Research
I am a health economist with an interest and, recently, a focus on applied research using human mobility data derived from smartphones.
Published papers
Does paid sick leave encourage staying at home? Evidence from the United States during a pandemic (2023)
Authors: Martin Andersen, Johanna Catherine Maclean, Michael F. Pesko, Kosali SimonRead the paper
Abstract
We study the impact of a temporary U.S. paid sick leave mandate that became effective April 1st, 2020 on self-quarantining, proxied by physical mobility behaviors gleaned from cellular devices. We study this policy using generalized difference-in-differences methods, leveraging pre-policy county-level heterogeneity in the share of workers likely eligible for paid sick leave benefits. We find that the policy leads to increased self-quarantining as proxied by staying home. We also find that COVID-19 confirmed cases decline post-policy.Texas Senate Bill 8 significantly reduced travel to abortion clinics in Texas (2023)
Authors: Martin S. Andersen, Christopher Marsicano, Mayra Pineda Torres, David SluskyRead the paper
Abstract
The Dobbs v. Jackson decision by the United States Supreme Court has rescinded the constitutional guarantee of abortion across the United States. As a result, at least 13 states have banned abortion access with unknown effects. Using “Texas” SB8 law that similarly restricted abortions in Texas, we provide insight into how individuals respond to these restrictions using aggregated and anonymized human mobility data. We find that “Texas” SB 8 law reduced mobility near abortion clinics in Texas by people who live in Texas and those who live outside the state. We also find that mobility from Texas to abortion clinics in other states increased, with notable increases in Missouri and Arkansas, two states that subsequently enacted post-Dobbs bans. These results highlight the importance of out-of-state abortion services for women living in highly restrictive states.A prospective examination of health care costs associated with posttraumatic stress disorder diagnostic status and symptom severity among veterans (2022)
Authors: Kelly L. Harper, Samantha Moshier, Stephanie Ellickson-Larew, Martin S. Andersen, Blair E. Wisco, Colin T. Mahoney, Terence M. Keane, Brian P. MarxRead the paper
Abstract
Posttraumatic stress disorder (PTSD) is associated with increased health care costs; however, most studies exploring this association use PTSD diagnostic data in administrative records, which can contain inaccurate diagnostic information and be confounded by the quantity of service use. We used a diagnostic interview to determine PTSD diagnostic status and examined associations between PTSD symptom severity and health care costs and utilization, extracted from Veteran Health Administration (VHA) administrative databases. Using a nationwide longitudinal sample of U.S. veterans with and without PTSD (N = 1,377) enrolled in VHA health care, we determined the costs and utilization of mental health and non–mental health outpatient, pharmacy, and inpatient services for 1 year following cohort enrollment. Relative to veterans without PTSD, those with PTSD had higher total health careUtilization Management in the Medicare Part D Program and Prescription Drug Utilization (2022)
Authors: Martin S. AndersenRead the paper
Abstract
Medicare Part D has significantly enhanced access to prescription drugs among Medicare beneficiaries. However, the recent rapid rise of utilization management policies in the Medicare Part D program may have adversely affected access to prescription drugs. I study the effects of expected and observed exposure to utilization management in prescription drug utilization using Medicare Part D claims data from 2009 to 2016 and an instrumental variables strategy based on the interaction of lagged health status and the set of plans available to each beneficiary. I find that the expected share of spending subject to utilization management increases the observed share, with the smallest effect for prior authorization. Increases in the expected share of drug spending subject to prior authorization increases Part D spending by $122.27 per percentage point, with almost three-quarters of this increase being paid by the Medicare program, rather than beneficiaries or plans. Comparable increases in step therapy and quantity limit exposure increase spending by $46 and decrease spending by $31, respectively. Interestingly, increased exposure to prior authorization and quantity limits increases the average price per 30-day prescription.Effects of utilization management on health outcomes: evidence from urinary tract infections and community-acquired pneumonia (2022)
Authors: Martin Andersen, Anurag PantRead the paper
