Improved Healthcare Access Reduces Requirements for Surgery in Indigent IBD Patients Using Biologic Therapy: A ‘Safety-Net’ Hospital Experience

Round 1

Reviewer 1 Report

 

Statistical values (odds ratio, 95% CI, and p value) shown in lines 28, 192, 194, and 263 should be checked again. Some of them must be wrong.

 

As the authors stated that “The rate of therapy maintenance may affect IBD related surgery” (lines 237-238), the rate must be different in groups. They should explain this point clearly.

 

Author Response

Statistical values (odds ratio, 95% CI, and p value) shown in lines 28, 192, 194, and 263 should be checked again. Some of them must be wrong.

This has been corrected, thank you for catching the error. The numbers in line 28 are identical to numbers in line 192, 263, and Table 3 marked by ? (alpha). The errors have been corrected and the values adjusted.

We recalculated the values in line 194 and updated the OR and 95%CI and added the associated p-value.

Reviewer 2 Report

I have the following remarks: 

1. The first part of the discussion repeats the data from the introduction.

2. Is the number of imaging test related only to health care, like suggested by Authors,  or it may also result from the more severe course of IBD?

 

 

 

Author Response

Reviewer 2

The first part of the discussion repeats the data from the introduction.

We regret the redundancy. This has now been corrected by editing the discussion and introduction sections. They should now read somewhat more differently while discussing similar concepts.

2. Is the number of imaging test related only to health care, like suggested by authors,  or it may also result from the more severe course of IBD?

This is an interesting point. The number of imaging studies may be related to healthcare but could also reflect the more severe course of IBD in the non-FAP patients.

We now address this point stating:

"To identify potential explanations for the difference in surgical rates, we found that the overall patient set (both FAP and non-FAP) had a higher cutoff value for imaging studies compared to non-FAP patients alone (3.5 vs. 2.5 studies) prior to undergoing surgery in the decision tree analysis. These findings suggest that the number of imaging studies was the best predictor of surgery among the two groups, and that the average patient who required surgery underwent more necessary imaging prior to surgery than non-FAP patients alone. This possibly suggests that the number of imaging studies is a manifestation of differences in healthcare system access between FAP and non-FAP patients, resulting in differing risks of surgery. In other words, it may be easier for FAP patients to obtain required imaging studies compared to non-FAP patients due to lower bur-den from prior authorization or co-payments. We speculate that because FAP patients are better able to get necessary imaging studies to inform treatment adjustments following initiation of biologic therapy, they had better outcomes and less need for surgery. On the other hand, non-FAP patients potentially had more difficulties obtaining necessary imaging studies to monitor their disease until it was severe enough to require surgery.  The finding of fewer imaging studies in the FAP vs. non-FAP group is at face value somewhat surprising. While not yet clear, it is also possible that the lower number of imaging studies in the FAP group could reflect the lower overall disease severity in this group which created a less frequent need for imaging, compared to the non-FAP group which may have received more imaging studies based on somewhat less well controlled disease. We observed that FAP patients had a lower number of complex CD behaviors (50.7% for FAP versus 70% for non-FAP) while similar average age and average body mass index, but shorter disease duration than our non-FAP population. While additional studies are needed to validate this, it may suggest that FAP administration may reduce disease severity and the associated costs of managing treatment."

Reviewer 3 Report

A quite important and comprehensive study with significant practical implications.

A minor suggestion: In the discussion section, please comment on why non-FAP patients were more likely to be female than FAP patients, as well as why they were more likely to have complex CD behavior (lines 174, 175). Also about the cause of the significant differences observed among the two groups with respect to race 176 (lines 176, 177). 

1.    What is the main question addressed by the research?

Although previous studies suggest that reduced ability to access health care services is an important risk factor for poor outcome in patients with IBD, it is not clear whether improvement of this ability results in improved disease outcome and proper use of resources in these patients. Given that IBD patients who require biologic agents for disease management are at increased risk of poor outcome, the authors of this study evaluated the effects of implementing a financial assistance program (FAP) in a group of indigent IBD patients who required treatment with biologic agents.

