Systematic Review of Health Insurance and Survival among Adolescent and Young Adult Cancer Patients

Abstract

Cancer is the leading cause of disease-related death among female adolescents and young adults (AYAs) and the second leading cause after heart disease among males in the United States. Despite recent improvements in survival outcomes, AYAs continue to experience poorer cancer survival rates compared with pediatric and older adult populations. Health care access and quality are substantially influenced by insurance coverage, and AYAs represent the age group with the highest proportion of uninsured individuals. The objective of this review was to examine the association between insurance status and cancer survival among AYAs to inform future policy interventions. Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, peer-reviewed studies published in the United States between 2018 and 2024 were identified from PubMed, CINAHL, and Web of Science. Eligible studies examined associations between insurance status and survival outcomes among the AYA cancer population. Ten studies, encompassing 468,583 AYAs with cancer (60.1% female), were included. Individuals with Medicaid/public insurance, as well as those uninsured, exhibited poorer survival outcomes compared with those privately insured across multiple cancer types, including head and neck cancer, lymphoma, central nervous system tumors, hepatic carcinomas, sarcoma, germ cell, breast, thyroid, cervical, colorectal, uterine, ovarian, kidney, lung, and melanoma. Several associations appeared to be age-dependent (e.g., lymphoma). Findings from this review suggest that insurance status is associated with both short- and long-term cancer outcomes among AYAs. The results also highlight persistent challenges in accurately assessing and categorizing insurance status within this population.

Introduction

In the United States, cancer is the leading cause of disease-related death among female adolescents and young adults (AYAs; ages 15–39 years) and the second leading cause after heart disease among male AYAs.¹ AYAs account for approximately 4.4% of all cancer diagnoses nationwide, translating to about 85,890 new cases annually.¹ The prevalence of cancer in this population is increasing, with incidence rates rising by 0.3% per year between 2010 and 2019.^2,3 Despite improvements in survival, reflected by an 86% relative survival rate across all cancers and a 0.9% annual decline in mortality from 2011 to 2020, AYAs continue to experience poorer cancer outcomes compared with pediatric and older adult populations.^2,3 Furthermore, AYA patients who are uninsured or underinsured, belong to racial or ethnic minority groups, or face socioeconomic disadvantages have been shown to have poorer survival outcomes.^4–7

Insurance coverage varies substantially across the AYA population. AYAs occupy a distinct developmental stage characterized by transitions in education, employment, and personal relationships, which increases their vulnerability to instability in insurance coverage.⁸ Despite recent policy changes in the United States, AYAs remain the largest uninsured age group.⁹ Cancer patients who are uninsured or covered by Medicaid are more likely to be diagnosed at a later stage, have lower rates of clinical trial participation, experience limited access to specialized cancer centers, and receive different types of treatment compared with those who have private insurance.^2,10–13 These disparities can significantly influence cancer survival among AYAs. For example, Keegan et al.² reported that AYA cancer patients who were enrolled in Medicaid at diagnosis were 2.5 times more likely to be diagnosed with later-stage disease compared with those with private insurance.

Given that cancer is the leading cause of disease-related death among AYAs and its incidence continues to rise, it is crucial to examine the association between insurance status and survival rates within this population. The purpose of this review was to evaluate the correlation between insurance status (Medicaid/public or uninsured, and privately insured) and cancer survival among AYAs, as reported in the existing literature. Findings from this review will shed light on this important topic, highlighting the role of insurance status in AYA survival outcomes. We hope that bringing more attention to this important topic will serve as an avenue for meaningful policy changes that will result in better survival rates for AYA cancer patients.

Methods

Search strategy

A systematic literature review of peer-reviewed studies was conducted in January 2025 to examine the association between insurance status and cancer survival among AYAs. Articles published between 2018 and 2024 were identified through PubMed, CINAHL, and Web of Science. A research librarian was consulted to refine the search strategy and develop appropriate keywords (Table 1). The Boolean operator “AND” was used to combine search terms such as adolescents and young adults, AYA, cancer, insurance, and survival. The review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Two reviewers (T.M. and Z.L.) independently screened the titles, abstracts, and full texts of identified studies. Discrepancies were resolved through discussion to minimize selection bias. Inter-rater reliability was assessed using Cohen’s kappa statistic.

Table 1.

