Impact of the COVID-19 Pandemic on Medical Oncology Workload: A Provincial Review

Round 1

Reviewer 1 Report (Previous Reviewer 1)

Thank you for the opportunity to review this resubmission detailing how medical oncology workload in Nova Scotia over a 7 year period (including the effects of COVID-19).

I thank the authors for considering my previous comments and making substantial efforts to revise the manuscript. I believe the manuscript has significantly improved as a result and makes a much stronger contribution to the literature.

 I have the following comments on the current submission:

 

Methods – Think it would aid the reader for you to clarify what data OPIS covers – I am assuming just the Nova Scotia Cancer Centre, but it is a little unclear.

Results – I think adding in the additional analysis by estimated tumor site and by MO is extremely valuable and enlightening, thank you for this additional analysis. The only thing that perhaps needs to be clarified in the manuscript is that for any tumor specific data, the numbers are estimates only based on treating physicians typical patient load. While you have noted this in the methods, I think the reader should be reminded in the results.

 

Discussion – the first paragraph contains useful background information relevant to calculation of workload and would be best added to the introduction to give additional context to the justification for the current work. It may be useful to the reader to remind them of the study aim at the beginning of the discussion, and then continue with a summary of your findings.

Line 495-500 – this is largely repeating the information in 377-382 so could be removed.     

Author Response

Dear Current Oncology Editors.

Thank you for reviewing our manuscript entitled “Impact of the COVID-19 Pandemic on Medical Oncology Workload: A Provincial Review” by Drs Sheridan, Colwell, Lamond, Macfarlane, Rayson, Snow, Wood and Ramjeesingh. We have reviewed the reviewers and editors’ comments and have made the following changes (in bold):

Reviewer 1:

Thank you for your valuable opinions. We agree that your comments have contributed to what we believe is a much stronger manuscript. To address your concerns:

1. Methods – Think it would aid the reader for you to clarify what data OPIS covers – I am assuming just the Nova Scotia Cancer Centre, but it is a little
   1. Yes, OPIS is used for the We have added that clarification into the methods section
2. Results - I think adding in the additional analysis by estimated tumor site and by MO is extremely valuable and enlightening, thank you for this additional analysis. The only thing that perhaps needs to be clarified in the manuscript is that for any tumor specific data, the numbers are estimates only based on treating physicians typical patient While you have noted this in the methods, I think the reader should be reminded in the results.
   1. Thank you for your comment. We have incorporated the following line into page 5 “Additionally, it should be emphasized that these are estimates based on treating physician’s typical patient load.” We hope this address your concern.
3. Discussion – the first paragraph contains useful background information relevant to calculation of workload and would be best added to the introduction to give additional context to the justification for the current work. It may be useful to the reader to remind them of the study aim at the beginning of the discussion, and then continue with a summary of your

 

1. Thank you for your comment. We have added the following line at the beginning of the discussion “This study has attempted to provide an account of the changes in MO workload over time and review the impact of the COVID-19 pandemic on the workload in hopes to aid in future discussions on resource allocation planning in ”

4. Line 495-500 – this is largely repeating the information in 377-382 so could be
   1. There are no lines 495-500 in the manuscript (it ends currently at 487) so we are unable to make any changes.

 

We hope these changes are sufficient and meets the reviewers and editor approval.

I would also like to take the opportunity to reconfirm the author list and their affiliations. We look forward to hearing from you soon about the manuscript.

Reviewer 2 Report (Previous Reviewer 2)

Following revisions, it reads very well, and merits publication. No further comments.

Author Response

Dear Reviewer 2:

 

We thank you for reviewing the manuscript and your valuable comments from before which we feel enriched this manuscript. There were no comments to follow up on after the revision and no changes made.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Thank you for the opportunity to review this manuscript which describes data over the last 9 years regarding the number and type of patient encounters in a medical oncology service in Nova Scotia. Strengths of this paper included examination of the data over an extended time period, allowing helpful comparisons of increases in encounters over time. However, as the paper is currently crafted, I felt it sat somewhere between a piece of original research and a commentary, but didn’t clearly fit in to either category which meant I found it difficult to know what to take away or conclude from it. I believe there was a missed opportunity to explore the data in more a detailed manner, which could have substantially added to the authors argument for revisions to workload metrics.

