Abstract
PURPOSE
Medicare Advantage (MA) beneficiaries newly diagnosed with cancer may want to switch from MA into traditional Medicare and purchase supplemental insurance (Medigap), which provides important financial protections. However, beneficiaries in states without Medigap consumer protections may be denied or effectively denied (through high premiums) Medigap, potentially trapping them in MA. This study examined whether MA beneficiaries with newly diagnosed cancer were more likely to switch into traditional Medicare in states with stronger Medigap consumer protections.
METHODS
This retrospective study analyzed Medicare encounter data for beneficiaries age 65 years and older, continuously enrolled in MA from 2018 to 2019, and received a new cancer diagnosis in 2019. The primary exposure was state-level Medigap protections: no protections, community rating only (some protections), or both community rating and guaranteed issue (strongest protections). The outcome of interest was beneficiary-level disenrollment from MA into traditional Medicare in 2020.
RESULTS
Among 426,248 MA beneficiaries newly diagnosed with cancer, 2.1% switched into traditional Medicare in states with both community rating and guaranteed issue, 1.1% switched in states with community rating only, and 0.8% switched in states with no Medigap protections. The odds of disenrollment were higher in states with both community rating and guaranteed issue (odds ratio [OR], 2.73 [95% CI, 2.53 to 2.94]; P < .001) and states with community rating only (OR, 1.48 [95% CI, 1.30 to 1.68]; P < .001), compared with states with no protections.
CONCLUSION
MA beneficiaries with newly diagnosed cancers had higher rates of disenrollment from MA into traditional Medicare when state-level Medigap consumer protections were present. Beneficiaries appear more likely to switch if states make it financially easier to obtain Medigap.
INTRODUCTION
Cancer is a significant health burden on the Medicare population, affecting nearly 15% of beneficiaries.1 Although Medicare provides nearly universal coverage for this population, Medicare beneficiaries experience differential access to affordable, timely, and high-quality cancer care depending on their type of Medicare coverage. Beneficiaries have the option to enroll in traditional Medicare or Medicare Advantage (MA), a privately administered alternative to traditional Medicare that now covers 54% of Medicare enrollees.2,3 Although MA plans often offer lower out-of-pocket costs and additional benefits, such as gym memberships,4 they also use provider networks and utilization management strategies, such as previous authorization,4 which may impede access to specialists or advanced care at leading cancer hospitals.5-8 Such barriers can be particularly consequential for patients who require complex and timely cancer treatment.9-12
CONTEXT
Key Objective
Are Medicare Advantage (MA) beneficiaries with newly diagnosed cancer more likely to switch into traditional Medicare in states with stronger supplemental insurance (Medigap) consumer protections?
Knowledge Generated
For MA beneficiaries with newly diagnosed cancer, greater state Medigap protection levels were associated with an increased likelihood of switching into traditional Medicare. Switching rates were highest in states with both community rating and guaranteed issue (strongest protections) and lowest in states with no protections.
Relevance
Our findings suggest that after a new cancer diagnosis, MA beneficiaries may seek to switch into traditional Medicare but face a constrained choice architecture because of the lack of Medigap protections. Policymakers seeking to support beneficiaries' health during serious illness should consider reforms that ensure both access to health care and affordable supplemental coverage when medical needs become more complex.
Given these challenges, some MA beneficiaries with cancer may consider switching to traditional Medicare, which offers access to a broader range of providers and fewer utilization restrictions.9-12 These factors are particularly important for beneficiaries seeking complex, specialized oncology care or participation in clinical trials.9,13 However, traditional Medicare typically comes with higher out-of-pocket costs unless these beneficiaries purchase supplemental Medigap insurance.14,15 For example, Medigap Plan G, the most popular and comprehensive Medigap option, provides nearly full financial coverage for Part A and B out-of-pocket expenses.16 Therefore, beneficiaries with newly diagnosed cancer who are enrolled in traditional Medicare with Medigap may face lower out-of-pocket costs and fewer managed care restrictions compared with those enrolled in MA. Because of these health care access and financial benefits, MA beneficiaries with newly diagnosed cancer may be motivated to switch into traditional Medicare and purchase Medigap.
Federal and state-level policies may make Medigap inaccessible to some beneficiaries, making it harder for beneficiaries to leave MA. Under federal law, Medicare beneficiaries age 65 years and older are provided a one-time, 6-month Medigap open enrollment period when beneficiaries first enroll in Medicare.17 During this period, Medigap insurers must price Medigap premiums at community rates (ie, insurers must charge all policyholders the same premium regardless of age or health status) and provide guaranteed issue (ie, insurers must offer coverage regardless of age or health status).16 After this initial 6-month period, Medicare beneficiaries who did not initially purchase a Medigap plan may be subject to medical underwriting, meaning insurers can individually rate the beneficiary (ie, vary premiums on the basis of the beneficiary's age and health status, or outright choose to not sell insurance to the beneficiary).16 However, some states provide additional consumer protections beyond federal standards.16 Currently, eight states have laws that require Medigap insurers to provide community rates, and four of these eight states also require guaranteed issue (Appendix Fig A1, online only).16
MA beneficiaries who live in the 42 states with no Medigap consumer protection policies may feel trapped in MA, since they can be denied or effectively denied (through high premiums) Medigap in traditional Medicare. This concept of the MA trap has been identified in the popular press and been empirically observed in past literature (Liu, et al 2024).18,46 We examined whether MA beneficiaries with a new cancer diagnosis in states with Medigap consumer protections (community rating only or both community rating and guaranteed issue) were more likely to disenroll from MA into traditional Medicare compared with those in states without Medigap consumer protection policies (Fig 1).
FIG 1.
State Medigap protections and the odds of disenrollment from MA into traditional Medicare by cancer type. The ORs and 95% CIs were estimated using univariate logistic regression with disenrollment from MA to traditional Medicare as the dependent variable and different levels of state Medigap protections as the independent variables. MA, Medicare Advantage; OR, odds ratio.
METHODS
Data Sources and Study Population
In this retrospective cohort study, we identified MA beneficiaries with a new cancer diagnosis in 2019 and assessed their enrollment status in 2020 to determine whether they remained in MA or switched into traditional Medicare. We then evaluated whether rates of disenrollment from MA into traditional Medicare were higher in states with Medigap consumer protection policies. This study was approved by the Johns Hopkins Medicine Institutional Review Board (approval number 11318) and followed the Strengthening the Reporting of Observational Studies in Epidemiology reporting guidelines.