2. What does it add to the subject area compared with other published
material?

This study is probably the first of its kind to evaluate the effectiveness of interventions aimed at improving access to health care for patients with IBD and poor economic status. Due to the nature of the sample it seems that the monitoring of the various aspects of the patients' treatment was essential and effective.

4. Any specific improvements the authors should consider regarding the
methodology? What further controls should be considered?

The retrospective nature is a major drawback of this otherwise interesting paper. Possibly, the study of the same parameters in patients with poor income who need biological agents (e.g. patients with rheumatoid arthritis) would probably increase the strength of the results obtained in the patients with IBD.

The authors found that patients, who participated in the financial assistance program and were treated with biologic agents, needed fewer surgeries to treat the disease and had a greater number of imaging procedures performed. The authors should provide more detailed and clearer explanations for these results.

5. Are the conclusions consistent with the evidence and arguments
presented and do they address the main question posed?

Transferring the results to real economic data would possibly lend more prestige to the study and help the reader to better understand the meaning of the study.

6. Are the references appropriate?

 

The bibliographic references are sufficient and relevant to the topic.

Author Response

A quite important and comprehensive study with significant practical implications.

A minor suggestion: In the discussion section, please comment on why non-FAP patients were more likely to be female than FAP patients, as well as why they were more likely to have complex CD behavior (lines 174, 175). Also about the cause of the significant differences observed among the two groups with respect to race 176 (lines 176, 177). 

1. What is the main question addressed by the research?

Although previous studies suggest that reduced ability to access health care services is an important risk factor for poor outcome in patients with IBD, it is not clear whether improvement of this ability results in improved disease outcome and proper use of resources in these patients. Given that IBD patients who require biologic agents for disease management are at increased risk of poor outcome, the authors of this study evaluated the effects of implementing a financial assistance program (FAP) in a group of indigent IBD patients who required treatment with biologic agents.

2. 
   What does it add to the subject area compared with other published
   material?

This study is probably the first of its kind to evaluate the effectiveness of interventions aimed at improving access to health care for patients with IBD and poor economic status. Due to the nature of the sample it seems that the monitoring of the various aspects of the patients' treatment was essential and effective.

4. 
   Any specific improvements the authors should consider regarding the
   methodology? What further controls should be considered?

The retrospective nature is a major drawback of this otherwise interesting paper. Possibly, the study of the same parameters in patients with poor income who need biological agents (e.g. patients with rheumatoid arthritis) would probably increase the strength of the results obtained in the patients with IBD.

The authors found that patients, who participated in the financial assistance program and were treated with biologic agents, needed fewer surgeries to treat the disease and had a greater number of imaging procedures performed. The authors should provide more detailed and clearer explanations for these results.

5. 
   Are the conclusions consistent with the evidence and arguments
   presented and do they address the main question posed?

Transferring the results to real economic data would possibly lend more prestige to the study and help the reader to better understand the meaning of the study.

6. 
   Are the references appropriate?

 

The bibliographic references are sufficient and relevant to the topic.

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

The authors have performed a retrospective cohort study comparing outcomes in indigent IBD patients who received a financial assistance program and those who did not. There were multiple differences between the groups at baseline, but it was found that there was no difference in corticosteroid free remission rates, but surgical rates were significantly lower in the FAP group but confidence intervals on this estimate were wide (OR: 0.13, 95%CI [0.02-0.82]). The findings of the study are interesting, but the major limitation of the study is the retrospective nature and the potential for confounding, which is acknowledged in the discussion. I have the following suggestions:

 

• For remission and use of physician global assessment, were any validated tools used? What was the timeframe to be corticosteroid free and were oral and topical corticosteroids included?
• Was any consideration made for propensity matching to control for demographic differences between the cohorts?
• Were surgeries only considered in the first 12 months
• Combining Crohn’s disease and UC for surgery leads to significant heterogeneity, were these considered separately and do the authors think the findings are valid if they were combined between CD and UC?