Search Strategy: Keywords for Systematic Literature Review

Concept	Keywords
1. Populations: adolescents and young adults	“adolescent* and young adult*” OR “AYA*” OR “adolescent*” OR “young adult*” OR “adult*”
2. Disease: cancer	“cancer” OR “tumor*” OR “neoplasm” OR “carcinoma” OR “malignancy” OR “sarcoma” OR “hematological malignancy” OR “leukemia” OR “lymphoma” OR “melanoma” OR “brain tumor” OR “glioma”
3. Outcome: survival	“survival” OR “mortality” OR “outcome*”
4. Comparison: insurance	“insurance” OR “Medicaid” OR “Medicare” OR “public insurance” OR “uninsured” OR “Tricare” OR “military health insurance” OR “private insurance” OR “privately insured”

^*truncation symbol.

Searches combined with AND: 1 AND 2 AND 3 AND 4.

Eligibility criteria

Studies were included if they (1) focused on AYAs with cancer in the United States, (2) reported AYA-specific outcomes, (3) examined the association between insurance status and cancer survival, (4) were published in English, (5) employed a cohort or population-based design, (6) included any cancer type, and (7) were published between January 1, 2018, and December 31, 2024. Studies were excluded if they (1) were not published in English, (2) focused on populations outside the United States, (3) were reviews, dissertations, or conference abstracts, (4) did not include cancer survival as an outcome, (5) did not specifically address the AYA population, or (6) did not examine the impact of insurance status.

Data extraction

For each study, data were extracted on publication year, authors, data source, age group, year of cancer diagnosis, sample size, sex distribution, cancer type, measures of insurance status, outcome measures, covariates, statistical methods, and study results.

Data analysis

Descriptive analyses were used to summarize the review findings. Studies were categorized according to the insurance status examined: uninsured (patients without any insurance coverage), insured (patients with either public or private insurance), public insurance (coverage through Medicaid, Medicare, or Tricare), and private insurance (coverage provided by private health insurance companies). The outcome measures included survival or mortality comparisons between uninsured or Medicaid/public patients and those with private insurance, expressed as adjusted odds ratios (aOR), adjusted hazard ratios (aHR), and 95% confidence intervals (CI). In one study that did not use private insurance as the reference group, effect estimates were converted accordingly.¹⁰ Associations between insurance status and survival were classified as positive, negative, or null based on the reported effect estimates and a significance threshold of p < 0.05, where applicable, as defined by the original studies. Meta-analysis was not performed for two reasons: (1) heterogeneity in insurance classification, cancer types, and outcome measures, and (2) overlap of study populations due to multiple studies using the same data sources (e.g., Surveillance, Epidemiology and End Results [SEER], California Cancer Registry [CCR]).

Assessment of quality

Study quality was evaluated by one author (Z.L.) using a modified Newcastle–Ottawa Scale (NOS; score range 1–9, with 1–3 indicating low quality, 4–6 moderate quality, and 7–9 high quality).¹⁴ Confidence in the quantitative findings was evaluated according to sample size adequacy, precision of estimates, robustness of statistical analyses, and the level of data sources (national, state, or regional). Because no meta-analysis was conducted, formal statistical tests for publication bias (e.g., funnel plot, Egger’s test) were not applicable.

Results

Search results

Search results are summarized in Figure 1. A total of 1022 articles were identified across the databases: 816 from PubMed, 94 from CINAHL, and 112 from Web of Science. After removing 182 duplicates using Covidence, a web-based software that streamlines the systematic review process,¹⁵ 852 articles remained for screening. Based on title and abstract review, 834 studies were excluded. Of the remaining 18 articles assessed in full text, 4 were excluded because they did not compare cancer survival by insurance type^16–19 and 4 were excluded because their study populations included patients younger than 15 or older than 39 years.^20–23 A total of 10 studies met the inclusion criteria and were included in the final review.^{10,11,24–31} Inter-rater agreement for full-text screening was high, with a kappa statistic of 0.88. Of the 10 included studies, 8 were of high quality (NOS scores 7–9) and 2 were of moderate quality (scores 5–6) (Table 2). Overall, the included studies demonstrated high quality, with an average NOS score of 7.3.

FIG. 1.

PRISMA flowchart. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Table 2.