To identify MA beneficiaries with cancer, we used 100% MA carrier, inpatient, and outpatient encounter data for 2019, which capture MA beneficiary encounters at physician offices, overnight hospital stays, and outpatient hospital visits, respectively. Cancer diagnoses among these MA beneficiaries were identified using the International Statistical Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes.19 We focused on the 11 most common cancers in the United States, which represent approximately 75% of new cancer diagnoses annually: breast, prostate, lung and bronchus, colorectal, melanoma, bladder, non-Hodgkin lymphoma, kidney and renal pelvis, uterine, leukemia, and pancreas.20 Appendix Table A1 provides a list of ICD-10-CM codes used to identify each type of cancer included in our study.
We used the 2018-2020 Medicare Beneficiary Summary Files to determine beneficiary enrollment in Medicare. We included beneficiaries continuously enrolled in MA from January 1, 2018, to December 31, 2019, and observed their subsequent enrollment status in 2020 (MA v traditional Medicare). We restricted our cohort to beneficiaries who met the following criteria: (1) alive from 2018 to 2020; (2) age 65 years or older in 2018, as not all states allow those younger than 65 years to purchase Medigap21; (3) not dually eligible for Medicare and Medicaid and not enrolled in Medicare Part D low-income subsidy, as these beneficiaries have access to special enrollment periods not available to the general Medicare population22; (4) living in the 50 states and Washington, DC; (5) not disabled; and (6) did not move to a new state during the time period.
New Cancer Diagnosis
We identified new cancer diagnoses in 2019 using the 2018 and 2019 MA carrier, inpatient, and outpatient encounter data. Beneficiaries were defined as having a new cancer diagnosis if they had one or more claims across any of the three encounter files with an ICD-10-CM diagnosis code for cancer after a 365-day washout period in which no previous cancer diagnoses were recorded. The presence of any type of cancer during the 365-day washout period would disqualify a subsequent cancer diagnosis from being considered a new cancer diagnosis. For instance, if a beneficiary received a lung cancer diagnosis but had a breast cancer diagnosis code within the previous 365 days, then the lung cancer would not be considered a new cancer diagnosis.
Key Variables
The main outcome was a binary variable describing whether the beneficiary stayed in MA or switched into traditional Medicare in 2020. The main independent variable was a categorical, ordinal variable describing the Medigap state type for the state where the beneficiary resided: (1) no Medigap consumer protection policies (42 states and District of Columbia), (2) community rating only (Arkansas, Minnesota, Vermont, and Washington), and (3) both community rating and guaranteed issue (Connecticut, Maine, Massachusetts, and New York).16 States without Medigap consumer protection policies served as the control group (Appendix Fig A1).
Statistical Analysis
We described beneficiary characteristics and compared them by state Medigap protection policies. We then performed unadjusted logistic regressions to model the association between state Medigap protection policies and beneficiary disenrollment from MA into traditional Medicare. To improve the interpretability of the logistic regression models, we estimated the average marginal effects (AMEs). Next, we examined the association between state Medigap protection policies and beneficiary disenrollment from MA into traditional Medicare for specific cancer types.
We performed sensitivity analyses using three alternative model specifications: (1) 365-day washout period restricted to beneficiaries with only one specific cancer, (2) 180-day washout period, and (3) no washout period. Each model specification was examined using both unadjusted univariate and adjusted multivariate analyses. The multivariate analysis included the following covariates: age, sex, race/ethnicity, comorbid medical conditions using the Elixhauser Comorbidity Index,23 and zip code–level data on high-school graduation rate and percentage below the Federal Poverty Level as measured by the 2016-2020 American Community Survey.
Statistical significance was defined as P < .01. Analysis was conducted in Stata version 17 (StataCorp LLC) and R version 4.3.1.
RESULTS
Sample Characteristics
Among the 426,248 MA beneficiaries with a new cancer diagnosis in 2019 (Appendix Fig A2), 358,885 (84.2%) resided in states with no Medigap protections, 24,024 (5.6%) in states with community rating only, and 43,339 (10.2%) in states with both community rating and guaranteed issue. The most common new cancer diagnoses in the overall sample were prostate (32.3%), breast (24.1%), and colorectal (8.5%; Table 1). 94.9% of beneficiaries had one cancer and only one beneficiary (<0.1%) had six cancers (Appendix Fig A3).
TABLE 1.