Abstract
Background Utilization management policies are pervasive in the Medicare Part D program. We assess the effect of utilization management restrictions in the Medicare Part D program on the quality of care in two clinical areas – community-acquired pneumonia (CAP) and urinary tract infections (UTI).Methods In this study, we identified new cases of CAP and UTI from Medicare claims data from 2010 to 2016. We assessed the relationship between exposure to utilization management for antibiotic medications suitable for treating these conditions and adverse health outcomes, based on the Agency for Healthcare Research and Quality prevention quality indicators.Results We identified 147,526 cases of CAP and 632,407 UTI cases in our data. In these samples, the adverse event rate varied from 3.6 to 5.7%. The probability of an adverse event increased by 0.75 (p = 0.061) percentage points for each ten percentage point increase in exposure to quantity limits (one form of utilization management) among people with CAP. There was no relationship between utilization management and adverse events in the UTI cohort.Conclusions In some circumstances, exposure to utilization management policies–particularly quantity limits–may adversely affect health.Association of opioid utilization management with prescribing and overdose (2022)
Authors: Martin S. Andersen, Vincent Lorenz, Anurag Pant, Jeremy W. Bray, G. Caleb AlexanderRead the paper
Abstract
OBJECTIVES: Deaths from prescription opioids have reached epidemic levels in the United States, yet little is known about how insurers’ coverage policies may affect rates of fatal and nonfatal overdose among individuals filling an opioid prescription. STUDY DESIGN: Retrospective cohort study using 2010-2016 Medicare claims data for beneficiaries with 1 or more filled prescriptions for a Schedule II opioid. METHODS: Outcomes were opioid volume dispensed in morphine milligram equivalents (MME), number of days supplied, and number of pills dispensed on each prescription and emergency department or inpatient stay associated with an opioid overdose during a prescription or within 7 days of the end of the prescription. RESULTS: A total of 7.03 million prescriptions for Schedule II opioids were dispensed over 1.87 million Part D beneficiary-years. The 7.03 million opioid prescriptions were associated with 8.5 opioid overdoses per 10,000 prescriptions. Prior authorization was associated with larger opioid volumes per prescription (103.6 MME; 95% CI, 36.2-171.0). Step therapy was associated with a greater number of days supplied (0.62 days; 95% CI, 0.10-1.13) and more pills dispensed (6.12 pills; 95% CI, 2.17-10.1). Quantity limits were associated with smaller opioid volumes (24.3 MME; 95% CI, 12.3-36.3) and fewer pills dispensed (2.35 pills; 95% CI, 1.77-2.93). In adjusted models, beneficiaries filling an opioid requiring prior authorization experienced 3.3 fewer overdoses per 10,000 prescriptions (95% CI, 0.41-6.2). CONCLUSIONS: Opioid utilization management among these beneficiaries was associated with mixed effects on opioid prescribing, and prior authorization was associated with a decreased likelihood of subsequent overdose. Further work exploring the impact of utilization management and insurer policies is needed.College openings in the United States increase mobility and COVID-19 incidence (2022)
Authors: Martin Andersen, Ana I. Bento, Anirban Basu, Christopher R. Marsicano, Kosali I. SimonRead the paper
Abstract
School and college reopening-closure policies are considered one of the most promising non-pharmaceutical interventions for mitigating infectious diseases. Nonetheless, the effectiveness of these policies is still debated, largely due to the lack of empirical evidence on behavior during implementation. We examined U.S. college reopenings’ association with changes in human mobility within campuses and in COVID-19 incidence in the counties of the campuses over a twenty-week period around college reopenings in the Fall of 2020. We used an integrative framework, with a difference-in-differences design comparing areas with a college campus, before and after reopening, to areas without a campus and a Bayesian approach to estimate the daily reproductive number (Rt). We found that college reopenings were associated with increased campus mobility, and increased COVID-19 incidence by 4.9 cases per 100,000 (95% confidence interval [CI]: 2.9–6.9), or a 37% increase relative to the pre-period mean. This reflected our estimate of increased transmission locally after reopening. A greater increase in county COVID-19 incidence resulted from campuses that drew students from counties with high COVID-19 incidence in the weeks before reopening (χ2(2) = 8.9Impacts of state COVID-19 reopening policy on human mobility and mixing behavior (2021)
