Cost-Utility of the eHealth Application ‘Oncokompas’, Supporting Incurably Ill Cancer Patients to Self-Manage Their Cancer-Related Symptoms: Results of a Randomized Controlled Trial

Round 1

Reviewer 1 Report

Dear authors,

Thank you for the opportunity to review the manuscript, “Cost-utility of the eHealth application ‘Oncokompas’, supporting incurably ill cancer patients to self-manage their cancer-related symptoms: results of a randomized controlled trial". The manuscript is well written, follows the steps outlined in a preceding protocol article, and contributes to the evidence on cost-utility of eHealth applications for incurably ill patients. I have no major comments, but have added some specific comments below, and hope they will be useful to the authors.

 

Minor comments:

·        1) Results: Were any of the sociodemographic and clinical characteristics statistically different between groups at baseline? Even if they were not, I would suggest mentioning this shortly in the text.

 

·         2) Table 1: Could putting % in brackets make the table easier to read? Employment: I would suggest to put “Absent from work” with an indent under “Yes”, as this is a subsample of “yes”.

 

·         3) Table 2: Formatting: Should “Control (N=61)” at t2 be in bold? Should it be a line under the text?

 

·         4) Table 2: I would suggest putting SDs in brackets make the table easier to read.

 

·         5) Table 3: Why is T1 mentioned as a time-point when it’s not a part of the analysis? The text has different formatting within the table. Should the brackets (n=xx) behind “3 months follow” be removed or is something missing?

Author Response

Reviewer 1

Dear authors,

Thank you for the opportunity to review the manuscript, “Cost-utility of the eHealth application ‘Oncokompas’, supporting incurably ill cancer patients to self-manage their cancer-related symptoms: results of a randomized controlled trial". The manuscript is well written, follows the steps outlined in a preceding protocol article, and contributes to the evidence on cost-utility of eHealth applications for incurably ill patients. I have no major comments, but have added some specific comments below, and hope they will be useful to the authors.

 

Response: Thank you for reviewing our paper. We are pleased to read that you found the article well written.

 

Minor comments:

•       1) Results: Were any of the sociodemographic and clinical characteristics statistically different between groups at baseline? Even if they were not, I would suggest mentioning this shortly in the text.

 

Response: we adapted Table 1 by adding a column with p-values for each item. None of the sociodemographic and clinical characteristics were statistically different between both groups at baseline. We adapted the Results section. The text is now as follows: “No significant differences in sociodemographic and clinical characteristics were found between the intervention and control group at baseline (Table 1).”.

 

• 2) Table 1: Could putting % in brackets make the table easier to read? Employment: I would suggest to put “Absent from work” with an indent under “Yes”, as this is a subsample of “yes”.

 

Response: we adapted Table 1. The percentages were put into brackets and the “absent from work”-row is adapted with an indent.

 

• 3) Table 2: Formatting: Should “Control (N=61)” at t2 be in bold? Should it be a line under the text?

 

Response: we adapted Table 2 and put ‘Control’ and ‘Intervention’ in bold, to increase the readability of the table.

 

• 4) Table 2: I would suggest putting SDs in brackets make the table easier to read.

 

Response: we adapted Table 2 and put the SD in brackets to increase the readability of the table.

 

• 5) Table 3: Why is T1 mentioned as a time-point when it’s not a part of the analysis? The text has different formatting within the table. Should the brackets (n=xx) behind “3 months follow” be removed or is something missing?

 

Response: thank you for mentioning this. These texts (‘T1’ and ‘(n=xx)’) should have been removed before submission. We removed the text.

Reviewer 2 Report

Thank you for the opportunity to review this very interesting and well written paper paper. I have only a few comments.

Abstract

The authors should move the description of Onkocompass and the need it addresses to the first part of the abstract.

the abstract should provide an interpretation of the non-significant findings.

the background is well written and nicely frames the study.

line 215 references  a reason for non-participation as 'too confronting' it is not clear what this means. Please clarify.

 

Author Response

Reviewer 2

Thank you for the opportunity to review this very interesting and well written paper paper. I have only a few comments.