Characteristics of Beneficiaries Enrolled in MA With a New Cancer Diagnosis in 2019
| Characteristic | No Protections (n = 358,885) | Community Rating (n = 24,024) | Community Rating + Guaranteed Issue (n = 43,339) | Overall (N = 426,248) |
|---|---|---|---|---|
| Enrollment status in 2020, No. (%) | ||||
| Stayed in MA | 356,148 (99.2) | 23,754 (98.9) | 42,449 (97.9) | 422,351 (99.1) |
| Switched to traditional Medicare | 2,737 (0.8) | 270 (1.1) | 890 (2.1) | 3,897 (0.9) |
| Cancer type,a No. (%) | ||||
| Breast | 86,777 (24.2) | 5,701 (23.7) | 10,362 (23.9) | 102,840 (24.1) |
| Prostate | 115,999 (32.3) | 7,630 (31.8) | 13,895 (32.1) | 137,524 (32.3) |
| Lung and bronchus | 23,682 (6.6) | 1,707 (7.1) | 3,119 (7.2) | 28,508 (6.7) |
| Colorectal | 30,927 (8.6) | 1,871 (7.8) | 3,377 (7.8) | 36,175 (8.5) |
| Melanoma | 23,243 (6.5) | 1,574 (6.6) | 2,577 (5.9) | 27,394 (6.4) |
| Bladder | 27,586 (7.7) | 1,797 (7.5) | 4,139 (9.6) | 33,522 (7.9) |
| Non-Hodgkin lymphoma | 20,532 (5.7) | 1,621 (6.7) | 2,398 (5.5) | 24,551 (5.8) |
| Kidney and renal pelvis | 16,532 (4.6) | 979 (4.1) | 1,861 (4.3) | 19,372 (4.5) |
| Uterine | 12,118 (3.4) | 857 (3.6) | 1,646 (3.8) | 14,621 (3.4) |
| Leukemia | 16,624 (4.6) | 1,343 (5.6) | 1,839 (4.2) | 19,806 (4.6) |
| Pancreas | 3,880 (1.1) | 238 (1.0) | 503 (1.2) | 4,621 (1.1) |
| Age, years, No. (%) | ||||
| ≤70 | 71,043 (19.8) | 5,136 (21.4) | 8,815 (20.3) | 84,994 (19.9) |
| 71-75 | 105,241 (29.3) | 7,334 (30.5) | 12,769 (29.5) | 125,344 (29.4) |
| 76-80 | 88,949 (24.8) | 5,869 (24.4) | 10,410 (24.0) | 105,228 (24.7) |
| 81-85 | 54,949 (15.3) | 3,408 (14.2) | 6,660 (15.4) | 65,017 (15.3) |
| ≥86 | 38,703 (10.8) | 2,277 (9.5) | 4,685 (10.8) | 45,665 (10.7) |
| Sex, No. (%) | ||||
| Female | 161,758 (45.1) | 10,899 (45.4) | 19,837 (45.8) | 192,494 (45.2) |
| Race/ethnicity, No. (%) | ||||
| Asian/Pacific Islander | 7,938 (2.2) | 342 (1.4) | 923 (2.1) | 9,203 (2.2) |
| Black or African American | 35,553 (9.9) | 535 (2.2) | 4,020 (9.3) | 40,108 (9.4) |
| Hispanic | 21,715 (6.1) | 193 (0.8) | 1,753 (4.0) | 23,661 (5.6) |
| Non-Hispanic White | 285,498 (79.6) | 22,298 (92.8) | 35,104 (81.0) | 342,900 (80.4) |
| Other/unknown | 8,181 (2.3) | 656 (2.7) | 1,539 (3.6) | 10,376 (2.4) |
| Elixhauser Comorbidity Index, mean (SD) | 5.1 (3.2) | 4.2 (2.8) | 4.6 (2.9) | 5.0 (3.1) |
| Percent of high school graduation, mean (SD) | 86.4 (9.9) | 91.3 (6.1) | 86.0 (10.1) | 86.6 (9.8) |
| Percent of below federal poverty line, mean (SD) | 8.8 (5.6) | 7.1 (4.1) | 8.9 (6.6) | 8.7 (5.6) |
NOTE. Beneficiaries were required to be enrolled in MA from January 1, 2018, for the 365-day washout period.
Abbreviations: MA, Medicare Advantage; SD, standard deviation.
Beneficiaries can be diagnosed with more than one type of cancer, and cancer types may add up to more than 100%.
In the overall MA population, 1.3% switched into traditional Medicare in states with both community rating and guaranteed issue, 1.1% in community rating only states, and 0.9% in states with no protections (Appendix Table A2). Among MA beneficiaries with a new cancer diagnosis, 2.1% switched into traditional Medicare in states with both community rating and guaranteed issue, 1.1% in community rating only states, and 0.8% in states with no protections (Table 1). The difference in disenrollment rates between MA beneficiaries with cancer and the overall MA population was most pronounced in states with both community rating and guaranteed issue (2.1% v 1.3%, respectively). Among the 3,897 MA beneficiaries with newly diagnosed cancer who switched into traditional Medicare, 22.8% resided in states with both community rating and guaranteed issue, although those beneficiaries account for only 10.2% of beneficiaries with cancer (Appendix Table A3).
Disenrollment From MA by State-Level Medigap Protections
In the unadjusted analysis, compared with states with no protections, the odds of disenrollment were 173% higher (odds ratio [OR], 2.73 [95% CI, 2.53 to 2.94]; P < .001) in states with both community rating and guaranteed issue. In states with both community rating and guaranteed issue, this corresponds to MA beneficiaries with cancer being 1.29 percentage points more likely to switch into traditional Medicare (AME, 1.29; 95% CI, 1.15 to 1.43; P < .001) than those in states with no protections, which had a switching rate of 0.8%. In states with community rating only, the odds of disenrollment were 48% higher (OR, 1.48 [95% CI, 1.30 to 1.68]; P < .001) compared with states with no Medigap protections. Comparing the two Medigap protection levels, the odds of disenrollment were 84% higher (OR, 1.84 [95% CI, 1.61 to 2.12]; P < .001) in states with both community rating and guaranteed issue compared with states with community rating only. We found similar results in the adjusted multivariate analysis (Table 2).
TABLE 2.
Association Between State Medigap Protection and Disenrollment From Medicare Advantage Into Traditional Medicare After a New Cancer Diagnosis
| Variable | Unadjusted | Adjusted | ||
|---|---|---|---|---|
| OR | AME | OR | AME | |
| Community rating v no protections | 1.48 (1.30 to 1.68); P < .001 | 0.36 (0.23 to 0.50); P < .001 | 1.53 (1.35 to 1.74); P < .001 | 0.40 (0.26 to 0.54); P < .001 |
| Community rating + guaranteed issue v no protections | 2.73 (2.53 to 2.94); P < .001 | 1.29 (1.15 to 1.43); P < .001 | 2.78 (2.58 to 3.00); P < .001 | 1.33 (1.19 to 1.46); P < .001 |
| Community rating + guaranteed issue v community rating | 1.84 (1.61 to 2.12); P < .001 | 0.93 (0.74 to 1.12); P < .001 | 1.81 (1.58 to 2.08); P < .001 | 0.92 (0.73 to 1.12); P < .001 |
| Age, years (v <71) | ||||
| 71-75 | 0.87 (0.80 to 0.95); P = .002 | –0.14 (–0.23 to –0.05); P = .002 | ||
| 76-80 | 0.78 (0.71 to 0.85); P < .001 | –0.25 (–0.34 to –0.16); P < .001 | ||
| 81-85 | 0.66 (0.59 to 0.74); P < .001 | –0.38 (–0.47 to –0.28); P < .001 | ||
| >85 | 0.71 (0.63 to 0.80); P < .001 | –0.32 (–0.43 to –0.21); P < .001 | ||
| Sex (v male) | ||||
| Female | 0.96 (0.90 to 1.03); P = .265 | –0.03 (–0.09 to 0.03); P = .264 | ||
| Race (v non-Hispanic White) | ||||
| Asian | 1.53 (1.27 to 1.83); P < .001 | 0.47 (0.23 to 0.71); P < .001 | ||
| Black | 1.13 (1.01 to 1.25); P = .033 | 0.11 (0.00 to 0.22); P = .042 | ||
| Hispanic | 0.94 (0.80 to 1.09); P = .401 | –0.06 (–0.19 to 0.07); P = .388 | ||
| Other/unknown | 1.03 (0.84 to 1.24); P = .788 | 0.02 (–0.15 to 0.20); P = .791 | ||
| Elixhauser Comorbidity Index | 1.04 (1.03 to 1.06); P < .001 | 0.04 (0.03 to 0.05); P < .001 | ||
| Percent of high school graduation | 1.00 (1.00 to 1.01); P = .465 | 0.00 (–0.00 to 0.01); P = .465 | ||
| Percent of below federal poverty line | 1.00 (1.00 to 1.01); P = .571 | 0.00 (0.00 to 0.01); P = .571 | ||
NOTE. Results are presented as OR with 95% CI and AME with 95% CI.
