Comorbidities and Risk Factors of Patients Diagnosed with CRC after Positive Fecal Test in Real Life

Round 1

Reviewer 1 Report

Thank you for the opportunity to review this manuscript titled “Comorbidities and risk factors of patients diagnosed with CRC 2 after Positive Fecal Test in real life”. The authors conducted a retrospective cohort study to investigate the risk factors for CRC among patients with a positive FOBT in the real life. The authors found that compared to patients with No CRC after positive FOBT, patients with CRC had lower levels of hemoglobin, iron and ferritin. Additionally, the authors found that age, anemia, family history of colorectal cancer, and previous colorectal cancer are factors for having colorectal cancer among those with positive fecal testing. Overall, the manuscript is well written and clearly of relevance. However, the authors used an incorrect regression model i.e., used Cox regression model but reported OR for risk factors. I have few comments for the authors to address.

 

1.     Line 94: Did the authors apply any exclusion criteria?

2.     Line 97: The authors should discuss more about the Israel’s HMO and describe the data source.

3.     Line 106: When was the demographic and laboratory tests measured? Please provide details on the time period for measurement of key variables.

4.     Line 115: The authors should discuss more about the data source Clalit Health Services.

5.     Line 129: Which regression model was used? Please provide details on the model fit and other statistics.

6.     Line 172-175: Age, family history of CRC and previous CRC were associated with CRC diagnosis. When was CRC diagnosis measured? How long were patients followed?

7.     Line 182: Please provide a note on the variables that were controlled for in the multivariable analysis. Why are results demonstrated only for few variables? Please explain why the OR was NA for variables like diabetes mellitus.

8.     Line 183: The title mentions use of Cox regression model which is a time to event model. However, the results reported are Odds Ratio which is generated from Logistic regression. How did the authors determine association between positive FOBTs and diagnosis of colorectal cancer when everyone included in the study had a positive FOBTs. What was the comparator group? I think the authors are confusing different regression models. Please revise appropriately.

9.     Line 273: Please add a limitation on the possibility that not all patients may have complete data capture.

 

Author Response

attached file 

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors analyzed participants in a large population-based screening program for colorectal cancer in Israel studying the characteristics of participants with a positive fecal occult blood test (FOBT). Specifically, they compared participants’ characteristics between those with a positive FOBT result + a subsequent colorectal cancer diagnosis and those with a positive FOBT results but without a subsequent colorectal cancer diagnosis. They report that iron deficiency anemia, a family history of colorectal cancer, and previous diagnosis of colorectal cancer are associated with a colorectal cancer diagnosis after FOBT. The manuscript has weaknesses.

Comments:

The authors report that 847,550 FOBTs were performed and 45,500 returned a positive result. Of those participants with a positive FOBT, 1502 were diagnosed with colorectal cancer. Question: what does it mean having a positive FOBT results but not having a subsequent colorectal cancer diagnosis? Did these participants undergo the same procedures that the participants received with a colorectal cancer diagnosis? Did they have any ascertained follow up to characterize them as not having colorectal cancer? If not, the current analysis is not meaningful.

It is also a weakness of the manuscript that it contains so little context to the published literature on the performance of the FOBT besides Israel – this type of analysis has been done in other screened populations. What came out of it?

In the introduction we should learn:

What are the screening guidelines for colorectal cancer in Israel? How is FOBT used versus colonoscopy?

What are the screening guidelines after a previous colorectal cancer diagnosis?

The authors briefly introduce findings from references 8 & 9. What do we learn from the current study that is different from the other two studies?

The findings that a previous diagnosis of colorectal cancer and a family history of colorectal cancer associates with a colorectal cancer diagnosis after a positive FOBT is not surprising – an expected result. The reviewer would expect that a recommendation for colorectal cancer screening among high-risk subjects with a previous diagnosis of colorectal cancer and a family history of colorectal cancer already exists in Israel. I would expect that high-risk participants are recommended to have a colonoscopy or colonoscopy follow-up. What is learned?

The authors recommend that subjects with both a positive FOBT and iron deficiency anemia should have priority for colonoscopy. The association of anemia with a colorectal cancer diagnosis after a positive FOBT is not statistically significant and seeing the odd ratio this association appears weak in general. Seeing the data this reviewer is not convinced that such a recommendation should be done with the data in hand. Please discuss.

Table 1, CVA – 2.1 not 21.

Lines 216-18, sentence not fully clear. What does 2.7% vs. 0.4% mean? Why would anemia so be so different from iron deficiency anemia?

Author Response

please find the attached file 

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Comments have been addressed. 

Reviewer 2 Report

Thank you for the comments and changes to the manuscript. No further requests for revision.