancer care continuum (which are reflected in the data used to make model decisions). Excluding race/ethnicity lowered the bias of the models but did not affect disparate impact. Developing analytical strategies to improve fairness would in turn improve the utility of machine learning approaches analyzing population-based cancer data
Understanding the 2020 pediatric cancer deficit: Insights from the National Childhood Cancer Registry
No full textHead and Neck Cancer Incidence in the United States Before and During the COVID-19 Pandemic
No full textImportance Research about population-level changes in the incidence and stage of head and neck cancer (HNC) associated with the COVID-19 pandemic is sparse.
Objective To examine the change in localized vs advanced HNC incidence rates before and during the first year of the pandemic.
Design, Setting, and Participants In this cross-sectional study of patients in the US diagnosed with HNC from 2017 to 2020, the estimated number with cancer of the oral cavity and pharynx (floor of mouth; gum and other mouth; lip; oropharynx and tonsil; and tongue) and larynx were identified from the SEER cancer registry. Subgroup analyses were stratified by race and ethnicity, age, and sex. Data were analyzed after the latest update in April 2023.
Exposure The COVID-19 pandemic in 2020.
Main Outcomes and Measures The primary outcomes were the annual incidence rates per 100 000 people for localized HNC (includes both localized and regional stages) and advanced HNC (distant stage) and weighted average annual percentage change from 2019 to 2020. Secondary outcomes included annual percentage change for 2017 to 2018 and 2018 to 2019, which provided context for comparison.
Results An estimated 21 664 patients (15 341 [71%] male; 10 726 [50%] ≥65 years) were diagnosed with oral cavity and pharynx cancer in 2019 in the US, compared with 20 390 (4355 [70%] male; 10 393 [51%] ≥65 years) in 2020. Overall, the HNC incidence rate per 100 000 people declined from 11.6 cases in 2019 to 10.8 in 2020. The incidence rate of localized cancer declined to 8.8 cases (−7.9% [95% CI, −7.5 to −8.2]) from 2019 to 2020. The localized cancer incidence during the first year of the pandemic decreased the most among male patients (−9.3% [95% CI, −9.2 to −9.5]), Hispanic patients (−12.9% [95% CI, −12.9 to −13.0]), and individuals with larynx cancer (−14.3% [95% CI, −13.6 to −15.0]). No change in the overall incidence rate was found for advanced HNC.
Conclusions and Relevance In this cross-sectional study, the incidence of localized HNC declined during the first year of the pandemic. A subsequent increase in advanced-stage diagnoses may be observed in later years
Sex & marital differences in delayed pharyngeal cancer treatment before and after medicaid expansion
No full textObjective: In the United States, pharyngeal cancer has become the most common type of head and neck cancer, with 80 % of cases found in males. Although disparities in treatment delays have been observed in pharyngeal patients, less is known about how policies facilitate timely care. This study aimed to estimate the association between Medicaid expansion and delaying initiation of pharyngeal cancer treatment. Methodology: We extracted Surveillance, Epidemiological, End Results (SEER) case data to analyze pharyngeal cancers diagnosed between 2000 and 2018. The outcome of interest was a binary variable indicating if the patient initiated treatment two or more months after diagnosis. We conducted subgroup analyses by sex, marital status, and type of treatment received (surgery, radiation, chemotherapy, post-operative radiation, systemic therapy). We implemented the Matrix Completion algorithm to account for staggered rollout of Medicaid expansion within our difference-in-differences design. Results: Our sample included 79,433 patients diagnosed with pharynx cancer. Delayed treatment was lowest among married females receiving systemic therapy (5 %), and highest among married males and females not recommended to receive surgery (43 %). Generally, there was no association between Medicaid expansion and changes in delayed treatment. Subgroup analyses show that Medicaid expansion was associated with reduced treatment delays in unmarried females receiving systemic therapy (−4.5%-points), and married males receiving chemotherapy (Est. = −2.6%-points), radiotherapy (Est. = −3.1%-points), and married males not recommended to receive surgery (Est. = −4.6%-points). Conclusions: Given the importance of timely pharyngeal cancer treatment, health systems must identify and address the drivers of treatment delays to advance cancer equity
Eliminating Medicaid dental benefits and early-stage oral cancer diagnoses