Summary of Studies’ Characteristics and Findings

Author (year of publication)	Data source	Age group	Year of diagnosis	AYA sample N = 468583	Female N (%) = 281463 (60.1)	Male N (%) = 187210 (39.9)	Type of insurance, N (%)
							Private	Medicaid/public	Uninsured	Unknown
Berkman et al. (2023)10	SEER-17	15–39	2000–2016	268,501	166,291 (62)	102,210 (38)	149,038 (55.5)	9761 (3.6)		109,702 (40.9)
Challapalli et al. (2018)27	SEER-18	15–39	2007–2014	1777	639 (36)	1138 (64)	1186 (66.7)	351 (19.8)	149 (8.4)	91 (5.1)
Colton et al. (2018)	SEER-18	15–39	2007–2014	66,556	39,007 (58.6)	27,549 (41.4)	48,328 (72.6)	18.228 (27.4)		Excluded
Cuglievan et al. (2021)24	MD Anderson Cancer Registry	15–29	2001–2003	201	101 (50.2)	100 (49.8)	151** (75.1)	50 (24.9)	Excluded	Excluded
Dhir et al. (2023)25	SEER-18	15–39	2007–2014	13,146	5931 (45.1)	7215 (54.9)	9201**** (70.0)	2466 (18.76)	1077 (8.19)	402 (3.06)
Huang et al. (2022)28	SEER-18	15–39	2007–2016	21,149	9265 (43.8)	11,884 (56.2)	13,427** (63.5)	6140* (29.03)	1582 (7.48)	Excluded
Murphy et al. (2021)26	TCR	15–39	2007–2012***	22,184***	14,074 (63.4)	8110 (36.6)	14,274 (64.34)	4056* (18.28)	3854 (17.37)	Excluded
Parsons et al. (2020)29	CCR	15–39	2005–2014	62,218	38,155 (61.3)	24,063 (38.7)	40,752** (65.5)	18,031* (29.0)	1838 (3.0)	1597 (2.6)
Singh et al. (2023)30	CCR	15–39 and ≥ 0	2000–2018	346#	167 (48.3)	179 (51.7)	242 (69.9)	91 (26.3)		13 (3.8)
Wojcik et al. (2023)31	CCR	15–39 and 40–64	2004–2015	12,505#	7833 (62.6)	4672 (37.4)	8389 (67.1)	853* (6.8)	316 (2.5)	2947 (23.6)