Authors: Thuy D. Nguyen, Sumedha Gupta, Martin Andersen, Ana I. Bento, Kosali I. Simon, Coady WingRead the paper
Abstract
This study quantifies the effect of the 2020 state COVID economic activity reopening policies on daily mobility and mixing behavior, adding to the economic literature on individual responses to public health policy that addresses public contagion risks. We harness cellular device signal data and the timing of reopening plans to provide an assessment of the extent to which human mobility and physical proximity in the United States respond to the reversal of state closure policies. We observe substantial increases in mixing activities, 13.56% at 4 days and 48.65% at 4 weeks, following reopening events. Echoing a theme from the literature on the 2020 closures, mobility outside the home increased on average prior to these state actions. Furthermore, the largest increases in mobility occurred in states that were early adopters of closure measures and hard-hit by the pandemic, suggesting that psychological fatigue is an important barrier to implementation of closure policies extending for prolonged periods of time.Requiring Versus Recommending Preparation Before Class: Does It Matter? (2018)
Authors: Martin S. Andersen, Dora Gicheva, Jeffrey SarbaumRead the paper
Abstract
Asking students to come to class prepared is quite common in undergraduate and graduate education. We use a quasiexperimental design to assess whether requiring undergraduate students in an introductory course to review prior to lecture the material that will be taught in class enhances their understanding of key concepts. We find that requiring rather than recommending preparation before class increases exam scores by about a quarter of a standard deviation, or roughly a third of a letter grade, for students in the second and third quartiles of the ability distribution but has little impact on very high- or low-ability students. We also estimate local average treatment effects, from which we draw a similar conclusion: reviewing the material before lecture benefits students in the middle of the ability distribution but has essentially no impact on the top and bottom quartiles.Effects of Medicare coverage for the chronically ill on health insurance, utilization, and mortality: Evidence from coverage expansions affecting people with end-stage renal disease (2018)
Authors: Martin S. AndersenRead the paper
Abstract
I study the effect of the 1973 expansions of Medicare coverage among individuals with end-stage renal disease (ESRD) on insurance coverage, health care utilization, and mortality. I find that the expansions increased insurance coverage by between 22 and 30 percentage points, in models that include trends in age, with the increase explained by Medicare coverage, and increased physician visits by 25–35 percent. These expansions also decreased mortality due to kidney disease in the under 65 population by between 0.5 and 1.0 deaths per 100,000. Lastly, I provide evidence for two mechanisms that affected mortality: an increase in access to and use of treatment, which may be due to changes in insurance coverage; and an increase in entry of dialysis clinics and transplant programs.Physician Preferences for Aggressive Treatment at the End of Life and Area-Level Health Care Spending: The Johns Hopkins Precursors Study (2017)
Authors: Joseph J. Gallo, Martin S. Andersen, Seungyoung Hwang, Lucy Meoni, Ravishankar JayadevappaRead the paper
Abstract
Objective: To determine whether physician preferences for end-of-life care were associated with variation in health care spending. Method: We studied 737 physicians who completed the life-sustaining treatment questionnaire in 1999 and were linked to end-of-life care data for the years 1999 to 2009 from Medicare-eligible beneficiaries from the Dartmouth Atlas of Health Care (in hospital-related regions [HRRs]). Using latent class analysis to group physician preferences for end-of-life treatment into most, intermediate, and least aggressive categories, we examined how physician preferences were associated with health care spending over a 7-year period. Results: When all HRRs in the nation were arrayed in quartiles by spending, the prevalence of study physicians who preferred aggressive end-of-life care was greater in the highest spending HRRs. The mean area-level intensive care unit charges per patient were estimated to be US$1,595 higher in the last 6 months of life and US$657 higher during the hospitalization in which death occurred for physicians who preferred the most aggressive treatment at the end of life, when compared with average spending. Conclusions: Physician preference for aggressive end-of-life care was correlated with area-level spending in the last 6 months of life. Policy measures intended to minimize geographic variation in health care spending should incorporate physician preferences and style.Effect of Prescription Drug Coupons on Statin Utilization and Expenditures: A Retrospective Cohort Study (2017)
