Review Reports - Virtual Clinics in Cardiology: Do They Provide Equivalent Care and Reduce Travel?

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

I would like to thank the authors for their efforts in conducting this study and for their valuable contribution to the literature on virtual cardiac care.

Introduction: The narrative largely restates the necessity of virtual appointments due to COVID-19 and their ease. This may be condensed to make it more concise. The introduction would benefit from a more precise explanation of the knowledge gap—what precisely is unknown—even though several research are included (e.g., long-term impact on investigations, carbon footprint, and equity in clinical outcomes).

- One way to improve the phrasing "equitable with face-to-face visits" is to use "comparable in clinical effectiveness and quality of care."
Avoid repetition in lines 31–42 by condensing the discussion of the use of virtual care in relation to COVID-19.

Strongen the connection between less travel and environmental impact in lines 43–48, but make it clear that this is still just hypothetical.
Lines 49–52: Offiah et al.'s study raises a significant point regarding possible bias; elaborate on how this limitation will be addressed in the current investigation.
According to the methodology,

lines 60–67: Make it clear if the study is multi-center or single-center.
- The assertion that clinic space was used for allocation without any selection criterion is significant, but it has to be backed up by a more official explanation of how allocation or randomisation bias was reduced.

(Lines 74–76): Virtual consultations: if possible, please specify the ratio of video to telephone consultations, as this may have an impact on the results.

Lines 77–85: Although 2023–2025 is listed as the data timeframe, the study appears to be retrospective. Please specify the precise time period that was covered and the date that data extraction was carried out (Lines 87–94): Although the allocation method is said to be random by availability, the claim that it is "expected to lead to a random distribution" has to be empirically confirmed. Please describe the methodology used to test for demographic comparability between groups.

(Lines 95–107): Please describe the data extraction procedure (automatic vs. human queries) and whether validation was used in addition to the 50-patient random check.

(Lines 108–109): Although the validation with 50 patients is praiseworthy, further information is required: How many differences were discovered? What adjustments were made to the approach? In this validation, was inter-rater reliability examined?

To enhance clarity and completeness, I suggest aligning this section with the STROBE checklist (Strengthening the Reporting of Observational Studies in Epidemiology).

In the dissucion;
• The discussion offers a logical understanding of the results, especially the carbon savings and travel distance reduction. Some passages, nevertheless, could be condensed and are repetitious (for example, lines 210–217 reiterate the similarity of caring without offering fresh perspectives).
• While the assertion that there is "no evidence" for some expected characteristics (homelessness, frailty, etc.) is significant, it would be more persuasive if it were backed up by citations to earlier research or a note of the study's shortcomings in gathering such data.
• While the comparison between patient-initiated preferences and booking clerks is intriguing, it needs to be made clearer how allocation decisions are actually made; otherwise, it is only conjectural.
• For readers who might not be familiar with this analysis technique, the PCA reference (line 201) has to be explained more clearly.
• Although the environmental impact (CO₂ reductions) is a significant point, it should be seen in the context of earlier research to emphasise its uniqueness and scale in relation to related studies.
In the conculsion:
• The key findings are succinctly outlined in the conclusion, especially the similar level of care provided throughout virtual and in-person consultations.
The generalisation to other specialities (lines 228–233), however, might be exaggerated in the absence of comparable data; this has to be reworded with more caution.
• Although the trip decrease is characterised as "relatively small," it would be more accurate to measure its clinical or policy importance (e.g., how such savings could scale across healthcare systems).
• While the idea of improved inquiry coordination is commendable, it might be strengthened with a specific illustration or suggestion (e.g., bundled visits, hybrid models).

Author Response

Dear Reviewer,

We wish to express our gratitude for your time and diligence in reviewing our work. We are grateful for your comments and hope that the following responses will be appropriate.

This is a helpful rephrasing and changed on lines 22, 61, 242

Avoid repetition in lines 31–42 by condensing the discussion of the use of virtual care in relation to COVID-19.

Changes made accordingly to line 40 - 41

Strongen the connection between less travel and environmental impact in lines 43–48, but make it clear that this is still just hypothetical.

Changes made to line 43

Lines 49–52: Offiah et al.'s study raises a significant point regarding possible bias; elaborate on how this limitation will be addressed in the current investigation.

Changes made to line 50

According to the methodology,

lines 60–67: Make it clear if the study is multi-center or single-center.

Changes made to line 63

- The assertion that clinic space was used for allocation without any selection criterion is significant, but it has to be backed up by a more official explanation of how allocation or randomisation bias was reduced.

Changes made to line 64

(Lines 74–76): Virtual consultations: if possible, please specify the ratio of video to telephone consultations, as this may have an impact on the results.

Changes made to line 75

Lines 77–85: Although 2023–2025 is listed as the data timeframe, the study appears to be retrospective. Please specify the precise time period that was covered and the date that data extraction was carried out (Lines 87–94): Although the allocation method is said to be random by availability, the claim that it is "expected to lead to a random distribution" has to be empirically confirmed. Please describe the methodology used to test for demographic comparability between groups.

