Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The authors examined the relationship between the timing of enrollment in Medicaid health insurance and cancer-specific tumor stage and survival using the SEER database for breast, colorectal, and lung cancers. It's a good and valuable documentation. Suggestions:

1. Why was the restriction on tumor patients over 65 years of age implemented? This should be explained.
2. What is the reason for the low screening rates among Medicaid users? This should be explained.
3. Regarding the rates and detection of late-stage lung cancer, it is stated that a widespread screening program has not been implemented. Can this statement be clarified? Was this only in a specific region, or was this situation specific only to Medicaid insurance? Or was it not applied to anyone at all?
4. When the Discussion section mentions 6.4 million non-elderly individuals, what is the age limit? This should be clarified.
5. The conclusion states that Medicaid-insured individuals are more likely to seek medical attention in the later stages of the disease, and a solution suggested is to increase enrollment rates among uninsured individuals. This suggestion contradicts the fact that already registered patients are detected in the late stages. This situation needs to be clarified.
6. Since the article is entirely related to Medicaid insurance, it would be better to provide more detailed information about the content, scope, and operation of Medicaid insurance so that authors in countries where this insurance is not used can better understand the article.

Author Response

Please see attached 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Regarding clarity I would recommendthe addition of a short “Conceptual framework” sentence in the Introduction: why timing plausibly affects stage/survival (screening, care access, delays), to strengthen narrative cohesion.

For refferences, ensure that policy and screening-related claims (e.g., uninsured eligible but unenrolled; NBCCEDP underutilization) are backed by the most authoritative sources (CDC/KFF/peer-reviewed). They appear to be, but the authors should avoid over-relying on web reports unless paired with peer-reviewed evidence.

For study design, there are some changes i would recommend:

• Exposure definition nuance: “Pre-diagnosis” is defined as ≥12 months before diagnosis, while “post-diagnosis” is month of/within 12 months after. This is reasonable but creates heterogeneity. The Discussion does note that post-diagnosis includes newly eligible due to diagnosis vs eligible-but-unenrolled. Consider strengthening this limitation and, if possible, adding sensitivity analyses (e.g., alternative windows: 6 months/24 months).
• Comparison group heterogeneity: “Non-Medicaid” combines privately insured and uninsured, acknowledged as potentially biasing toward the null. This is a key limitation; authors could propose future work stratifying by insurance type if data allow 

Also, please clarify the random intercept/frailty specification (state-level) more explicitly: e.g., which level is random (state), how many states, and whether SEER registry clustering overlaps with state clustering.

Table 1 lists “Census Tract Median Income” as “$5.75” etc., later explained as “tens of thousands.” This is confusing. Prefer presenting as $57,500 directly (or label clearly in the table header). They state using Gray’s test for cancer-specific survival; but later use cause-specific Cox models. This is fine, but add one sentence clarifying the relationship: Gray’s test pertains to cumulative incidence in competing risks; cause-specific Cox estimates hazard among those at risk.

Author Response

Please see attached

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors used the SEER database for breast, colorectal, and lung cancers to examine the relationship between timing of enrollment in Medicaid health insurance and cancer-specific tumor staging and survival. I do not see an adequate and understandable response to my criticisms numbered 2, 3, 4, and 6. I believe that a study examining survival and tumor staging with certain types of cancer, involving an insurance company belonging to only one country, will have limited scientific contribution to readers worldwide.

 

Author Response

Comment 1: The authors used the SEER database for breast, colorectal, and lung cancers to examine the relationship between timing of enrollment in Medicaid health insurance and cancer-specific tumor staging and survival. I do not see an adequate and understandable response to my criticisms numbered 2, 3, 4, and 6. I believe that a study examining survival and tumor staging with certain types of cancer, involving an insurance company belonging to only one country, will have limited scientific contribution to readers worldwide.

Response 1: We acknowledge that our analysis draws on U.S. data (SEER linked with Medicaid) and is therefore most directly relevant to the U.S. context. Our study was designed to examine relationships between timing of Medicaid enrollment and cancer stage at diagnosis and survival within the U.S. health system. While findings may inform hypotheses in other settings with similar eligibility structures, we agree that global generalizability is limited and have clarified this in the manuscript’s Strengths and Limitations and Generalizability sections. We added a sentence noting that the results are principally applicable to U.S. policy and delivery contexts and may not extrapolate to countries with different insurance structures. This can be found in lines 410-412. As to the other comments, specifically comments 2,3, and 4. For comment 2, We appreciate the concern; however, screening rates were not an outcome analyzed in this study, and we do not report screening uptake or prevalence by insurance type. Our outcomes are tumor stage at diagnosis and survival. We have reviewed the text to ensure we do not imply that screening rates were measured, so we are unsure what this comment is referring to. For comment 3, thank you for pointing out the need for clarity. Our intention was to reference national screening guidelines (e.g., USPSTF) rather than to describe a specific “program” implemented by Medicaid or by a specific region. We have revised the language to avoid the term “program” and instead refer to “national screening guidelines. For comment 4, we use the standard U.S. convention of “non‑elderly” to refer to individuals under age 65 and have noted that in the paper. Again, for comment 6, we understand that there is concern that this study does not generalize to other parts of the world; however, this is not the intention of this paper. The goal of this paper is to describe this specific program as it related to cancer outcomes in the United States. We believe that we have provided sufficient detail within the scope of the paper about the Medicaid program for readers who are unfamiliar with it to understand the basics of the program and with citations to read further if they would like.

Reviewer 2 Report

Comments and Suggestions for Authors

I accept the publication in the current form 

Author Response

Comment 1: I accept the publication in the current form 

Response 1: We appreciate this reviewer's time and effort. They raised valid and useful critiques that we believe truly improved our manuscript.