 

I have listed some general comments below, which I hope the authors might find helpful when revising the paper.

 

Introduction

Overall I thought the introduction covered the relevant points at only a superficial level and could have benefited from some additional context and relevant data to build the case for the importance and relevance of this study.

In the paragraph from line 47-54, you mention limitations in care delivery, increased demands and impact but this is all very vague – provision of some concrete examples and definitions of what you are referring to would aid your introduction considerably.

I was also unclear about the specific aim of the study. Is it to highlight that current workload metrics do not capture workload accurately? Is it to capture changes in workload as a result of COVID? Is it to argue that more resources are needed, and if so what type of resources?

 

Methods

What was the justification for choosing 2014 as the start date?

Line 79 – I’m assuming that the number of medical oncologists increased from 9 to 13 across the study period,although this is not made clear. But this has significant implications for how the data is interpreted.

 

Results

Continuing on from my previous comment regarding a potential increase (?) in the number of medical oncologists over the time period, is this accounted for anywhere in your results? While I don’t have the data obviously so I am just hypothesising if there was an increase from 9 to 13 oncologists (an increase of >30% in staff resource) then for at least consults and follow-ups workload presumably would have changed very little per FTE medical oncologist.   

I think there is a missed opportunity here to enrich this dataset with some additional contextual data. For example, is there differences in the patterns of encounters dependant on tumour/treatment type, or subgroups of the population? If so, perhaps any workload metric should account for these factors – indeed a line in the discussion (line 144) suggests tumour type is accounted for in the metric. Given the manner of data collection (effectively a record audit) this should be relatively straightforward to extract.

Discussion:

As it currently reads, I found the discussion more like a commentary rather than a considered and reflective account of the study results, particularly in terms of consistency (or not) with existing data and interpretations of specific findings.

For example, does the increase in new consults map closely to reported increased incidence rates? It has also been widely reported that COVID affected the number of people participating in cancer screening, seeking advice for unusual symptoms etc which all likely affect the number of people getting diagnosed with cancer in 2020. As your centre also experienced a drop in new consults in 2020 compared to 2019, did this align with other data regarding less screening etc?

In lines 161-175 you discuss changes in treatment regimes etc and how these factors are leading to more CFT encounters. While I don’t doubt this is the case, it would be helpful to back up this argument with some relevant citations. In addition, how much of the increase does this explain over time? This is likely difficult to quantify but I think an interesting point which has implications for understanding workload. Eg the number of teletoxicities encounters doubled from 2016-2018. What drove this? Was a new chemotherapy drug introduced in that period? Were more nurses hired during this time so more calls to patients could be made?    

 

Minor comments:

Abstract – include some description of what data is covered – eg is it national data across Canada?

Line 26 – ensure you back up all claims with citations (eg cancer is leading cause of death in Canada)

Line 79 – what is a physician-extender? Can you clarify for readers not familiar with the term

Author Response

Dear Reviewer 1,

Thank you very much for your comments and revision suggestions.

We have made an effort to include the revisions within the document itself and have addressed all points within this letter of response.

Introduction:

We have added some more detail in the introductory paragraph about the limitations on surgery/radiation oncology during the pandemic, for clarification, as well as supportive citations.

The aim was to demonstrate the impacts of COVID 19 on cancer care and to use this as an example to highlight the need for an improved workload metric.

Methods:

2014 was initially chosen to be representative of a time pre-dating regular use of immunotherapy in Canada. The first approval of immunotherapy in Nova Scotia was for Nivolumab in 2016. We had intended to capture a few years’ worth of data before this, as well as after. We wanted to be able to provide some sort of information regarding how regular use of immunotherapy may alter workload.

We have provided some further insight into the number of physicians during the study period (including new hires) and some evaluation of the implications on the interpretation of the data/results.