Abbreviations: AME, average marginal effect; OR, odds ratio.
Cancer-Specific Disenrollment Patterns
Disenrollment patterns varied considerably by cancer diagnosis. In states with both community rating and guaranteed issue versus no protections, beneficiaries with lung and bronchus cancer had the highest odds of disenrollment (OR, 4.87 [95% CI, 3.88 to 6.12]; P < .001), while those with breast cancer had the lowest (OR, 2.34 [95% CI, 1.97 to 2.77]; P < .001) (Fig 1 and Appendix Table A4).
Sensitivity Analyses
Across all specifications, the association between state Medigap protections and disenrollment was similar. Reducing the washout period from 365 to 180 days or eliminating it entirely slightly attenuated the effect sizes. Restricting analysis to beneficiaries without multiple cancer diagnoses also slightly reduced the magnitude of the effect size. The adjusted models consistently demonstrated slightly larger effect sizes compared with unadjusted models (Fig 2).
FIG 2.
Odds of disenrollment from MA into traditional Medicare for states with both community rating and guaranteed issue versus no protections by model specification. The ORs and 95% CIs for the unadjusted models were estimated using logistic regression with disenrollment from MA to traditional Medicare as the dependent variable and community rating + guaranteed issue versus no state Medigap protections as the independent variable. The adjusted models also included beneficiaries' sex, age, race/ethnicity, comorbidity severity as measured by the Elixhauser Comorbidity Index, and zip code–level data on high-school graduation percentage and percentage below the Federal Poverty Level. MA, Medicare Advantage; OR, odds ratio.
DISCUSSION
We find that state-level Medigap policies are associated with MA disenrollment among beneficiaries with newly diagnosed cancer. Those who live in states with more protection policies (community rating only and both community rating and guaranteed issue) are more likely to disenroll from MA compared with states with no protections. These findings suggest that access to Medigap is an important factor influencing whether MA beneficiaries with newly diagnosed cancer switch into traditional Medicare.
Within traditional Medicare, Medigap plans offer critical financial safeguards for traditional Medicare beneficiaries, especially beneficiaries who require complex care. For instance, Medigap Plan G offers nearly complete coverage of Part A and B out-of-pocket costs.16,24 However, the high cost of Medigap premiums likely presents a significant financial obstacle for many Medicare beneficiaries. In 2023, the average annual Medigap premium was nearly $2,000 in US dollars (USD), and the median annual income for Medicare beneficiaries was around $36,000 USD.25,26 Considering that just under a quarter of beneficiaries in traditional Medicare are already struggling to pay their traditional Medicare premiums,27 it is likely that Medigap is unaffordable for many Medicare beneficiaries with cancer, especially those living in states with no protections who face medical underwriting.
Although our study did not survey Medicare beneficiaries with newly diagnosed cancer, there are some notable characteristics of MA plans that may explain why these beneficiaries may be switching from MA into traditional Medicare. First, many MA plans have limited access to in-network physicians, surgeons, and hospitals relative to traditional Medicare. Notably, 41% of MA plans in counties with a National Cancer Institute (NCI)–designated cancer center did not include these centers in their network in 2016.9 NCI-designated cancer centers are recognized for their excellence in both research and clinical care,13 tend to have superior survival rates,28,29 and have greater access to clinical trials.9 MA beneficiaries are also less likely to have access to or use high-volume cancer surgery hospitals,10,12 top-ranked cancer hospitals,11 and teaching hospitals.9 As a result, MA beneficiaries with cancer are likely to have higher mortality rates than those enrolled in traditional Medicare.12 Thus, for beneficiaries requiring access to cancer care outside of their existing MA network, there is a strong incentive to switch their coverage to traditional Medicare with Medigap to access the care they need.
Another significant concern for MA beneficiaries with cancer is the impact of previous authorization policies. MA plans increasingly rely on previous authorization policies to manage aspects of cancer care, including medications, surgeries, and imaging.30 Among MA insurers, hematology and oncology medications composed the highest proportion of Part B medication costs as well as the second highest proportion of overall Part B spending costs subject to previous authorization.31 Although previous authorization in oncology may reduce low-value care, a growing body of research suggests that the administrative burden of previous authorizations may result in delays in cancer care,32-34 which are associated with higher mortality.35 Thus, beneficiaries with cancer who are enrolled in MA plans may face more delays in care because of previous authorization than those enrolled in traditional Medicare, which has minimal previous authorization requirements.5-7,36 Therefore, access to affordable Medigap coverage is an important factor for MA beneficiaries seeking relief from previous authorization policies that may be disrupting their care.