Authors: Matthew Daubresse, Martin Andersen, Kevin R. Riggs, G. Caleb AlexanderRead the paper
Abstract
Importance Drug coupons are widely used, but their effects are not well understood. Objective To quantify the effect of coupons on statin use and expenditures. Design Retrospective cohort analysis of IMS Health LRx LifeLink database. Setting U.S. retail pharmacy transactions. Participants Incident statin users who initiated branded atorvastatin or rosuvastatin between June 2006 and February 2013. Main Outcomes and Measures Monthly statin utilization (pill-days of therapy), switching (filling a different statin), termination (failure to refill statin for 6 mo), and out-of-pocket and total costs. Results Of 1.1 million incident atorvastatin and rosuvastatin users, 2% used a coupon for at least one statin fill. At 1 year, compared with noncoupon users, those who used a statin coupon on their first fill were dispensed an equal number of monthly pill-days (23.7 vs 23.8), were less likely to switch statins (14.4% vs 16.3%), and were less likely to have terminated statin therapy (31.3% vs 39.2%). At 4 years, coupon users were more likely to have switched (45.5% vs 40.8%) and less likely to have terminated statin therapy (50.6% vs 61.1%) compared with noncoupon users. Those who used greater numbers of coupons were substantially less likely to switch and terminate statin therapies. Monthly out-of-pocket costs were lower among coupon than noncoupon users at 1 year ($9.7 vs $15.1), but total monthly costs were qualitatively similar ($115.5 vs $116.9). At 4 years, monthly out-of-pocket costs among coupon users remained lower ($14.3 vs $16.6) compared with noncoupon users. Sensitivity analyses supported the main results. Conclusions Coupons for branded statins are associated with higher utilization and lower rates of discontinuation and short-term switching to other statin products.On the failure of scientific research: an analysis of SBIR projects funded by the U.S. National Institutes of Health (2017)
Authors: Martin S. Andersen, Jeremy W. Bray, Albert N. LinkRead the paper
Abstract
The Small Business Innovation Research (SBIR) program is the primary source of public funding in the United States for research by small firms on new technologies, and the National Institutes of Health (NIH) is a major contributor to that funding agenda. Although previous research has explored the determinants of research success for NIH SBIR projects, little is known about the determinants of project failure. This paper provides important, new evidence on the characteristics of NIH SBIR projects that fail. Specifically, we find that firms that have a founder with a business background are less likely to have their funded projects fail. We also find, after controlling for the endogenous nature of woman-owned firms, that such firms are also less likely to fail.Constraints on Formulary Design Under the Affordable Care Act (2017)
Authors: Martin AndersenRead the paper
Abstract
I study the effect of prescription drug essential health benefits (EHB) requirements from the Affordable Care Act on prescription drug formularies of health insurance marketplace plans. The EHB regulates the number of drugs covered but leaves other dimensions (cost sharing and utilization management) of the formulary unregulated. Using data on almost all formularies in the country, I demonstrate that requiring insurers to cover one additional drug adds 0.22 drugs (3.3%) to the average formulary, mostly owing to firms increasing the number of drugs covered to comply with the EHB requirement. The EHB requirement also increases the probability that a drug is subject to utilization management and is assigned to a higher (more costly) formulary tier. My results suggest that newly covered drugs are 22.3 percentage points more likely to be subject to utilization management, compared to 36.7% for the average covered drug. Using formularies for Medicare Advantage plans, which are subject to uniform, nationwide benefit design standards, and the formulary status of newly approved drugs that do not satisfy the EHB requirement, I reject the hypotheses that consumer demand or effects on plan entry can explain my results. Copyright © 2017 John Wiley & Sons, Ltd.Impact of HIE Drug Formularies on Patient Out-of-Pocket Costs (2016)