Author (year of publication)	Qualitya	Type of cancer(s)		Outcome measures	Key findings
Berkman et al. (2023)10	7	All AYA cancers grouped into three categories: Hematological malignancies CNS tumors Solic tumors		Early death: mortality within 2 months of diagnosis	Uninsured AYA cancer patients had increased risk of early death in each of cancer types.
Challapalli et al. (2018)27	9	Head and Neck Cancer		8-year cancer-specific mortality	AYAs with Medicaid or uninsured had an increased hazard of death from head and neck cancer, compared with those with private insurance. Outcomes for those with Medicaid are not better than the uninsured. Health insurance status is the main nonclinical factor associated with survival among AYAs with head and neck cancer.
Colton et al. (2018)	8	Lymphoid leukemia AML HL NHL Astrocytoma Other gliomas Hepatic carcinomas Fibrosarcoma, Peripheral nerve, and other fibrous Malignant gonadal germ cell tumors Other unspecified		Mortality: not specified.	An increased risk of death among those with public or no insurance compared to private insurance for most cancer types and age groups. Specifically, for lymphoid leukemia (age 15–19, 30–34, and 35–39), acute myeloid leukemia (age 15–19 and 25–29), Hodgkin lymphoma (all ages), non-Hodgkin lymphoma (age 20–24, 25–29, 30–34, and 35–39), astrocytomas (age 30–34), other gliomas (age 25–29, 30–34, and 35–39), hepatic carcinomas (age 25–29), fibrosarcomas, peripheral nerve and other fibrous tumors (age 30–34), malignant gonadal germ cell tumors (age 20–24, 25–29, 30–34, and 35–39), other and unspecified carcinomas (age 20–24, 25–29, 30–34, and 35–39)
Cuglievan et al. (2021)24	5	Bone sarcoma  Leukemia  Brain tumor  NHL  HL  Soft-tissue sarcoma  Thyroid cancer		Overall survival: from time of diagnosis  Late overall survival: from the date of 5-year anniversary of cancer diagnosis	Patients with private insurance had significantly better late overall survival than patients with public insurance. Patients with brain tumors or leukemia, those who had private insurance had better overall survival than those who had public insurance. Patients with thyroid cancers, those who had private insurance had better late overall survival.
Dhir et al. (2023)25	9	Lymphoma: HL NHL		5-year overall survival	Both HL and NHL AYA patients with Medicaid or uninsured were at significantly increased risk of death compared to their insured counterparts
Huang et al. (2022)28	8	Lymphoma: HL NHL		5-year overall survival	Risk of death was significantly higher for AYA lymphoma patients with Medicaid or no insurance than for those with private insurance. The ACA has improved insurance coverage and survival rates. AYA patients with either no insurance or Medicaid both had worse death rates. AYA patients from most age groups being uninsured or having Medicaid compared with privately insured had higher risk of death
Murphy et al. (2021)26	7	Women Breast Thyroid Cervix Melanoma Colorectum Osteosarcoma HL NHL Uterus Ovary Kidney Pelvis Astrocytoma AML Lung ALL	Men Testis HL NHL Melanoma Colorectum Thyroid Kidney Pelvis Astrocytoma ALL Kaposi Sarcoma AML Lung Soft Tissue Sarcoma Other Glioma Bladder	5-year relative survival: the ratio of observed survival to expected survival in cancer-free population. 5-year overall survival	Women with Medicaid had lower 5-year relative survival compared with women with private insurance for breast, cervical, colorectal, uterine, ovarian, kidney or renal pelvis, lung cancer, melanoma, Hodgkin lymphoma, Non-Hodgkin lymphoma, and ALL. Women with no insurance had lower 5-year relative survival compared with women with private insurance for breast, cervical, colorectal, uterine cancer, melanoma, and Hodgkin lymphoma. Men with Medicaid had lower 5-year relative survival compared with men with private insurance for testicular, colorectal, and kidney or renal pelvis cancer, and non-Hodgkin lymphoma. Having Medicaid, other government insurance, or no insurance at diagnosis statistically significantly increased all-cause mortality in both women and men compared with having private insurance.
Parsons et al. (2020)29	7	Breast Thyroid Melanoma Testicular NHL HL Cervical	Colorectal Ovarian Sarcoma CNS ALL AML	5-year relative survival: the ratio of observed survival to expected survival in cancer-free population. 5-year overall survival	For all cancers, compared with those with private/military insurance, individuals with Medicaid insurance had significantly worse both overall and cancer-specific survival regardless of when coverage began. For all cancers, those who obtained Medicaid at diagnosis or had discontinuous Medicaid enrollment had worse overall and cancer-specific survival relative to those with were continuously enrolled. When stratified by cancer site, similar patterns for many cancers were identified (see details in Supplementary Table S1).
Singh et al. (2023)30	6	Synovial sarcoma		Overall survival Cancer-specific survival	Health insurance status (public/uninsured) was not associated with overall and cancer specific survival among AYA patients with synovial sarcoma.
Wojcik et al. (2023)31	7	Melanoma		Overall survival Cancer-specific survival	Medicaid, Medicare, or military insurance was associated with a higher risk of death compared with privately insured AYA melanoma patients.

^*Medicaid and other public insurance were grouped together.

^**Military insurance was grouped into private insurance category.

^***Only data used for the insurance analysis were included. The original study used data from 1995–2016 with 88,000 patients, but only data from 2007–2012, involving 22,184 patients, were used to examine the relationship between insurance status and survival outcomes.

^****All insureds were grouped into privately insured category except Medicaid.

^#Only AYA samples were included.

^aQuality assessed according to a modified Newcastle–Ottawa Scale (range 1–9, with 1–3 indicating low quality, 4–6 indicating moderate quality, and 7–9 indicating high quality).

AML, acute myeloid leukemia; CNS, central nervous system; CCR, California Cancer Registry; HL, Hodgkin lymphoma; NHL, non-Hodgkin lymphoma; SEER, Surveillance, Epidemiology and End Results; TCR, Texas Cancer Registry.

Study characteristics

Of the 10 included studies, 7 focused exclusively on the AYA population.^{10,11,25–29} One study focused specifically on individuals aged 15–29 years.²⁴ The remaining two studies included broader age ranges that encompassed older adults but were retained in the analysis because they reported separate results for the AYA subgroup.^30,31 For these two studies, only AYA-specific sample sizes and data were extracted.^30,31

Collectively, the 10 studies included a total of 468,583 AYA patients, comprising 281,463 females (60.1%) and 187,210 males (39.9%) (Table 2). Five studies utilized national cancer registry data from SEER-17¹⁰ or SEER-18^11,25,27,28; four used state-level cancer registries, including three based on the California Cancer Registry (CCR)^29–31 and one on the Texas Cancer Registry (TCR);²⁶ and one study relied on institutional data from the MD Anderson Cancer Registry.²⁴ The years of cancer diagnosis across studies ranged from 2000 to 2018, with five of them focusing on diagnoses made between 2007 and 2016^11,25–28 (Table 2 and Supplementary Fig. S1).