Our results reveal substantial heterogeneity in MA disenrollment by cancer type. In states with both community rating and guaranteed issue, beneficiaries with lung and bronchus cancers had almost five times higher odds of switching from MA, while those with other types of cancers had at least two times higher odds of switching. This heterogeneity in switching rates may be, in part, due to the initial costs associated with different types of cancer care. Indeed, in states with both community rating and guaranteed issue, we observe that MA beneficiaries diagnosed with cancers with higher average first-year treatment costs were more likely to disenroll from MA into traditional Medicare compared with those diagnosed with cancers that have lower average first-year treatment costs (Appendix Fig A4).37 This further strengthens our hypothesis that the financial burden of cancer care may be a factor driving MA beneficiaries with cancer to switch into traditional Medicare and seek Medigap coverage. These findings align with existing literature, which finds that MA beneficiaries with more complex and costly health needs are more likely to switch into traditional Medicare.18,38-40
The variation in disenrollment likelihood by cancer type may also be attributed to differences in the complexity of treatment. Lung and pancreatic cancers, which had among the highest odds of disenrollment from MA in states with both community rating and guaranteed issue, typically necessitate a treatment combination of surgery, radiation therapy, and/or chemotherapy.41,42 Thus, treatment for these cancers often requires access to multiple oncology subspecialists, who might not always be available within MA networks.9-12 Thus, for these beneficiaries requiring more complex cancer care, Medigap access is crucial, as coverage through traditional Medicare provides access to the care they need.
Conversely, beneficiaries with cancer types that typically require less complex treatment regimens were less likely to disenroll from MA in states with both community rating and guaranteed issue compared with states with no protections. For instance, melanoma has a survival rate upward of 90% and can often be treated with surgery alone.20,43 For prostate cancer, most patients present with localized disease, for which active surveillance is typically the preferred approach.44,45 The less complex nature of these treatments may make it easier to find adequate care within MA networks, which reduces the incentive to switch into traditional Medicare and thereby minimizes the importance of Medigap coverage. Although this paper provides empirical evidence on MA disenrollment by state-level Medigap policies, future work (both quantitative and qualitative) could further examine disenrollment mechanisms by specific cancer types.
Our study has several limitations. First, we are unable to determine the specific reasons explaining why MA beneficiaries switched into traditional Medicare. The absence of these data prevents us from identifying whether factors such as the inability to access preferred providers or services and prescription drug denials because of previous authorizations played a role in disenrollment decisions. Second, although we observed an association between new cancer diagnoses and MA disenrollment, we cannot establish a causal relationship. Third, we did not possess individual-level data on Medigap denials, approvals, and premium costs. To approximate Medigap accessibility, we used state-level Medigap policies as a proxy measure. Data at the individual level would have enabled us to more accurately estimate whether Medigap accessibility affected disenrollment from MA into traditional Medicare.
In conclusion, Medicare beneficiaries with newly diagnosed cancer are more likely to disenroll from MA into traditional Medicare in states providing stronger Medigap consumer protections. The increased likelihood of switching in states with stronger Medigap consumer protection policies suggests that beneficiaries with new cancer in states without such protections may face barriers in accessing affordable Medigap coverage, which limits their ability to disenroll from MA. Thus, policymakers should engage in efforts to improve health care coverage options to Medicare beneficiaries with cancer, such as improving Medigap accessibility, strengthening MA regulations on provider networks, and reforming MA previous authorization processes.
APPENDIX
TABLE A1.
ICD-10-CM Prefixes Used to Identify Cancer Types
| Cancer Type | ICD-10-CM Prefixes |
|---|---|
| Breast | C50 |
| Prostate | C61 |
| Lung and bronchus | C34 |
| Colorectal | C18, C19, C20, C21 |
| Melanoma | C43 |
| Bladder | C67 |
| Non-Hodgkin lymphoma | C82, C83, C84, C85 |
| Kidney and renal pelvis | C64, C65 |
| Uterine | C54, C55 |
| Leukemia | C91, C92, C93, C94, C95 |
| Pancreas | C25 |
Abbreviation: ICD-10-CM, International Statistical Classification of Diseases, Tenth Revision, Clinical Modification.
TABLE A2.
Characteristics of All Beneficiaries Enrolled in MA in 2019
| Characteristic | No Protections (No. = 9,618,449) | Community Rating (No. = 581,815) | Community Rating + Guaranteed Issue (No. = 1,045,079) |
|---|---|---|---|
| Enrollment status in 2020, No. (%) | |||
| Stayed in MA | 9,529,446 (99.1) | 575,577 (98.9) | 1,031,582 (98.7) |
| Switched to traditional Medicare | 89,003 (0.9) | 6,238 (1.1) | 13,497 (1.3) |
| Age, Years, No. (%) | |||
| <71 | 2,898,125 (30.1) | 189,196 (32.5) | 322,496 (30.9) |
| 71-75 | 2,761,422 (28.7) | 172,757 (29.7) | 301,101 (28.8) |
| 76-80 | 1,952,814 (20.3) | 113,545 (19.5) | 205,833 (19.7) |
| 81-85 | 1,147,673 (11.9) | 60,913 (10.5) | 121,340 (11.6) |
| >85 | 858,415 (8.9) | 45,404 (7.8) | 94,309 (9.0) |
| Sex, No. (%) | |||
| Female | 5,475,065 (56.9) | 326,033 (56.0) | 591,313 (56.6) |
| Race and ethnicity, No. (%) | |||
| Asian/Pacific Islander | 306,506 (3.2) | 13,911 (2.4) | 27,011 (2.6) |
| Black or African American | 875,555 (9.1) | 14,237 (2.4) | 85,086 (8.1) |
| Hispanic | 713,723 (7.4) | 7,733 (1.3) | 49,369 (4.7) |
| Non-Hispanic White | 7,473,448 (77.7) | 528,084 (90.8) | 840,629 (80.4) |
| Other/unknown | 249,217 (2.6) | 17,850 (3.1) | 42,984 (4.1) |
| Elixhauser Comorbidity Index, mean (SD) | 86.0 (10.3) | 91.0 (6.4) | 2.9 (2.6) |
| Percent of high school graduation, mean (SD) | 8.9 (5.7) | 7.3 (4.2) | 86.0 (10.2) |
| Percent of below federal poverty line, mean (SD) | 3.1 (2.8) | 2.5 (2.5) | 8.8 (6.7) |
Abbreviations: MA, Medicare Advantage; SD, standard deviation.
TABLE A3.
Rates of Disenrollment From MA Into Traditional Medicare by State Medigap Protections
| State Medigap Protections | Number of Beneficiaries Switching From MA Into Traditional Medicare (%) | Total Number of Beneficiaries (%) |
|---|---|---|
| No protections | 2,737 (70.2) | 358,885 (84.2) |
| Community rating | 270 (6.9) | 24,024 (5.6) |
| Community rating + guaranteed issue | 890 (22.8) | 43,339 (10.2) |
| Total | 3,897 (100) | 426,248 (100) |
Abbreviation: MA, Medicare Advantage.