Authors: Martin S. Andersen, Christine Buttorff, G. Caleb AlexanderRead the paper
Abstract
ABSTRACT Objectives: To assess the generosity of drug coverage in federally facilitated Health Insurance Exchanges (HIEs). Study Design: We linked publicly available benefit design data from 36 federally facilitated HIEs with drug formulary data for 10 therapeutic categories from Managed Markets Insight & Technology. We used health plans as our unit of analysis. Methods: We created a generosity index reflecting the ratio of patients’ out-of-pocket payments to the total drug price (higher values indicate less generous coverage). Because patients’ total out-of-pocket spending changes throughout the year due to deductibles and out-of-pocket maximums, we also examined generosity by varying levels of annual drug spending. Results: Platinum plans covered nearly 90% of drugs that we studied, while bronze, silver, and gold plans covered approximately 80%; results by therapeutic category were similar. Nine in 10 plans used 4- to 6-tier drug formularies, and approximately 24% of branded drugs were associated with some type of utilization management. Bronze plans were less generous than other metal levels, requiring consumers to pay about 94% of the total pharmaceutical costs prior to the deductible being met, while platinum plans required consumers to pay about 43% of such costs. As consumers’ out-of-pocket spending increased throughout the year, differences in generosity across plans narrowed. Conclusions: Benefit structure and formulary design contribute to significant variation in prescription drug coverage generosity within HIEs. These characteristics of health plans are important for individuals with high out-of-pocket costs.Modernizing Medicare’s Benefit Design and Low-Income Subsidies to Ensure Access and Affordability. (2015)
Authors: C. Schoen, K. Davis, C. Buttorff, M. AndersenRead the paper
Abstract
Insurance coverage through the traditional Medicare program is complex, fragmented, and incomplete. Beneficiaries must purchase supplemental private insurance to fill in the gaps. While impoverished beneficiaries may receive supplemental coverage through Medicaid and subsidies for prescription drugs, help is limited for people with incomes above the poverty level. This patchwork quilt leads to confusion for beneficiaries and high administrative costs, while also undermining coverage and care coordination. Most important, Medicare’s benefits fail to limit out-of-pocket costs or ensure adequate financial protection, especially for beneficiaries with low incomes and serious health problems. This brief, part of a series about Medicare’s past, present, and future, presents options for an integrated benefit for enrollees in traditional Medicare. The new benefit would not only reduce cost burdens but also could potentially strengthen the Medicare program and enhance its role in stimulating and supporting innovations throughout the health care delivery system.Comparing Employer-Sponsored And Federal Exchange Plans: Wide Variations In Cost Sharing For Prescription Drugs (2015)
Authors: Christine Buttorff, Martin S. Andersen, Kevin R. Riggs, G. Caleb AlexanderRead the paper
Abstract
Just under seven million Americans acquired private insurance through the new health insurance exchanges, or Marketplaces, in 2014. The exchange plans are required to cover essential health benefits, including prescription drugs. However, the generosity of prescription drug coverage in the plans has not been well described. Our primary objective was to examine the variability in drug coverage in the exchanges across plan types (health maintenance organization or preferred provider organization) and metal tiers (bronze, silver, gold, and platinum). Our secondary objective was to compare the exchange coverage to employer-sponsored coverage. Analyzing prescription drug benefit design data for the federally facilitated exchanges, we found wide variation in enrollees’ out-of-pocket costs for generic, preferred brand-name, nonpreferred brand-name, and specialty drugs, not only across metal tiers but also within those tiers across plan types. Compared to employer-sponsored plans, exchange plans generally had lower premiums but provided less generous drug coverage. However, for low-income enrollees who are eligible for cost-sharing subsidies, the exchange plans may be more comparable to employer-based coverage. Policies and programs to assist consumers in matching their prescription drug needs with a plan’s benefit design may improve the financial protection for the newly insured.Policy Options To Expand Medicare’s Low-Income Provisions To Improve Access And Affordability (2015)
Authors: Cathy Schoen, Christine Buttorff, Martin Andersen, Karen DavisRead the paper