We expected the allocation to be random, but tested that using the PCA analysis. That analysis showed that whilst the 2 groups were very similar the Virtual Clinic group travelled slightly further (seen in Figure 1). As part of responding to the reviewers’ questions we have provided further demographic information as a Supplementary Document. That document shows that Virtual Clinic patients are also a little older. Both sets of data show that the allocation between the 2 groups was not in fact random. They were very similar, but not identical. We speculate that this is because of patients contacting the organisation and asking for a specific appointment type because of immobility or distance that needs to be driven.

Changes made to Lines 78 to 85

(Lines 95–107): Please describe the data extraction procedure (automatic vs. human queries) and whether validation was used in addition to the 50-patient random check.

Changes made to line 102

(Lines 108–109): Although the validation with 50 patients is praiseworthy, further information is required: How many differences were discovered? What adjustments were made to the approach? In this validation, was inter-rater reliability examined?

Changes made to line 102

To enhance clarity and completeness, I suggest aligning this section with the STROBE checklist (Strengthening the Reporting of Observational Studies in Epidemiology).

STROBE documentation is included with the submission

In the dissucion;
• The discussion offers a logical understanding of the results, especially the carbon savings and travel distance reduction. Some passages, nevertheless, could be condensed and are repetitious (for example, lines 210–217 reiterate the similarity of caring without offering fresh perspectives).

Embarrassingly fair point and changes made in lines 210 -217

While the assertion that there is "no evidence" for some expected characteristics (homelessness, frailty, etc.) is significant, it would be more persuasive if it were backed up by citations to earlier research or a note of the study's shortcomings in gathering such data.

Change made to line 207 to emphasise that the conclusions are directly from our data.

While the comparison between patient-initiated preferences and booking clerks is intriguing, it needs to be made clearer how allocation decisions are actually made; otherwise, it is only conjectural.

Change made to line 64

For readers who might not be familiar with this analysis technique, the PCA reference (line 201) has to be explained more clearly.

Changes made to line 128

Although the environmental impact (CO₂ reductions) is a significant point, it should be seen in the context of earlier research to emphasise its uniqueness and scale in relation to related studies.

Changes made to line 227

In the conculsion:
• The key findings are succinctly outlined in the conclusion, especially the similar level of care provided throughout virtual and in-person consultations.
The generalisation to other specialities (lines 228–233), however, might be exaggerated in the absence of comparable data; this has to be reworded with more caution.

Changes made to line 234

Although the trip decrease is characterised as "relatively small," it would be more accurate to measure its clinical or policy importance (e.g., how such savings could scale across healthcare systems).

Change made to line 239

While the idea of improved inquiry coordination is commendable, it might be strengthened with a specific illustration or suggestion (e.g., bundled visits, hybrid models).

Change made to line 241

Reviewer 2 Report

Comments and Suggestions for Authors

Overall

The authors studied the role of virtual visits in cardiology clinics and tried to answer if this leads to more investigations for the patients and does it save the travel while providing equitable care. This study is important and has clinical meaning. The results provided confidence in continuing to utilize virtual visits in addition to saving travel, ultimately helping with the environment. The paper is written in good english with clear representation of the results with tables and graphs.

Strengths

Adequate power of the study (9000 patients)

Limiting to cardiology clinics is a strength given that the results are not confounded by including several types of clinics from different specialities

The article already discussed the impact of travel more than 20 kms which could lead to differences in the groups but fortunately the number of patients are very few and does not affect the results and analysis has been done by excluding them.

Other Weaknesses / Things to Clarify

There is no limitations section. This should be added for the readers to be aware of what the limitations of the study are.

In line 48 from the introduction section, the paper says more tests are ordered for virtual visits and cites reference 8. But reference 8 actually showed that fewer tests are ordered with virtual consults. The reference is right about more follow-up visits, but not about investigations.

In the discussion (line 211), the authors say care is similar only because the number of follow-up visits was the same. Looking at the number of follow up visits is not enough to state that care is similar.. Typically need to compare outcomes such as hospitalization or mortality etc

Line 88: Expand short forms like GP and A+E for readers outside the UK.

It looks like the analysis was based only on the first visit (virtual or face-to-face). If that is the case, please mention this explicitly in the methods section. This was somewhat mentioned in the discussion.

Regarding travel time, it was discussed/identified that there is a slight decrease in travel because of virtual visits with doctors but overall travel time for investigations are the same. But what about the travel time for in person visits to get the blood draws and other tests done at the same place as doctor consultation, can this be clarified further? If the blood draws and doctor visits happen during same visit for some of these patients, then the advantage of virtual visit saving travel time is negated.

Possible Limitations to Add

Telephone vs video visits are not delineated in the study, this could possibly be a limitation since telephone visits could lead to more investigations than video for obvious reasons.