Insurance categorization

Insurance status was a key variable across all included studies, although the methods of categorization and reporting varied (Table 2). Private insurance was the most commonly reported type and was included in all studies, with proportions ranging from 55.5% in Berkman et al.¹⁰ to 75.1% in Cuglievan et al.²⁴ Three studies classified military insurance as private insurance.^24,28,29 Dhir et al.²⁵ categorized all insured individuals as having private insurance, except those covered by Medicaid.

Among public insurance types, Medicaid was the most frequently reported, ranging from 6.8% to 29.03% (Table 2). Huang et al.²⁸ and Murphy et al.²⁶ grouped Medicaid with other government programs (e.g., federal or state plans, Tricare). The proportion of uninsured patients ranged from 2.5% to 17.37%.^26,31 Three studies combined Medicaid/publicly insured and uninsured individuals into a single category.^10,11,30 Cuglievan et al.²⁴ excluded uninsured individuals from its analysis.

Cancer types and outcomes

A variety of cancer types and outcome measures were examined across the included studies (Table 2). Five studies^{25,27,28,30,31} focused on a single cancer type, while the remaining five^{10,11,24,26,29} examined a broad range of AYA cancers. Two studies specifically investigated hematological malignancies,^25,28 and three focused on solid tumors.^24,27,30 Survival outcomes varied, including short-term survival (e.g., within 2 months of diagnosis)¹⁰ and long-term survival outcomes (e.g., 8 years²⁷ and over 15 years²⁴ post-diagnosis). Three studies assessed the association between insurance status and 5-year survival, including overall survival and/or cancer-specific survival.^25,26,28 The remaining four studies did not specify a time frame for survival.^11,29–31

Medicaid/public versus privately insured

Seven studies examined overall survival among AYA cancer patients with Medicaid/public insurance versus those with private insurance, using Cox proportional hazards models to report aHRs and 95% CIs (Fig. 2 and Supplementary Table S1).^25–31 Of these, six studies found that individuals with Medicaid/public insurance experienced significantly worse survival outcomes across various cancer types compared with those holding private insurance.^25–29,31 Specifically, Huang et al.²⁸ reported an aHR of 1.08 (95% CI: 1.03–1.13) for lymphoma, while Wojcik et al.³¹ observed an aHR of 2.47 (95% CI: 1.96–3.12) for melanoma. Murphy et al.²⁶ reported aHRs of 2.17 (95% CI: 1.98–2.37) for females (across 15 cancer types: breast, thyroid, cervix, melanoma, colorectum, osteosarcoma, Hodgkin lymphoma [HL], non-Hodgkin lymphoma [NHL], uterus, ovary, kidney, pelvis, astrocytoma, acute myeloid leukemia [AML], lung, and acute lymphoblastic leukemia [ALL]) and 2.07 (95% CI: 1.83–2.94) for males (testis, HL, NHL, melanoma, colorectum, thyroid, kidney/pelvis, astrocytoma, ALL, Kaposi sarcoma, AML, lung, soft tissue sarcoma, other glioma, and bladder). Dhir et al.²⁵ observed aHRs of 2.23 (95% CI: 1.76–2.81) for HL and 1.95 (95% CI: 1.67–2.27) for NHL. Challapalli et al.²⁷ reported an aHR of 1.61 (95% CI: 1.22–2.12) for head and neck cancer.

FIG. 2.

Comparison of reported adjusted hazard ratios (aHR) and 95% confidence intervals (CIs) between Medicaid/public insured and privately insured AYA cancer patients across seven studies. AYA, adolescent and young adult; HL, Hodgkin lymphoma; NHL, non-Hodgkin lymphoma.