Abstract
For fifty years Medicare has enhanced the health and financial security of seniors. Yet in 2014 an estimated 40 percent of low-income beneficiaries spent 20 percent or more of their incomes on out-of-pocket expenditures for premiums and medical care, while one-third were underinsured based on their out-of-pocket spending for medical care alone. These high burdens reflect Medicare’s limited benefits and restrictive income eligibility levels for supplemental Medicaid coverage. We examined the impacts of illustrative policies designed to improve beneficiaries’ financial protection and access to care by reducing Medicare premiums and cost sharing for covered benefits on a sliding scale for all beneficiaries with incomes up to 200 percent of the federal poverty level. We estimate that these policies could improve the affordability of health care for eleven million people. Designed to be aligned with the Affordable Care Act’s subsidy approach for the population younger than age sixty-five, these policies also have the potential to smooth transitions into Medicare, reduce administrative costs, and provide a more secure and equitable foundation for Medicare’s future.Heterogeneity and the effect of mental health parity mandates on the labor market (2015)
Authors: Martin S. AndersenRead the paper
Abstract
Health insurance benefit mandates are believed to have adverse effects on the labor market, but efforts to document such effects for mental health parity mandates have had limited success. I show that one reason for this failure is that the association between parity mandates and labor market outcomes vary with mental distress. Accounting for this heterogeneity, I find adverse labor market effects for non-distressed individuals, but favorable effects for moderately distressed individuals and individuals with a moderately distressed family member. On net, I conclude that the mandates are welfare increasing for moderately distressed workers and their families, but may be welfare decreasing for non-distressed individuals.Predictors of Venous Thromboembolism in Patients with Advanced Common Solid Cancers (2009)
Authors: Isaac E. Hall, Martin S. Andersen, Harlan M. Krumholz, Cary P. GrossRead the paper
Abstract
There is uncertainty about risk heterogeneity for venous thromboembolism (VTE) in older patients with advanced cancer and whether patients can be stratified according to VTE risk. We performed a retrospective cohort study of the linked Medicare-Surveillance, Epidemiology, and End Results cancer registry in older patients with advanced cancer of lung, breast, colon, prostate, or pancreas diagnosed between 1995–1999. We used survival analysis with demographics, comorbidities, and tumor characteristics/treatment as independent variables. Outcome was VTE diagnosed at least one month after cancer diagnosis. VTE rate was highest in the first year (3.4%). Compared to prostate cancer (1.4 VTEs/100 person-years), there was marked variability in VTE risk (hazard ratio (HR) for male-colon cancer 3.73 (95% CI 2.1–6.62), female-colon cancer HR 6.6 (3.83–11.38), up to female-pancreas cancer HR 21.57 (12.21–38.09). Stage IV cancer and chemotherapy resulted in higher risk (HRs 1.75 (1.44–2.12) and 1.31 (1.0–1.57), resp.). Stratifying the cohort by cancer type and stage using recursive partitioning analysis yielded five groups of VTE rates (nonlocalized prostate cancer 1.4 VTEs/100 person-years, to nonlocalized pancreatic cancer 17.4 VTEs/100 patient-years). In a high-risk population with advanced cancer, substantial variability in VTE risk exists, with notable differences according to cancer type and stage.Racial disparities in cancer therapy (2008)
Authors: Cary P. Gross, Benjamin D. Smith, Elizabeth Wolf, Martin AndersenRead the paper
Abstract
The purpose of this study was to determine whether racial disparities in cancer therapy had diminished since the time they were initially documented in the early 1990s.The authors identified a cohort of patients in the SEER-Medicare linked database who were ages 66 to 85 years and who had a primary diagnosis of colorectal, breast, lung, or prostate cancer during 1992 through 2002. The authors identified 7 stage-specific processes of cancer therapy by using Medicare claims. Candidate covariates in multivariate logistic regression included year, clinical, and sociodemographic characteristics, and physician access before cancer diagnosis.During the full study period, black patients were significantly less likely than white patients to receive therapy for cancers of the lung (surgical resection of early stage, 64.0% vs 78.5% for blacks and whites, respectively), breast (radiation after lumpectomy, 77.8% vs 85.8%), colon (adjuvant