Phase-Based and Lifetime Health System Costs of Care for Patients Diagnosed with Leukemia and Lymphoma: A Population-Based Descriptive Study

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Using linked datasets from 2005 to 2019, this study estimated phase-specific and lifetime costs for patients with leukemia and lymphoma. The findings of this study provide valuable insights into the economic burden of these diseases. However, there are several considerations and limitations to be aware of. I have listed my specific comments below:

·      This study didn’t include a control group (e.g. non-cancer patients). Without this, it's hard to tell which costs were due to cancer and which were from other health issues.

·      Similarly, without a control, the study didn’t address how it handles confounding factors like comorbidities, which can greatly affect healthcare costs, especially in older patients.

·      The reference to Canada's Consumer Price Index (CPI) is inaccessible (404 error) and titled "The Daily — Consumer Price Index: Annual review, 2020," which likely doesn't include the 2023 CPI. It is recommended to add a table in the appendix with the CPI details.

·      The study mentions using a hybrid costing methodology but provides little detail on its implementation or potential limitations.

·      One major limitation of using claims data is missing information. The flowchart only shows the number of patients excluded due to birthdate and death. However, in the methods, the authors mention that patients with missing or invalid data of some other variables were excluded as well (e.g. sex, date of diagnosis…).

·      While the study presents the results of some subgroup analyses (e.g., by age and gender), it's not clear if these were pre-specified or how multiple comparisons were handled statistically in the methods section.

·      SD and confidence intervals are not prominently reported for some results.

·      Table 1: why doesn’t the median age at diagnosis in years (Q1, Q3) have a number?

·      Table 1: Male (57.9%) and female (42.2%) in the Leukemia group doesn’t sum up 1. I know this could be due to rounding up, but it may be explained in the footnotes.

·      Table 2: Would it be possible to add SD or 95% CI for the Months in phase and their costs in this table as well? So does Table A2.

·      Figure 2: This figure needs to improve its readability, as the numbers and bars are overlapping. Additionally, an error bar for SD is needed.

·      Figure 3: It appears that the figure legend is missing for this figure.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

 This paper describes a major work done that aimed to evaluate the cost of illness of patients with leukemia and lymphoma from the public payer perspective in Ontario, Canada.

My major concern about this paper is that the costs recorded for this work are “all-cause healthcare costs” and not disease (cancer) specific costs. Recording all healthcare costs while reporting on “health system costs for leukemia and lymphoma care” seems to assume that patients with leukemia and lymphoma have no other diseases with treatment costs (diabetes, CHF, HTN etc.)  and that all health-costs derived from the electronic databases for these patients are related to their leukemia and lymphoma.

This of course is not the case. Moreover, since the median age of diagnosis of the patients with leukemia and lymphoma was 64 years and 66 years in the study population, for sure many of them came with other chronic illnesses. Not even considering emergency healthcare expenses such as accidents and others.

This major limitation is not mentioned at all by the authors in the “limitations” paragraph, and the authors did not report and did not do any analysis to control for background illnesses when comparing results between the different leukemia and lymphoma disease subtypes.

In any case, because of this limitation, the title of the paper should be revised to: “Phase-based and lifetime health system costs of patients diagnosed with leukemia and lymphoma: a population-based descriptive study”.

 

Other major comments:

11. The authors did an effort to link costs data from many different health administrative databases. For completeness of the description of the study design, the authors should add a table naming and describing each database that was used, the data that was retrieved from each database for this study and (the most important) a clarification regarding completeness of data. A remark in the limitations paragraph that “…we only had data on outpatient drugs covered by the publicly funded ODB program, which mainly serves those aged 65 and older, limiting our ability to capture all drug-related costs borne by private insurance or patients” cannot be acceptable only showing up somewhere at the end of the discussion. If I realize right, the data does not include outpatient drugs in all patients under the age of 65- which is a major drawback on the dataset!!! This limitation should have been mentioned specifically in the methods paragraph. When I have read the manuscript, I should have understood from the beginning that the data does not include these costs.

 

22. Table 2: I would have expected that if I divided the “cost by phase” for every cancer subtype by the “number of months in phase”, I should have got the values that appear in figure 2 (mean monthly costs associated with different cancer subtypes). However, this is not the case. Please explain.

 

33. Missing a conclusion paragraph after the discussion.

 

Minor comments:

1.1.  In the introduction, the sentence:” The costs are particularly higher for lymphoma and leukemia than for other types of cancer” is not referenced. How do we know that it is true?

 2.      In “study participants”: What is “OHIP eligible”- please explain.

 3.      In the “flow diagram” the first box, I assume, refers to patients with a de novo diagnosis during the study period, therefore please change from “Patients with a diagnosis of leukemia or lymphoma” to “patients diagnosed with leukemia or lymphoma…”. This way excluding patients that were diagnosed before the study period started.

 4.      In figures 2,3 table 2 and appendix tables and figures, missing data regarding the 3364 patients with “other leukemia”. This is a significant portion of patients missing results.

 5.      In the discussion, line 390: the costs measured in this study are total healthcare costs in cancer patients, you cannot talk specifically about “cost structure of cancer care”- since you don’t have the specific costs of the cancer care itself.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Thanks to the authors for addressing my previous comments. I have no further comment.

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you for sending me this updated version. I recommend now to accept as is.