Author Response

Reviewer 1

We appreciate the helpful comments and have incorporated changes into the revised manuscript using yellow highlighting.

1. For remission and use of physician global assessment, were any validated tools used? What was the timeframe to be corticosteroid free and were oral and topical corticosteroids included? 

 

Response: We regret not having presented this more clearly. These evaluations were based on inflammatory markers (C-reactive protein and fecal calprotectin), and clinical remission (Based on patient symptoms) and endoscopic remission. We assessed corticosteroid free remission at 6 and 12 months. We also included patients on oral prednisone. This is now stated in methods as:

 

2. Was any consideration made for propensity matching to control for demographic differences between the cohorts? 

Response: This is correct, no attempt was made to control for demographic characteristics of the cohorts.

 

3. Were surgeries only considered in the first 12 months. 

Response: No surgeries were beyond 12 months.

 

4. Combining Crohn’s disease and UC for surgery leads to significant heterogeneity, were these considered separately and do the authors think the findings are valid if they were combined between CD and UC?

 Response: Although, we combined both UC & CD, we also included data for both groups separately.

 

Reviewer 2 Report

In the manuscript by Gu, they have assessed the system, financial assistance program for the health care outcome of IBD patients requiring biologic therapy. The financial program must be attractive and useful for indigent IBD patients. As this kind of program should become prevalent in the world to improve the quality of the treatment of IBD patients, the presented study must be useful for the development of good systems in the future. Here are the comments to the authors:

 

Major

1. Although the system, financial assistance program is attractive, it is not clear if they needed to pay less amount compared to non-FAP patients who have the insurance. They must have explained a part by stating that biologic agents were provided at "significantly discounted co-pay" in patients enrolled in FAP. Please explain more detail of the program. Furthermore, it is not clear who could be the qualified individuals. The authors mentioned that it is based on household income in relation to the Federal Poverty Income Level. Is there any simple image of income amount that can predict the qualification?
2. The presented data showed that the remission rate of IBD was similar, but the incidence of surgery was less in patients enrolled in FAP compared with those without FAP. The maintenance therapy rate with biologic could affect the rate of IBD-related surgery. Not only the access to biologic but also the maintenance therapy could affect the rate of IBD-related surgery. Please indicate the rate of maintenance therapy in Table 1.
3. Did the data of Table 1 is from the number of patients or from the number of unique biologic prescriptions? The demographics should be shown by the number of patients, and the number of biologic prescriptions in each patient should be a factor of FAP and non-FAP groups. As the access to biologics is easier in the FAP group, the number of biologics used in each patient might be higher.
4. The data showed that a significantly lower proportion of FAP patients had previous IBD-related surgeries. What could be the cause of the data? Is the data a limitation the study or the effect of FAP? Please discuss this issue in the discussion.

 

Minor

1. Line 51. ED should be spelled out.
2. Line 55. HRU should be spelled out.
3. Line 83 What is "previous codes"? It is not clear indeterminate colitis was easily extracted.
4. Line 101. IRB approval number should be shown.
5. Line 151. the p values "0.464" and "0.337" were different from the data presented in Table 2.
6. Table 1. What is "a" indicated as superscript at the factor, "Complex CD behavior”.
7. Average number of imaging studies in a period could be listed in Table 2.

Author Response

We appreciate the useful comments and have incorporated these changes in the now revised manuscript which are shown in yellow highlighting.

Reviewer 2

1. Although the system, financial assistance program is attractive, it is not clear if they needed to pay less amount compared to non-FAP patients who have the insurance. They must have explained a part by stating that biologic agents were provided at "significantly discounted co-pay" in patients enrolled in FAP. Please explain more detail of the program. Furthermore, it is not clear who could be the qualified individuals. The authors mentioned that it is based on household income in relation to the Federal Poverty Income Level. Is there any simple image of income amount that can predict the qualification?