therapy for stage III, 52.1% vs 64.1%), and prostate (definitive therapy for early stage, 72.4% vs 77.2%, respectively). For both black and white patients, there was little or no improvement in the proportion of patients receiving therapy for most cancer therapies studied, and there was no decrease in the magnitude of any of these racial disparities between 1992 and 2002. Racial disparities persisted even after restricting the analysis to patients who had physician access before their diagnosis.There has been little improvement in either the overall proportion of Medicare beneficiaries receiving cancer therapies or the magnitude of racial disparity. Efforts in the last decade to mitigate cancer therapy disparities appear to have been unsuccessful. Cancer 2008. � 2008 American Cancer Society.Relation Between Medicare Screening Reimbursement and Stage at Diagnosis for Older Patients With Colon Cancer (2006)
Authors: Cary P. Gross, Martin S. Andersen, Harlan M. Krumholz, Gail J. McAvay, Deborah Proctor, Mary E. TinettiRead the paper
Abstract
Context Medicare’s reimbursement policy was changed in 1998 to provide coverage for screening colonoscopies for patients with increased colon cancer risk, and expanded further in 2001 to cover screening colonoscopies for all individuals. Objective To determine whether the Medicare reimbursement policy changes were associated with an increase in either colonoscopy use or early stage colon cancer diagnosis. Design, Setting, and Participants Patients in the Surveillance, Epidemiology, and End Results Medicare linked database who were 67 years of age and older and had a primary diagnosis of colon cancer during 1992-2002, as well as a group of Medicare beneficiaries who resided in Surveillance, Epidemiology, and End Results areas but who were not diagnosed with cancer. Main Outcome Measures Trends in colonoscopy and sigmoidoscopy use among Medicare beneficiaries without cancer were assessed using multivariate Poisson regression. Among the patients with cancer, stage was classified as early (stage I) vs all other (stages II-IV). Time was categorized as period 1 (no screening coverage, 1992-1997), period 2 (limited coverage, January 1998-June 2001), and period 3 (universal coverage, July 2001-December 2002). A multivariate logistic regression (outcome = early stage) was used to assess temporal trends in stage at diagnosis; an interaction term between tumor site and time was included. Results Colonoscopy use increased from an average rate of 285/100 000 per quarter in period 1 to 889 and 1919/100 000 per quarter in periods 2 (P{}.001) and 3 (P vs 2{}.001), respectively. During the study period, 44 924 eligible patients were diagnosed with colorectal cancer. The proportion of patients diagnosed at an early stage increased from 22.5% in period 1 to 25.5% in period 2 and 26.3% in period 3 (P{}.001 for each pairwise comparison). The changes in Medicare coverage were strongly associated with early stage at diagnosis for patients with proximal colon lesions (adjusted relative risk period 2 vs 1, 1.19; 95% confidence interval, 1.13-1.26; adjusted relative risk period 3 vs 2, 1.10; 95% confidence interval, 1.02-1.17) but weakly associated, if at all, for patients with distal colon lesions (adjusted relative risk period 2 vs 1, 1.07; 95% confidence interval, 1.01-1.13; adjusted relative risk period 3 vs 2, 0.97; 95% confidence interval, 0.90-1.05). Conclusions Expansion of Medicare reimbursement to cover colon cancer screening was associated with an increased use of colonoscopy for Medicare beneficiaries, and for those who were diagnosed with colon cancer, an increased probability of being diagnosed at an early stage. The selective effect of the coverage change on proximal colon lesions suggests that increased use of whole-colon screening modalities such as colonoscopy may have played a pivotal role.Working papers
Estimating the Impact of Temporary COVID-19 College Closures on the 2020 Census Count (2023)
Authors: Martin Andersen, Emefa Buaka, D. Sunshine Hillygus, Christopher R. Marsicano, Rylie C. MartinRead the paper
Abstract
Temporary college closures in response to the COVID-19 pandemic created an exodus of students from college towns just as the decennial census count was getting underway. We use aggregate cellular mobility data to evaluate if this population movement affected the distributional accuracy of the 2020 Census. Based on the outflow of devices in late March 2020, we estimate that counties with a college were undercounted by two percent, likely affecting Congressional apportionment.COVID-19 Restrictions Reduced Abortion Clinic Visits, Even in Blue States (2020)
Authors: Martin Andersen, Sylvia Bryan, David SluskyRead the paper