Results:

See above point in regard to variation in physician numbers of the study period. Adjustments accounting for the increase in physician number have been made.

As for the influence of various tumor types, we were somewhat limited by the way in which our Oncology Patient Information System logs patient encounters. They are not logged by disease site/tumor type, but rather by physician. We did break down the numbers by specific physician and also divided them into rough estimates of tumor type based on sites that that physician is known to treat (example – someone who sees breast and gastrointestinal cancers had their workload divided in half, with one half assigned to the “breast” group, the other to “gastrointestinal”). This was done in an effort to gather some information about the variation in tumor type and its impact on workload over the study period.

Discussion:

Thank you for your points regarding the discussion section. We have made several revisions, within the limitations of available data and reasonable adjustment.

We do not have the data to confirm or deny variation in screening rates during the COVID19 pandemic at our specific centre, though have been able to add some of the provincial data that has been recently released. It was not feasible for us to access this data otherwise. As cited, there does seem to be some correlation between decreased screening, diagnosis and thus new consults during the early pandemic. Also as cited, the system should be preparing to meet further increased demand in years to come. We have included some data to support this.

(161-175) These are all important points we included in an effort to suggest there are many important factors to consider in the landscape of care delivery and that they are variable over time. We included a few years of data prior to approval of immunotherapy in the province to try to capture the effect of this change, considering increased use of immunotherapy as one such example. We have included some discussion in this regard, though it is difficult to quantify and the analysis of potential variables impacting the encounter types (example date of approval for immunotherapy drugs, etc) will be considered for addition to a future review. 

In terms of personnel, other than the physician increase (with a nursing counterpart per new physician), there was no other staff increase. No manpower was added, even in response to that significant increase in encounters.

Minor comments:

Abstract - description of (provincial) data has been added

Line 26 - will add citation (thank you)

Line 79 - definition for physician extender has been added (in this case a Nurse Practioner)

Thank you very much for your thoughtful suggestions.

M. Sheridan

 

Reviewer 2 Report

The paper reads v well and makes a couple of very signficant points that the metric of the new cancer patients per year is not appropriate / accurate reflection of M Onc workload, and that virtual care is on the increase before and of course during the covid pandemic. The data are well presented, and there is no need to make any changes. 

Regarding the discussion session, and for this paper to become more of a reference paper on this topic, perhaps it would be desirable to discuss the M Onc workload issue in a more global framework, and explain the differences seen between low middle income and high income countries (e.g. Funduytus JGO). There are also publications from Spain and the UK, that can also be discussed and referenced.

Worth also discussing and referencing, how increased treatment options, which is mentioned eg immunotherapy and targeted treatments, have an impact on workload, perhaps by providing standard of care treatments in 2020 compared to 2014 in Canada (or recommendations from Cancer Care Ontario). Also worth discussing increasing patients/family expectations, how they impact on workload. There is a mention on telephone calls, may be interesting to know what per cent of those relate to direct patient care, and what percent to relatives questions, concerns etc.

Regarding the issue of how to measure workload accurately, it may be worth discussing what this proposed algorithm for estimating workload should include, eg different types of cancer may be associated with different workload, different stage, availability or absence of palliative care or adequate primary care, availability of specialist nurses or not. Also to discuss the need for regular review and change of this algorithm as new advances in cancer care, provide more avenues of treatment, e.g. now immunotherapy for locally advanced (not metastatic) NSCLC.

Finally a paragraph in the discussion dedicated to the impact of covid to cancer care, by reviewing the literature would also be welcome.

 

Author Response

Dear Reviewer 2,

Thank you for your comments and revision suggestions.

Your comments in regard to the discussion section are well-received and we have done our best to revise this section in particular, to become more of a reference paper on the topic. Further literature review has been conducted on primary topics, including the evolution of cancer care (novel treatments and broader applications), personnel resources, and consideration of disease site, as well as the impacts of COVID-19 on cancer care and various associated services.

Thank you kindly,

M. Sheridan