First consultations often require baseline labs, so both groups may look similar with respect to ordering number of tests/investigations. This could explain why the results are not the same as earlier studies that showed more tests after virtual visits.

Some cardiologists or clinics may order more tests than others. This could affect the results as a potential confounding factor.

Suggestion for future research

Discuss if a hybrid model (first in-person, then follow-up virtual) might lower test numbers and travel time while keeping good patient confidence.

Author Response

Dear Reviewer,

We wish to express our gratitude for your time and diligence in reviewing our work. We are grateful for your comments and hope that the following responses will be appropriate.

There is no limitations section. This should be added for the readers to be aware of what the limitations of the study are.

Limitations section added at the end of the Discussion

Change made to line 48 to clarify that the reference is in relation to the increase in follow up.

Change made to line 211 to clarify that this is about visits made for investigation.

Line 88: Expand short forms like GP and A+E for readers outside the UK.

Change made to line 88

Change made to line 92

There is analysis for both the numbers of tests that were ordered and the number of visits those tests required the patients to make. We accounted for patients having tests done on the same visit by identifying the date / time of the test and then grouping the tests together accordingly. We assumed that tests performed at the same time did not require a further visit. These are referred to in sections 3.4 and 3.5

Possible Limitations to Add

Telephone vs video visits are not delineated in the study, this could possibly be a limitation since telephone visits could lead to more investigations than video for obvious reasons.

Gratefully acknowledged and added to the new Limitations section

Acknowledged but no alteration made

Some cardiologists or clinics may order more tests than others. This could affect the results as a potential confounding factor.

We analysed the behaviour of individual cardiologists to look for this effect. There was surprisingly little variation between the cardiologists. However we felt that it detracted from the overall message of the paper without adding an important further learning. We chose not to include the data. It could be included but would represent a significant change to the paper.

Suggestion for future research

Discuss if a hybrid model (first in-person, then follow-up virtual) might lower test numbers and travel time while keeping good patient confidence.

Changes made to line 227

Reviewer 3 Report

Comments and Suggestions for Authors

I am unclear about the place where the virtual appointments were made. Can you clarify how many patients travelled to the doctor's office to see the doctor in person, and how many patients travelled to the doctor's office to communicate with the doctor by video or phone? How many patients communicated with the doctor by video or phone from home?

Author Response

Dear Reviewer,

We wish to express our gratitude for your time and diligence in reviewing our work. We are grateful for your comments and hope that the following responses will be appropriate.

There were 4672 patients who attended the cardiac appointments by virtual methods. These patients did not travel and were able to stay at their homes.

There were 4773 patients who attended the cardiac appointments Face to Face. These patients travelled to the Doctors Office.

The Clinicians were typically based in the hospital for either mode of consultation.

These are covered in the Methods section.

Reviewer 4 Report

Comments and Suggestions for Authors

Congratulations to the authors for their interesting manuscript providing new insight; anyway I have some comments about it:

Although the booking system may seem random, it is likely that patient self-selection driven by factors such as frailty, digital literacy, and socioeconomic status had a role in determining who opted for virtual appointments. The observed group differences in the PCA analysis further support the presence of this bias. Therefore the authors should acknowledge that the cohorts were not truly randomized and that this constitutes an uncontrolled confounder.

The study encompasses 9,445 patients; however, a numerical summary of baseline demographic characteristics is not provided. Authors should implement this aspect.

The phrase “statistical significance of less than 0.05” is unclear; does it indicate p > 0.05 or p < 0.05? This should be clarified.

The observation that patients in the virtual group tended to live farther away should have been adjusted for, as travel distance is inherently associated with the likelihood of attending a virtual clinic and may have influenced subsequent care patterns. Moreover authors are encouraged to include other successful experiences in virtual visit in cardiology (doi: 10.3390/jcm12020620) in order to empower their discussion.

Author Response

Dear Reviewer,

We wish to express our gratitude for your time and diligence in reviewing our work. We are grateful for your comments and hope that the following responses will be appropriate

This is acknowledged in section 3.1. There is a further expansion of 3.2 in order to explain the nature of the PCA and the presence of a difference between the 2 groups.

The study encompasses 9,445 patients; however, a numerical summary of baseline demographic characteristics is not provided. Authors should implement this aspect.

Further supplementary document provided with the submission including demographic characteristics. We were worried that including this data could prove unhelpful. It doesn’t add any particular learning to the work and so we had excluded it. We thought that including it as a supplementary document would provide the data in a way that wasn’t distracting but helpful for those who required it.

The phrase “statistical significance of less than 0.05” is unclear; does it indicate p > 0.05 or p < 0.05? This should be clarified.

Change made to line 159

Discussion in section 3.1 with an extended section 3.2 to explain PCA

Moreover authors are encouraged to include other successful experiences in virtual visit in cardiology (doi: 10.3390/jcm12020620) in order to empower their discussion.