TABLE A4.
Unadjusted Association Between State Medigap Protections and Disenrollment From MA Into Traditional Medicare After a New Cancer Diagnosis by Cancer Type
| Cancer Type | Variable | OR | AME |
|---|---|---|---|
| All cancers | Community rating v no protections | 1.48 (1.30 to 1.68); P < .001 | 0.36 (0.22 to 0.50); P < .001 |
| Community rating + guaranteed issue v no protections | 2.73 (2.53 to 2.94); P < .001 | 1.29 (1.15 to 1.43); P < .001 | |
| Community rating + guaranteed issue v community rating | 1.84 (1.61 to 2.12); P < .001 | 0.93 (0.74 to 1.12); P < .001 | |
| Breast | Community rating v no protections | 1.53 (1.18 to 1.98); P = .001 | 0.38 (0.10 to 0.66); P = .007 |
| Community rating + guaranteed issue v no protections | 2.34 (1.97 to 2.77); P < .001 | 0.95 (0.70 to 1.21); P < .001 | |
| Community rating + guaranteed issue v community rating | 1.53 (1.14 to 2.04); P = .004 | 0.57 (0.21 to 0.94); P = .002 | |
| Prostate | Community rating v no protections | 1.39 (1.11 to 1.74); P = .005 | 0.30 (0.06 to 0.53); P = .014 |
| Community rating + guaranteed issue v no protections | 2.44 (2.12 to 2.80); P < .001 | 1.09 (0.86 to 1.33); P < .001 | |
| Community rating + guaranteed issue v community rating | 1.76 (1.37 to 2.25); P < .001 | 0.8 (0.47 to 1.12); P < .001 | |
| Lung and bronchus | Community rating v no protections | 1.69 (1.10 to 2.59); P = .016 | 0.57 (0.00 to 1.14); P = .050 |
| Community rating + guaranteed issue v no protections | 4.87 (3.88 to 6.12); P < .001 | 3.11 (2.41 to 3.80); P < .001 | |
| Community rating + guaranteed issue v community rating | 2.88 (1.85 to 4.48); P < .001 | 2.54 (1.66 to 3.42); P < .001 | |
| Colorectal | Community rating v no protections | 1.56 (1.01 to 2.42); P = .047 | 0.42 (–0.08 to 0.92); P = .099 |
| Community rating + guaranteed issue v no protections | 3.22 (2.50 to 4.16); P < .001 | 1.64 (1.12 to 2.17); P < .001 | |
| Community rating + guaranteed issue v community rating | 2.07 (1.28 to 3.32); P = .003 | 1.22 (0.51 to 1.93); P = .001 | |
| Melanoma | Community rating v no protections | 1.81 (1.14 to 2.89); P = .013 | 0.57 (0.00 to 1.13); P = .049 |
| Community rating + guaranteed issue v no protections | 2.44 (1.75 to 3.42); P < .001 | 1.00 (0.49 to 1.51); P < .001 | |
| Community rating + guaranteed issue v community rating | 1.35 (0.79 to 2.30); P = .270 | 0.44 (–0.31 to 1.18); P = .251 | |
| Bladder | Community rating v no protections | 1.47 (0.93 to 2.34); P = .099 | 0.36 (–0.14 to 0.85); P = .160 |
| Community rating + guaranteed issue v no protections | 2.78 (2.16 to 3.58); P < .001 | 1.32 (0.87 to 1.77); P < .001 | |
| Community rating + guaranteed issue v community rating | 1.89 (1.16 to 3.08); P = .011 | 0.96 (0.31 to 1.62); P = .004 | |
| Non-Hodgkin lymphoma | Community rating v no protections | 1.19 (0.72 to 1.95); P = .505 | 0.16 (–0.35 to 0.67); P = .535 |
| Community rating + guaranteed issue v no protections | 3.02 (2.26 to 4.03); P < .001 | 1.74 (1.09 to 2.39); P < .001 | |
| Community rating + guaranteed issue v community rating | 2.55 (1.48 to 4.37); P < .001 | 1.58 (0.77 to 2.39); P < .001 | |
| Kidney and renal pelvis | Community rating v no protections | 1.54 (0.83 to 2.87); P = .172 | 0.39 (–0.28 to 1.06); P = .254 |
| Community rating + guaranteed issue v no protections | 3.90 (2.81 to 5.42); P < .001 | 2.06 (1.30 to 2.82); P < .001 | |
| Community rating + guaranteed issue v community rating | 2.53 (1.31 to 4.87); P = .005 | 1.67 (0.67 to 2.67); P = .001 | |
| Uterine | Community rating v no protections | 1.21 (0.56 to 2.62); P = .632 | 0.14 (–0.48 to 0.76); P = .658 |
| Community rating + guaranteed issue v no protections | 3.10 (2.07 to 4.63); P < .001 | 1.39 (0.69 to 2.09); P < .001 | |
| Community rating + guaranteed issue v community rating | 2.56 (1.13 to 5.80); P = .024 | 1.25 (0.33 to 2.16); P = .007 | |
| Leukemia | Community rating v no protections | 1.50 (0.88 to 2.58); P = .138 | 0.37 (–0.21 to 0.95); P = .208 |
| Community rating + guaranteed issue v no protections | 2.88 (2.01 to 4.15); P < .001 | 1.37 (0.70 to 2.05); P < .001 | |
| Community rating + guaranteed issue v community rating | 1.92 (1.05 to 3.49); P = .033 | 1.00 (0.14 to 1.87); P = .023 | |
| Pancreas | Community rating v no protections | 1.15 (0.41 to 3.19); P = .793 | 0.21 (–1.46 to 1.89); P = .805 |
| Community rating + guaranteed issue v no protections | 3.21 (1.96 to 5.26); P < .001 | 3.10 (1.24 to 4.97); P = .001 | |
| Community rating + guaranteed issue v community rating | 2.80 (0.96 to 8.2); P = .060 | 2.89 (0.44 to 5.34); P = .021 |
NOTE. Results are presented as OR (95% CI) and AME (95% CI). The logistic regression ORs and marginal effects were estimated using a univariate logistic regression with disenrollment from MA to traditional Medicare as the dependent variable and different levels of state Medigap protections as the explanatory variable.