Response: Thank you for reviewing our paper. We are pleased to read that you found the article interesting and well written.

• Abstract

The authors should move the description of Onkocompass and the need it addresses to the first part of the abstract.

Response: we removed the description of Oncokompas to the first part of the abstract.

• the abstract should provide an interpretation of the non-significant findings

Response: we adapted the abstract and added the following text: “Non-parametric bootstrapping with 5000 replications were used to obtain 95% confidence intervals around the Incremental costs and quality-adjusted life years (QALYs). A probabilistic approach was used because of the skewness of cost data.”

And: “Future research on the costs of eHealth in palliative cancer care is warranted to assess the generalizability of the findings of this study.”

• the background is well written and nicely frames the study.

Response: Thank you. We are pleased to read this.

• line 215 references a reason for non-participation as 'too confronting' it is not clear what this means. Please clarify.

Response: we adapted the text which is now as follows: “being too (emotionally) confronting".

Reviewer 3 Report

Thank you for the opportunity to review this paper that contributes knowledge about the cost-effectiveness of e-Health interventions. The paper advanced knowledge about cost-utility evaluation of e-Health in palliative care.

The paper would benefit from a few minor revisions, as follows:

Methods - Specify the statistical tests employed to ascertain differences between the intervention and control group.

Results – Table 1: Add a column with p-values for each item. This helps readers to verify that randomization worked well, variables were balanced and groups homogeneous.

Specify the p-value for the differences found in the EQ-5D utility scores (line 235) as done for total costs.

Table 3 – Is T1 a typo? Wasn’t EQ-5d measured only at the baseline and at 3-month follow-up?  Also, what do you mean by "(n = xx)" reported immediately after 3 months of follow-up?

Discussion – Less than 50% of eligible patients participated to the study as shown in Figure 1. This data and how it may further negatively affect the interpretation of results may deserve room for discussion.

Author Response

Reviewer 3

Thank you for the opportunity to review this paper that contributes knowledge about the cost-effectiveness of e-Health interventions. The paper advanced knowledge about cost-utility evaluation of e-Health in palliative care.

Response: Thank you for reviewing our paper.

The paper would benefit from a few minor revisions, as follows:

• Methods - Specify the statistical tests employed to ascertain differences between the intervention and control group.

Response: we adapted the Methods section. The text is now as follows: “Chi-square tests and independent t-tests were used to analyze whether randomization resulted in comparable groups of patient characteristics across study arms, as well as a Mann-Whitney U test when data were not normally distributed.”

• Results – Table 1: Add a column with p-values for each item. This helps readers to verify that randomization worked well, variables were balanced and groups homogeneous.

Response: we adapted Table 1 by adding a column with p-values for each item.

• Specify the p-value for the differences found in the EQ-5D utility scores (line 235) as done for total costs.

Response: we specified the p-value for the differences in EQ-5E utility scores: “At baseline, also no statistically significant differences were found in EQ-5D utility scores between the intervention group and control group (p-value = 0.35)”

 

• Table 3 – Is T1 a typo? Wasn’t EQ-5d measured only at the baseline and at 3-month follow-up?  Also, what do you mean by "(n = xx)" reported immediately after 3 months of follow-up?

Response: Thank you for mentioning this. These texts (‘T1’ and ‘(n=xx)’) should have been removed before submission. We removed the text.

• Discussion – Less than 50% of eligible patients participated to the study as shown in Figure 1. This data and how it may further negatively affect the interpretation of results may deserve room for discussion.

Response: It is not correct that less than 50% of eligible patients participated in our study. In total 293 patients were screened for eligibility of which 219 were eligible and could be approached to participate in the study. Of the 219 patients who were approached to participate 143 (65%) was willing to participate. Nevertheless, selection bias might have occurred which may have affected generalizability of the study findings. Unfortunately, due to privacy regulations, we were not able to compare data of responders versus non-responders. We added the following information to the Discussion section: “Additionally, selection bias might have occurred, which may affect generalizability of the study findings. Unfortunately, due to privacy regulations, no data was gathered on non-responders, hampering the possibility to compare characteristics of responders and non-responders.”