Predictors and Drivers of End-of-Life Medicare Spending Among Older Adults with Solid Tumors: A Population-Based Study

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

See review (attached)

Comments for author File: Comments.pdf

Author Response

We thank Reviewer 1 for their thoughtful comments. Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

The authors address an important topic in cancer patients’ care such as expenses incurred in the last month of life

I read with great interest the manuscript which, in general, already at the first approach, seemed to me well structured and complete in the description of the study’s purpose, the model used to answer the question, the methodology.

The statistical analysis is made very user-friendly for the reader and the conclusions are discussed in depth.

I can only suggest that the tables describing the characteristics of the sample and presenting the relationships between the covariates and the result should be made more readable.

 

Author Response

Comment 1: I can only suggest that the tables describing the characteristics of the sample and presenting the relationships between the covariates and the result should be made more readable.

Response 1: We thank Reviewer 2 for their thoughtful comments. We made some adjustments to the tables, and we are working with the editors to make them more easily digestible before publication.

Reviewer 3 Report

Comments and Suggestions for Authors

Brief Summary: This is a “population based retrospective cohort study” using USA Medicare data to “evaluate and compare the predictors of high-intensity end of life (Eol) spending” during the last 30 days of life for older adults with four types of cancer.  Predictors were selected “based on clinical significance, the existing literature, their relevance for healthcare delivery, and a conceptual framework for the determinants of treatment intensity among seriously ill patients.”  Findings supported previous research and provided some new insights into the problem of high-intensity Eol care for older adults with cancer.

General Comments: This manuscript is well written, logical and generally flows extremely well.

Specific Comments:

Abstract: Complete and adequate.

Introduction: This section lays out the argument for both the problem of expensive Eol care, and the design of this study including the variables selected and how “measured”, the statistical approach and how they addressed covariance, and assumptions of the linear regression. 

Methods Sample selection and outcomes: Inclusion and exclusion criteria are clear as well as a brief description of the database used (the SEER-Medicare database plus health claims data information) Covariates: Justification for particular set of variables is “clinical significance, existing literature, relevance for healthcare delivery, and a conceptual framework (patient-level clinical factors, patient-level demographic factors, area-level socioeconomic factors, physician/practice-related factors, and regional/geographic factors.  This framework is particularly helpful to organize the variable selection based on what was in the two data sets.  Lines 145-149 provide sufficient explanation as to how they determined “disability.” Sensitivity Analysis: Appropriate and clear.

Results:            

Descriptives: Figure 1 and Table 1 are complete and well summarized in lines 184-197.  It would be helpful with such a large table to put the headers on each page if the editor can do this. Figure 2 is also excellent and easy to understand.

Predictors: Table 2 coulld also use headings on each page to enhance understanding without having to go back and forth. Summary in lines 215-231 is also excellent.

Regression: Again, excellent summary in lines 238-256 and again, headings for Table 3

Discussion: Organized by the conceptual framework, compares to prior research and points out the differences found in this study.  The authors also discuss why they think they got the results they did (e.g., lines 295-301).

Recommendations: For professional organizations and decision support tools.  Would the authors also make recommendations for professional education (both initial and continuing education) and research to further explore why providers avoid difficult discussions and even older adults seek a diagnosis late and high intensity care more often?  To assume it is due to fear of discrimination seems premature.  There are also system biases that will not be changed with EMR reminders.  It seems there are cultural factors at work not fully addressed in the conceptual model used. 

Limitations: adequately addressed.

All references are timely appropriate.

 

Author Response

Comment 1: Figure 1 and Table 1 are complete and well summarized in lines 184-197.  It would be helpful with such a large table to put the headers on each page if the editor can do this. Figure 2 is also excellent and easy to understand.

Response 1: We thank Reviewer 3 for their thoughtful comments. We made some adjustments to the tables, and we are working with the editors to make them more easily digestible before publication.

Comment 2: Table 2 could also use headings on each page to enhance understanding without having to go back and forth.

Response 2: We thank Reviewer 3 for their thoughtful comments. We made some adjustments to the tables, and we are working with the editors to make them more easily digestible before publication.

Comment 3: Again, excellent summary in lines 238-256 and again, headings for Table 3

Response 3: We thank Reviewer 3 for their thoughtful comments. We made some adjustments to the tables, and we are working with the editors to make them more easily digestible before publication.

Comment 4: Would the authors also make recommendations for professional education (both initial and continuing education) and research to further explore why providers avoid difficult discussions and even older adults seek a diagnosis late and high intensity care more often?  To assume it is due to fear of discrimination seems premature. There are also system biases that will not be changed with EMR reminders. It seems there are cultural factors at work not fully addressed in the conceptual model used. 

Response 4: We completely agree with the reviewer’s suggestion regarding the need for more early and continuing training for providers. We added the following sentence in the “Recommendations for Policy & Practice” section: “Additionally, it may be necessary to invest more in early and continuing education programs that train providers on how to identify which patients are at greatest risk of high-intensity EoL care, effectively communicate about EoL decisions, and successfully refer patients to palliative care and hospice. Further research may also be needed to fully understand all of the factors that may contribute to the avoidance of EoL discussions by both patients and providers.”