No full textBackground: Despite the importance of regular dental visits for detecting oral cancer, millions of low-income adults lack access to dental services. In July 2009, California eliminated adult Medicaid dental benefits. We tested if this impacted oral cancer detection for Medicaid enrollees. Methods: We analyzed Surveillance, Epidemiology, and End Results-Medicaid data, which contains verified Medicaid enrollment status, to estimate a difference-in-differences model. Our design compares the change in early-stage (Stages 0–II) diagnoses before and after dropping dental benefits in California with the change in early-stage diagnoses among eight states that did not change Medicaid dental benefits. Patients were grouped by oropharyngeal cancer (OPC) and non-OPC (oral cavity cancer), type, and the length of Medicaid enrollment. We also assessed if the effect of dropping dental benefits varied by the number of dentists per capita. Results: Dropping Medicaid dental benefits was associated with a 6.5%-point decline in early-stage diagnoses of non-OPC (95% CI = −14.5, −3.2, p = 0.008). This represented a 20% relative reduction from baseline rates. The effect was highest among beneficiaries with 3 months of continuous Medicaid enrollment prior to diagnosis who resided in counties with more dentists per capita. Specifically, dropping dental coverage was associated with a 1.25%-point decline in the probability of early-stage non-OPC diagnoses for every additional dentist per 5000 population (p = 0.006). Conclusions: Eliminating Medicaid dental benefits negatively impacted early detection of cancers of the oral cavity. Continued volatility of Medicaid dental coverage and provider shortages may be further delaying oral cancer diagnoses. Alternative approaches are needed to prevent advanced stage OPC
Head and neck cancer treatment delays in 2021: Estimating distributional effects by site, surgery, and p16-positivity
No full textBackground Following decades of policies increasing access to high-quality cancer treatment, the COVID-19 pandemic upended the U.S. healthcare system. The pandemic\u27s disruption likely affected an often-overlooked dimension of quality cancer treatment: timely initiation. Timely treatment initiation is especially critical for head and neck cancer (HNC). We aimed to assess how the treatment interval (diagnosis to treatment initiation) changed in 2021 for different types of HNCs and treatment modalities. Methodology We analyzed Surveillance, Epidemiological, End Results (SEER) case data for years 2007–2021. Using ICD site codes, cancers were restricted to oropharynx, oral cavity, other pharynx, larynx. Oropharynx cancers were stratified by Human Papillomavirus (HPV) subtype based on p16-positivity codes. The outcome of interest was a variable measuring the number of days from diagnosis to treatment initiation. Cases were stratified by site and whether they received surgery as first course of treatment. To overcome validity threats from skewed treatment interval data and unobserved heterogeneity, we constructed an unconditional quantile regression model to estimate the effect of treatment in 2021 across the distribution of the treatment interval. Results 155,273 patients in SEER initiated HNC treatment between 2007 and 2021. The median treatment interval was 29 days (Interquartile Range = 2–48). Among patients not receiving surgery, 2021 was associated with delayed treatment for all sites except oral cavity. For patients receiving surgery, 2021 was only associated with delayed treatment for p16+ Oropharynx cancer. Discussion HNC patients overall, but HPV+ Oropharynx cancer patients especially, experienced treatment delays in 2021. These delays, and their consequences, warrant policymaking attention
Colorectal cancer in older adults after the USPSTF\u27s 2008 updated screening recommendation
No full textBackground: Colorectal cancer (CRC) screenings can improve detection and prevent precancerous polyps from becoming malignant tumors. In 2008, the United States Preventive Services Task Force (USPSTF) updated their policy and no longer recommended that adults over age 75 screen for CRC. We evaluated how this policy update impacted screening behaviors and CRC outcomes in older adults. Methods: We obtained data from the Behavioral Risk Factor Surveillance System to analyze blood stool and colonoscopy screening, the Surveillance, Epidemiological, End Results program to analyze CRC staging and survival, the National Association of Centralized Cancer Registries to analyze CRC incidence, and the National Center for Health Statistics to analyze mortality. With a difference-in-differences design, we compared the changes in outcome trends of the exposed group (age 75+), before and after 2008, with the changes in trends of a similar unexposed group (age 65–74). Results: There was no association between the 2008 update and blood stool tests in older adults. We did, however, find that the update was associated with a 3.0 %-point decline in the probability of older adults completing a colonoscopy within the past