Parsons et al.²⁹ observed elevated mortality hazards across 13 common cancers (breast, thyroid, melanoma, testicular, NHL, HL, cervical, colorectal, ovarian, sarcoma, central nervous system [CNS], ALL, and AML) for individuals with Medicaid, regardless of timing of enrollment. Specifically, those with continuous Medicaid enrollment had an aHR of 1.42 (95% CI: 1.33–1.52), discontinuous enrollment had an aHR of 1.64 (95% CI: 1.49–1.80), and those enrolling at diagnosis had an aHR of 1.51 (95% CI: 1.42–1.61), all compared with privately insured individuals.²⁹ Singh et al.³⁰ found no statistically significant difference in survival between Medicaid and privately insured patients with synovial sarcoma. Wojcik et al.³¹ additionally reported that melanoma patients with military insurance had an aHR of 2.03 (95% CI: 1.30–3.18) for all-cause mortality compared with those privately insured, which was slightly lower than that of patients with Medicaid (aHR = 2.47, 95% CI: 1.96–3.12) or Medicare (aHR = 2.99, 95% CI: 2.01–4.45) (Supplementary Table S1).

Uninsured versus privately insured

Six studies examined survival outcomes among AYA cancer patients, comparing uninsured individuals with those holding private insurance (Fig. 3 and Supplementary Table S1).^{10,25–28,31} Five of these studies used Cox proportional hazards models to report aHRs and 95% CIs.^25–28,31 Murphy et al.²⁶ reported significantly worse survival among uninsured females and males across 15 cancer types, with aHRs of 1.82 (95% CI: 1.66–1.99) and 1.44 (95% CI: 1.30–1.59), respectively. Wojcik et al.³¹ (2023) found a significantly increased mortality risk among uninsured individuals with melanoma (aHR = 1.60, 95% CI: 1.07–2.41). Similarly, Challapalli et al.²⁷ reported higher mortality among uninsured patients with head and neck cancer (aHR = 1.51, 95% CI: 1.03–2.21). Dhir et al.²⁵ (2023) observed a significantly elevated mortality risk for uninsured patients with HL (aHR = 1.88, 95% CI: 1.18–1.94).

FIG. 3.

Comparison of reported adjusted hazard ratios (aHR) and 95% confidence intervals (CIs) between uninsured and privately insured AYA cancer patients across seven studies. CNS, central nervous system; HL, Hodgkin lymphoma; NHL, non-Hodgkin lymphoma.

In contrast, Huang et al.²⁸ found no statistically significant difference in survival between privately insured and uninsured HL or NHL patients, with an aHR of 1.06 (95% CI: 0.99–1.14). Berkman et al.¹⁰ focused on early mortality (within 2 months of diagnosis) and found that uninsured patients had significantly higher odds of early death across cancer types, including hematological malignancies (aOR = 1.82, 95% CI: 1.49–2.22), CNS tumors (aOR = 1.92, 95% CI: 1.54–2.17), and solid tumors (aOR = 1.85, 95% CI: 1.54–2.17).

Age-related insurance status and survival

Two studies examined the relationship between insurance status and survival among AYA lymphoma patients across different age groups (Fig. 4 and Supplementary Table S1). Huang et al.²⁸ reported that uninsured individuals aged 18–39 diagnosed with HL or NHL had significantly higher hazards of death compared to those with private or military insurance. The elevated risks were observed in the 18–25 age group (aHR = 1.50, 95% CI: 1.11–2.03) and the 26–39 age group (aHR = 1.74, 95% CI: 1.47–2.06), whereas the association was not statistically significant for the 15–17 age group (aHR = 2.63, 95% CI: 0.88–7.81).²⁸

FIG. 4.

Comparison of reported adjusted hazard ratios (aHR) and 95% confidence intervals (CIs) between non-privately (Medicaid/public or uninsured) and privately insured AYA lymphoma (HL and NHL) patients across two studies, stratified by age group. AYA, adolescents and young adult; HL, Hodgkin lymphoma; NHL, non-Hodgkin lymphoma.

Individuals with Medicaid also faced increased mortality risks relative to those with private insurance across all age brackets.²⁸ Specifically, the aHRs were 1.85 (95% CI: 1.14–3.02) for ages 15–17, 1.81 (95% CI: 1.47–2.24) for ages 18–25, and 2.36 (95% CI: 2.09–2.66) for ages 26–39.²⁸ Similarly, Colton et al.¹¹ found significantly higher mortality among publicly insured or uninsured individuals with HL compared with privately insured patients across all age groups, with the highest risk observed in the 25–29 age group (aHR = 3.27, 95% CI: 1.81–5.94). For NHL, elevated risks were also reported among publicly insured or uninsured patients, except in the 15–19 age group, where the association did not reach statistical significance.¹¹

Discussion

In this study, we conducted a systematic review of published literature to examine the association between insurance status and survival outcomes among AYAs with cancer. We identified 10 observational studies with strong evidence showing that AYAs with Medicaid or no insurance had a higher risk of mortality and poorer survival outcomes than those with private insurance. Although the strength of these associations varied by cancer type and age group, the overall pattern remained consistent across multiple cancers. These findings underscore the complex relationship between insurance status and cancer survival among AYAs.