Abbreviations: AME, average marginal effect; MA, Medicare Advantage; OR, odds ratio.
FIG A1.
Map of state-level Medigap protection policies.
FIG A2.
Flowchart of the included sample. LIS, Medicare Part D Low-Income Subsidy; MA, Medicare Advantage; MBSF, Master Beneficiary Summary File; SES, socioeconomic status.
FIG A3.
Distribution of number of cancers.
FIG A4.
Initial cost of cancer treatment and the percent of MA beneficiaries disenrolling from MA in states with both community rating + guaranteed issue. Using data from states with both community rating and guaranteed issue, univariate linear regression was used to estimate the relationship between the initial cost of treatment (defined as the first 12 months of costs after diagnosis) and the percentage of MA beneficiaries who disenrolled from MA into traditional Medicare. The results of the regression find that an increase in the initial cost of treatment by $25,714 USD (95% CI, $13,932 USD to $37,496 USD; P < .001) is associated with a 1 percentage point increase in the probability of disenrolling from MA into traditional Medicare. MA, Medicare Advantage; USD, US dollars.
Stacy Cooper
Consulting or Advisory Role: Amgen, Jazz Pharmaceuticals
Ravi Gupta
Consulting or Advisory Role: Sunday Health
Angela Liu
Consulting or Advisory Role: Robbins Geller Rudman & Dowd LLP
Expert Testimony: Robbins Geller Rudman & Dowd LLP
No other potential conflicts of interest were reported.
DISCLAIMER
The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
SUPPORT
Supported by Arnold Ventures (grant G146501) and the Johns Hopkins School of Medicine Dean's Fund for Summer Research.
AUTHOR CONTRIBUTIONS
Conception and design: Brendon Lee, David Pittman, Gerard F. Anderson, Angela Liu
Financial support: Gerard F. Anderson
Administrative support: Gerard F. Anderson, Angela Liu
Provision of study materials or patients: Gerard F. Anderson
Collection and assembly of data: Brendon Lee, David Pittman, Gerard F. Anderson, Angela Liu
Data analysis and interpretation: All authors
Manuscript writing: All authors
Final approval of manuscript: All authors
Accountable for all aspects of the work: All authors
AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
State Medigap Policies and Disenrollment From Medicare Advantage Among Beneficiaries With Cancer
The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/op/authors/author-center.
Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).
Stacy Cooper
Consulting or Advisory Role: Amgen, Jazz Pharmaceuticals
Ravi Gupta
Consulting or Advisory Role: Sunday Health
Angela Liu
Consulting or Advisory Role: Robbins Geller Rudman & Dowd LLP
Expert Testimony: Robbins Geller Rudman & Dowd LLP
No other potential conflicts of interest were reported.
REFERENCES
- 1.Lam MB, Burke LG, Orav EJ, et al. : Proportion of patients with cancer among high-cost Medicare beneficiaries: Who they are and what drives their spending. Healthcare 6:46-51, 2018 [DOI] [PubMed] [Google Scholar]
- 2.Biniek JF, Freed M, Damico A, et al. : Half of all eligible Medicare beneficiaries are now enrolled in private Medicare Advantage plans. KFF, 2023. https://www.kff.org/policy-watch/half-of-all-eligible-medicare-beneficiaries-are-now-enrolled-in-private-medicare-advantage-plans/
- 3.Trish E, Valdez S, Ginsburg PB, et al. : Substantial growth in Medicare Advantage and implications for reform. Health Aff 42:246-251, 2023 [DOI] [PubMed] [Google Scholar]
- 4.Neuman T, Freed M, Biniek JF: 10 reasons why Medicare Advantage enrollment is growing and why it matters. KFF, 2024. https://www.kff.org/medicare/issue-brief/10-reasons-why-medicare-advantage-enrollment-is-growing-and-why-it-matters/
- 5.Jacobson G, Leonard F, Sciupac E, et al. : What do Medicare beneficiaries value about their coverage?: Findings from the Commonwealth Fund 2024 value of Medicare survey. Commonwealth Fund, 2024
- 6.Medicare Advantage appeal outcomes and audit findings raise concerns about service and payment denials. U.S. Department of Health and Human Services Office of Inspector General, 2018. https://oig.hhs.gov/reports/all/2018/medicare-advantage-appeal-outcomes-and-audit-findings-raise-concerns-about-service-and-payment-denials/
- 7.Some Medicare Advantage organization denials of prior authorization requests raise concerns about beneficiary access to medically necessary care. U.S. Department of Health and Human Services Office of Inspector General, 2022. https://oig.hhs.gov/reports/all/2022/some-medicare-advantage-organization-denials-of-prior-authorization-requests-raise-concerns-about-beneficiary-access-to-medically-necessary-care/
- 8.Medicare Advantage: Continued monitoring and implementing GAO recommendations could improve oversight. U. S. Government Accountability Office,2022. https://www.gao.gov/products/gao-22-106026
- 9.Jacobson G, Trilling A, Neuman T, et al. : Medicare Advantage hospital networks: How much do they vary? Kaiser Family Foundation, 2016. https://www.kff.org/medicare/report/medicare-advantage-hospital-networks-how-much-do-they-vary/
- 10.Raoof M, Jacobson G, Fong Y: Medicare Advantage networks and access to high-volume cancer surgery hospitals. Ann Surg 274:e315-e319, 2021 [DOI] [PubMed] [Google Scholar]
- 11.Kim D, Meyers DJ, Rahman M, et al. : Comparison of the use of the top-ranked cancer hospitals between Medicare Advantage and traditional Medicare. Am J Manag Care 27:e355-e360, 2021 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Raoof M, Ituarte PHG, Haye S, et al. : Medicare Advantage: A disadvantage for complex cancer surgery patients. J Clin Oncol 41:1239-1249, 2023 [DOI] [PubMed] [Google Scholar]
- 13.NCI-designated cancer centers. National Cancer Institute, 2012. https://www.cancer.gov/research/infrastructure/cancer-centers
- 14.Costs. Centers for Medicare & Medicaid Services. https://www.medicare.gov/basics/costs/medicare-costs [PubMed]
- 15.Ochieng N, Cubanski J, Neuman T: A snapshot of sources of coverage among Medicare beneficiaries. KFF, 2024. https://www.kff.org/medicare/issue-brief/a-snapshot-of-sources-of-coverage-among-medicare-beneficiaries/