two years (C.I. = −4.0, −2.0). Among older adults diagnosed with CRC, the update was associated with a 1.5 %-point increase in the probability of presenting at an advanced stage (C.I. = 1.1, 1.9). Finally, the update was also associated with lower CRC incidence (Est. = −13.9 cases/100,000 population; C.I. = −22.6, −5.1) and mortality rates (Est. = −5.6 deaths/100,000 population; C.I. = −10.1, −1.1). We observed the largest associations between the policy and CRC outcomes in adults age 85+. Discussion: The USPSTF\u27s 2008 recommendation was associated with reduced colonoscopies, especially in adults over age 85. Whether this recommendation, or the 2021 updated guidance, optimizes population health by reducing the burden of CRC screening in older adults remains unknown
Incentivizing dental services in healthy behaviour Medicaid waivers
No full textObjective: In the United States, adult dental benefits are optional in the state-managed, public insurance program, Medicaid. States also have the option to adapt their Medicaid program via waivers which pair healthy behaviour incentives (HBI) with cost-sharing. These waivers have proven ineffective, but the empirical evidence has ignored differences between states. This study aims to evaluate the impact of four state\u27s HBI Medicaid waiver on dental visits among low-income adult population subject to incentives and cost-sharing requirements by the HBI waiver. Methods: Analysing biannual data from the Behavioural Risk Factor Surveillance System\u27s Oral Health module (2008–2018) with a Difference-in-Differences design, this study estimated the effect of a Healthy Behaviour Incentive waiver on the probability of visiting the dentist in the past year. The three states that implemented an HBI Waiver (Indiana, Michigan and Wisconsin) were analysed separately. Secondary outcomes included being uninsured and having all teeth extracted. Matrix Completion methods accounted for dynamic treatment and tested for non-common trends. Inference was based on randomization inference tests. Results: Only in Michigan was an HBI waiver consistently associated with a significant increase in the probability of a dental visit (Est. = 5.6%-points, p =.01). There was little convincing evidence that HBI waivers were associated with being uninsured or having all teeth extracted. Conclusions: Between 2010 and 2019, many states have implemented an HBI waiver, each with a different approach to incentivizing dental visits. These implementation differences may explain the heterogeneous effects by state. More work is needed to evaluate how Medicaid waivers impact health outcomes in low-income populations
How did states in the United States adapt their cancer control plan in response to the COVID-19 pandemic?
No full textThe COVID-19 pandemic upended the delivery of cancer services across the care continuum. By outlining specific strategies for addressing cancer in the state, cancer control plans serve a critical role during a public health emergency. This policy analysis aims to understand how states updated their cancer control plan as a response to COVID-19. All plans from 50 states and the District of Columbia were reviewed for language related to “COVID.” Among the 51 cancer plans analyzed, 7 plans met the inclusion criteria (Illinois [IL], Iowa [IA], Maine [ME], Nevada [NV], North Carolina [NC], Utah [UT], and Vermont [VT]). These seven plans adapted their cancer control plan in response to the COVID-19 pandemic across three main themes: (1) improving care across the cancer care continuum, from prevention to screening and treatment; (2) improving cancer care service delivery by expanding telehealth, addressing workforce shortages, and investing in public health systems; and (3) achieving population health equity by addressing social determinants of health. Two states only adapted their plans by prioritizing future monitoring and evaluation activities as related to the COVID-19 pandemic (ME and VT). The other five states all took different approaches to improve cancer care by adapting their service delivery and addressing social determinants of health. IL prioritized access to cancer screenings through expanding equity informed telehealth models. IA also prioritized equitable screenings as well as clinical trial participation, by addressing workforce shortages. NV focused on prevention, leveraging telehealth and specifically targeted food security and job loss resulting from the pandemic. NC-directed cancer treatment efforts by addressing workforce shortages. UT integrated telehealth and equity initiatives to combat barriers like food insecurity and social disparities. Continued policy surveillance is needed to ensure that patients receive timely, appropriate cancer care during future public health emergencies. Research evaluating whether these plan adaptations improved outcomes or advanced equity remains warranted