Medicaid insured and uninsured

Medicaid was the primary public insurance reported in the included studies, accounting for up to 29% of cases.^28,29 Most studies in this review found that AYAs with Medicaid had a higher mortality risk than their privately insured counterparts.^25–29,31 These findings are consistent with broader evidence showing that, although Medicaid improves access to care relative to being uninsured, it does not eliminate disparities in cancer outcomes.^2,17 For example, Zhao et al.³² reported that adults aged 18–64 years with Medicaid were significantly more likely to be diagnosed with late-stage cancer across all stageable cancer types and had poorer short- and long-term survival at each stage. Medicaid recipients often face barriers such as limited provider participation, delayed treatment initiation, and reduced access to high-quality cancer centers, all of which contribute to survival gaps.^33–35 The findings suggest that Medicaid coverage alone is insufficient to ensure equitable cancer outcomes. Efforts to improve the quality and accessibility of care across insurance types, such as increasing Medicaid reimbursement rates and strengthening regulations on private insurance policies, may enhance provider participation and expand access to high-quality cancer care,^36,37 ultimately improving outcomes for Medicaid beneficiaries.

Uninsured AYAs had poorer survival outcomes than those with private insurance. Across cancer types, they faced a higher risk of both early and long-term mortality.^{10,25–27,31} These findings are consistent with Wu et al.,³⁸ who reported that uninsured cervical cancer patients had an increased risk of mortality compared with privately insured patients. They also align with Institute of Medicine (IOM) reports³⁹ showing that uninsured nonelderly adults experience poorer health and shorter life expectancy than those with insurance, as well as with evidence from the Annual Review of Public Health⁴⁰ indicating that health insurance improves health outcomes.

Overall, AYAs with Medicaid or no insurance had higher mortality risk (hazard ratios) than those with private insurance across multiple cancer types. The similar survival outcomes observed among AYAs with Medicaid and those uninsured (Figs. 2 and 3) likely reflect that, in many states, individuals who are newly diagnosed with cancer and initially uninsured often become eligible for Medicaid to cover part of their treatment. As Yabroff et al.⁴¹ reported, prior to the Affordable Care Act (ACA), 22%−60% of Medicaid recipients obtained coverage at or after their cancer diagnosis. Because some individuals enroll in Medicaid following diagnosis, and cancer registries typically record insurance status only once, this may partly explain the similar survival outcomes between Medicaid-insured and uninsured groups relative to those privately insured.

Additionally, the higher proportion of Medicaid-insured or uninsured AYAs may contribute to their significantly lower 5-year survival rates compared with children and older adults, as well as their slower survival gains over time.⁴² Previous studies have shown that having health insurance is associated with higher rates of cancer screening, earlier diagnosis, timely treatment initiation, and improved survivorship care,^{2,32,38,41,43–47} underscoring insurance as a protective factor across the cancer care continuum.⁴⁸ Addressing these structural inequities could meaningfully reduce survival disparities and improve long-term outcomes for this vulnerable population.

Cancer type-related insurance and survival

Although most cancer types consistently showed poorer survival outcomes among individuals with Medicaid or no insurance compared with those privately insured,^{10,11,24–30} certain cancers (e.g., synovial sarcoma) did not appear to be affected by insurance status.³⁰ This discrepancy may be attributed to cancer-specific factors (e.g., slower tumor growth), limited sample sizes in the included studies, or the absence of screening tests. Historically, lack of insurance has been associated with reduced access to cancer screening services, which are essential for early detection and improved survival outcomes.⁴⁹ Synovial sarcoma, a rare soft tissue cancer with no established population screening test,⁵⁰ may partly explain the lack of association observed between insurance status and survival. Such variations across cancer types underscore the importance of conducting disease-specific analyses when evaluating insurance-related disparities.