- 16.Boccuti C, Jacobson G, Orgera K, et al. : Medigap enrollment and consumer protections vary across states. KFF, 2018. https://www.kff.org/medicare/issue-brief/medigap-enrollment-and-consumer-protections-vary-across-states/
- 17.Malloy ML: Medigap: Background and Statistics. Congressional Research Service, 2023 [Google Scholar]
- 18.Liu A, Pittman D, Anderson G, et al. : Medigap guaranteed issue associated with Medicare Advantage disenrollment for beneficiaries administered a part B drug. Health Aff Scholar 2:qxae136, 2024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.World Health Organization : ICD-10: International Statistical Classification of Diseases and Related Health Problems: Tenth Revision. World Health Organization, 2004 [PubMed] [Google Scholar]
- 20.SEER cancer statistics factsheets: Common cancer sites. National Cancer Institute. https://seer.cancer.gov/statfacts/html/common.html
- 21.When can I buy a Medigap policy? Centers for Medicare & Medicaid Services. https://www.medicare.gov/health-drug-plans/medigap/ready-to-buy/when
- 22.Special enrollment periods. Centers for Medicare & Medicaid Services. https://www.medicare.gov/basics/get-started-with-medicare/get-more-coverage/joining-a-plan/special-enrollment-periods
- 23.Austin SR, Wong YN, Uzzo RG, et al. : Why summary comorbidity measures such as the Charlson Comorbidity Index and Elixhauser score work. Med Care 53:e65-e72, 2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Compare Medigap plan benefits. Centers for Medicare & Medicaid Services. https://www.medicare.gov/health-drug-plans/medigap/basics/compare-plan-benefits
- 25.Freed M, Ochieng N, Cubanski J, et al. : Key facts about Medigap enrollment and premiums for Medicare beneficiaries. KFF, 2024. https://www.kff.org/medicare/issue-brief/key-facts-about-medigap-enrollment-and-premiums-for-medicare-beneficiaries/
- 26.Cottrill A, Cubanski J, Neuman T, et al. : Income and assets of Medicare beneficiaries in 2023. KFF. https://www.kff.org/medicare/issue-brief/income-and-assets-of-medicare-beneficiaries-in-2023/
- 27.Leonard F, Jacobson G, Collins SR, et al. : Medicare’s Affordability Problem: A Look at the Cost Burdens Faced by Older Enrollees—Findings from the Commonwealth Fund Biennial Health Insurance Survey, 2022. The Commonwealth Fund, 2023 [Google Scholar]
- 28.Shulman LN, Palis BE, McCabe R, et al. : Survival as a quality metric of cancer care: Use of the National Cancer Data Base to assess hospital performance. JCO Oncol Pract 14:e59-e72, 2018 [DOI] [PubMed] [Google Scholar]
- 29.Wolfson JA, Sun CL, Wyatt LP, et al. : Impact of care at comprehensive cancer centers on outcome: Results from a population-based study. Cancer 121:3885-3893, 2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Lin NU, Bichkoff H, Hassett MJ: Increasing burden of prior authorizations in the delivery of oncology care in the United States. JCO Oncol Pract 14:525-528, 2018 [DOI] [PubMed] [Google Scholar]
- 31.Gupta R, Fein J, Newhouse JP, et al. : Comparison of prior authorization across insurers: Cross sectional evidence from Medicare Advantage. BMJ 384:e077797, 2024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Kyle MA, Keating NL: Prior authorization and association with delayed or discontinued prescription fills. J Clin Oncol 42:951-960, 2024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.American Society of Clinical Oncology : The state of cancer care in America, 2015: A report by the American Society of Clinical Oncology. JCO Oncol Pract 11:79-113, 2015 [DOI] [PubMed] [Google Scholar]
- 34.Agarwal A, Freedman RA, Goicuria F, et al. : Prior authorization for medications in a breast oncology practice: Navigation of a complex process. JCO Oncol Pract 13:e273-e282, 2017 [DOI] [PubMed] [Google Scholar]
- 35.Cone EB, Marchese M, Paciotti M, et al. : Assessment of time-to-treatment initiation and survival in a cohort of patients with common cancers. JAMA Netw Open 3:e2030072, 2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Cunningham-Cook M: The Medicare Advantage trap. The American Prospect. https://prospect.org/api/content/82f7ede4-8e34-11ee-9676-12163087a831/
- 37.Mariotto AB, Enewold L, Zhao J, et al. : Medical care costs associated with cancer survivorship in the United States. Cancer Epidemiol Biomarkers Prev 29:1304-1312, 2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Rahman M, Keohane L, Trivedi AN, et al. : High-cost patients had substantial rates of leaving Medicare Advantage and joining traditional Medicare. Health Aff 34:1675-1681, 2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Meyers DJ, Belanger E, Joyce N, et al. : Analysis of drivers of disenrollment and plan switching among Medicare Advantage beneficiaries. JAMA Intern Med 179:524-532, 2019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Li Q, Trivedi AN, Galarraga O, et al. : Medicare Advantage ratings and voluntary disenrollment among patients with end-stage renal disease. Health Aff 37:70-77, 2018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Lemjabbar-Alaoui H, Hassan O, Yang YW, et al. : Lung cancer: Biology and treatment options. Biochim Biophys Acta 1856:189-210, 2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Puckett Y, Garfield K: Pancreatic Cancer. Treasure Island, FL, StatPearls Publishing, 2024 [PubMed] [Google Scholar]
- 43.Joyce KM: Surgical management of melanoma, in Ward WH, Farma JM (eds): Cutaneous Melanoma: Etiology and Therapy. Brisbane, AU, Codon Publications, 2017 [PubMed] [Google Scholar]
- 44.Litwin MS, Tan HJ: The diagnosis and treatment of prostate cancer: A review. JAMA 317:2532-2542, 2017 [DOI] [PubMed] [Google Scholar]
- 45.Cassell A, Yunusa B, Jalloh M, et al. : A review of localized prostate cancer: An African perspective. World J Oncol 10:162-168, 2019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Tribble SJ. Older Americans say they feel trapped in medicare advantage plans. NPR. 2024. https://www.npr.org/sections/health-shots/2024/01/03/1222561870/older-americans-say-they-feel-trapped-in-medicare-advantage-plans [Google Scholar]