 

Response: FAP is available for any patient who applies and qualifies based on income, expenses, and numbers of household members. Access to additional approval criteria is not available. When patients are approved, they pay a copay for each infusion dose.

 

2. The presented data showed that the remission rate of IBD was similar, but the incidence of surgery was less in patients enrolled in FAP compared with those without FAP. The maintenance therapy rate with biologic could affect the rate of IBD-related surgery. Not only the access to biologic but also the maintenance therapy could affect the rate of IBD-related surgery.

 

Response: The rate of therapy maintenance may affect IBD related surgery which is one reason why we believe that access to PFA may improve outcomes for these patients. Escalating biologic therapy could enhance endoscopic remission, minimized surgery, hospitalizations and imaging required per patient.

 

3. Did the data of Table 1 is from the number of patients or from the number of unique biologic prescriptions? The demographics should be shown by the number of patients, and the number of biologic prescriptions in each patient should be a factor of FAP and non-FAP groups. As the access to biologics is easier in the FAP group, the number of biologics used in each patient might be higher. The data showed that a significantly lower proportion of FAP patients had previous IBD-related surgeries. What could be the cause of the data? Is the data a limitation the study or the effect of FAP? Please discuss this issue in the discussion.

 

Response: We now state: “Another possible limitation of this study is that our data show a significantly lower proportion of FAP patients had previous IBD-related surgeries. There are several causes of this which could include more limited access to HRU; it is possible but unlikely that these patients were simply ‘healthier’ given their similar rates of CFSR. The FAP patients had fewer IBD-related surgeries and had a shorter time between prescription and initiation of biologic which may point to and are consistent with more comprehensive therapy management. Additional studies which evaluate such patient characteristics are therefore warranted which may consider this possibility. Furthermore, this study reports a single center experience at a ‘safety-net’ hospital system in Dallas, TX, so the extent to which our findings may be generalizable to other indigent IBD populations in other regions of the US remains unclear.”

Minor

1. Line 51. ED should be spelled out.This has now been changed
2. Line 55. HRU should be spelled out.This has now been changed in lines 45 and 46
3. Line 83 What is "previous codes"? It is not clear indeterminate colitis was easily extracted.This has now been changed to reflect ICD9 coding as well.
4. Line 101. IRB approval number should be shown.Added
5. Line 151. the p values "0.464" and "0.337" were different from the data presented in Table

 

Response: This has now been changed – to values in Table 2, check which are correct

 

Table 1. What is "a" indicated as superscript at the factor, "Complex CD behavior”. 

 

Response: We now state:

“^aMultivariable logistic regression analysis performed adjusting for age, gender, race, IBD diagnosis, prior IBD-related surgery, disease duration, and complex CD behavior”

Average number of imaging studies in a period could be listed in Table 2. 

Response: This is described in more detail in Figures 1 and 2, therefore we have now added the following information to the results section:

“In FAP patients, CT and MRI of the abdomen/pelvis was the first ‘splitting’ variable, and a best cut-off level of 3.5 studies was identified. A lower best cut-off level of 2.5 imaging studies was identified compared to FAP patients (Fig. 1 and 2).”

 

Round 2

Reviewer 1 Report

The authors have addressed the suggested comments there are just a couple of remaining issues:

• in your response you suggested that surgeries beyond 12 months were included but in the paper you now say it is only in the first 12 months. Please clarify
• Please provide a definition of corticosteroid free remission in the paper

Reviewer 2 Report

The authors partially respond my queries.

 

1. As I mentioned in the original review, the rate of maintenance therapy could be stated in Table 1. They must be able to extract the data.
2. The number of biologics used in each patient might be higher in the FAP group. It could also be shown in the presented data.
3. Table 1. Is the explanation of "a" correct? Did they show the data of logistic regression analysis here? It is odd to add "a" just at the factor "Complex CD behavior ".
4. Average number of imaging studies in a period could be listed in Table 2. They just showed cut of values of imaging studies in Figure 1 and 2 and did not show the detailed numbers in each group.
5. Line 273. CFSR should be CSFR.