Age-related insurance and survival

This review also found that the association between insurance status and cancer survival among AYAs varied by age cohort, with the greatest mortality disparities observed in older AYA groups.^11,28 Among individuals with lymphomas (account for up to 30% AYA cancers,^51,52) under age 19, Medicaid and uninsured patients did not have significantly higher mortality risk compared with those with private insurance.^11,28 Those findings may reflect the protective effects of coverage through Children’s Health Insurance Program for those under 19. The higher mortality risk among older AYAs suggests that this population faces distinct barriers to cancer care, including aging out of dependent coverage, unstable or low-wage employment, and underutilization of preventive care due to cost.^53–55 A substantial body of evidence has shown that the dependent-coverage expansion was linked to reduced uninsured rates and improved survival following a cancer diagnosis.^56,57 These age-related disparities in cancer survival among AYAs highlight a critical intersection of developmental transitions, socioeconomic vulnerability, and insurance instability.⁸

Policy implications

Higher mortality rates among Medicaid-insured and uninsured AYAs underscore the urgent need for policies that enhance insurance continuity, expand access to affordable care, and strengthen provider participation in public insurance programs. Expanding Medicaid coverage—particularly in states that have not adopted Medicaid expansion—could substantially increase insurance access among AYAs. For example, Roth et al.⁵⁸ reported that provisions of the ACA, including Medicaid expansion and the Dependent Coverage Expansion, have improved both insurance coverage and cancer outcomes in this population. The ACA allows parents with private insurance to retain dependents on their plans until age 26,⁵⁹ thereby increasing coverage among AYAs. Further extending dependent coverage eligibility and increasing federal subsidies for ACA marketplace plans could further reduce uninsurance rates and improve health outcomes.

A recent population-based study found that individuals residing in states that expanded Medicaid eligibility under the ACA experienced greater improvements in cancer survival outcomes than those in non-expansion states.⁶⁰ In addition, the Centers for Disease Control and Prevention’s National Breast and Cervical Cancer Early Detection Program (NBCCEDP) has expanded access to guideline-concordant breast and cervical cancer screening, diagnosis, and treatment through Medicaid.⁶¹ However, Medicaid coverage provided under the NBCCEDP ends once cancer treatment is completed.⁶¹ Some policy leaders have advocated for a single-payer national health insurance system⁶² to provide universal coverage, a potential long-term solution for achieving equity in cancer care and health outcomes across all Americans.

Quality of evidence

The quality of included studies was generally moderate to high, with most employing population-based data and appropriate statistical adjustments. However, several studies used the same data source with overlaps on study populations.^{10,11,25,27,28} Two studies have relatively small sample size.^24,30 While most studies adequately defined populations and outcomes, residual confounding and incomplete adjustment for socioeconomic variables were common limitations. Consequently, the conclusions drawn from this review should be interpreted cautiously, and future high-quality longitudinal studies following insurance coverage are warranted to confirm these associations.

Strengths and limitations

To our knowledge, this is the first systematic review to examine the association between insurance status and cancer outcomes specifically among AYAs. We searched multiple scientific databases to comprehensively assess the relationship between insurance status and survival outcomes. All included studies utilized cancer registry data covering broad geographic regions and diverse populations, thereby enhancing the generalizability of the findings. Registry data also reflect real-world clinical practice, providing insights into the effects of health policies outside controlled trial settings.

Several limitations should be noted. First, due to the heterogeneity of insurance classifications, cancer sites, and confounders, we were unable to conduct quantitative meta-analyses. Second, we may have missed some relevant studies, and some of the measures we abstracted and reported may involve unavoidable subjectivity. Third, because all studies were observational and relied on secondary data, causality between insurance status and cancer survival outcomes cannot be inferred. Fourth, the absence of longitudinal data on changes in insurance status may obscure differences between coverage groups and underestimate the true impact of insurance. Fifth, although a formal assessment of publication bias was not feasible, the possibility of selective publication cannot be excluded. Studies with significant findings are more likely to be published, which may have influenced the overall conclusions. Finally, this review was not registered in PROSPERO because data extraction had been completed before registration was considered.

Conclusion

This systematic review found that AYAs with Medicaid or no insurance generally experienced poorer survival outcomes than those with private insurance across multiple cancer types. These findings underscore the critical role of health insurance in improving cancer survival among AYAs. Policymakers should prioritize strategies that expand access to affordable care and enhance the quality of care for publicly insured and uninsured populations. Future research should standardize insurance classifications, comprehensively adjust for potential confounders, and track insurance status over time. Linking claims and registry data and applying time-sensitive analytical methods could more accurately estimate the impact of insurance continuity on cancer outcomes among AYAs.

Supplementary Material

Supplementary Data