Provider beliefs, practices, and perceived barriers to dietary elimination therapy in eosinophilic esophagitis

Abstract

INTRODUCTION:

Despite effective dietary treatments, physicians prefer medications for eosinophilic esophagitis (EoE).

METHODS:

We conducted a web-based survey of providers to assess the perceived effectiveness, practice patterns, and barriers to EoE dietary therapy.

RESULTS:

Providers view diet as the least effective treatment. The greatest barrier was the belief that patients are disinterested and unlikely to adhere (58%). With less access to dietitians (56%), non-academic providers often manage diets without dietitian guidance (41%).

CONCLUSIONS:

Given high patient acceptance for diets and multiple treatment options for EoE, clinicians need evidence-based knowledge on EoE diets, access to dietitians, and awareness of patient preferences.

INTRODUCTION

Dietary therapy, which identifies and removes food antigen triggers, is an effective non-pharmacologic treatment for eosinophilic esophagitis (EoE). Studies demonstrate that the efficacy of elimination diet is up to 71%, similar to that of topical steroids; however, comparative trials are lacking.^1–4 Success with diet therapy can be augmented by partnering with a dietitian, but access is not universal and diet education often falls to the clinician.

Due to concerns about potential medication side effects and costs, patients may express interest in dietary approaches for managing EoE.^5–7 Studies of provider practice patterns in EoE consistently show that physicians prefer to use medications over diet.^8–12 We aimed to describe provider practices around dietary approaches in EoE and identify beliefs about, barriers and needed resources to support diet therapy.

METHODS

In this cross-sectional online survey, a primary cohort of gastroenterologists was recruited from American Gastroenterological Association, North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition, and the American Partnership for Eosinophilic Disorders listservs, and a secondary validation cohort of gastroenterologists and allergists was recruited from Medscape. The survey was developed by authors and iteratively refined with feedback from a dietitian with expertise in EoE, consisting of 20 questions related to beliefs about effectiveness, practice patterns, barriers and needs to recommending and executing dietary therapy for EoE (Supplementary Appendix). Data were collected between August 2021 to January 2022. Descriptive statistics were performed and bivariate analysis was used to compare differences between practice settings. Diet was defined as the providers’ preferred strategy, including 6-food (6FED), 4-food, 2-food, 1-food elimination, allergy test based, and elemental diets.

RESULTS

Respondents from the primary (n=94) and secondary validation (n=195) cohorts had a comparable spread across practice settings, geographic regions, and access to dietitian support (Table 1). Rating the estimated effectiveness of diet against topical corticosteroids (TCS) and proton pump inhibitor (PPI) on scale from 0% to 100%, the mean efficacy of diet (46.8%, SD 31.2) was lower than TCS (74.3%, SD 29.8, p=0.00) and PPI (56.5%, SD 26.8, p=0.062) for short-term treatment. For long-term treatment, diet therapy (mean 49.5%, SD 38.9) was rated as less effective than TCS (mean 64.4%, SD 31.6; p=0.012), but not different than PPI (53.3%, SD 27.4; p=0.72).

TABLE 1.

Overall provider characteristics.

		Primary Cohort		Validation Cohort		P-value
Provider Type	GI	94	70 (80.5%) Faculty/Staff 10 (11%) APP 5 (5.8%) Trainees	150	153 (78.5%) Faculty/Staff 23 (11.8%) APP 19 (9.7) Trainees	---
	Allergy	0		45
Practice Setting	Private	43 (53.8%)		113 (64.2%)		0.051
	Academic	26 (32.5%)		34 (19.3%)		0.062
	VA	0 (0%)		3 (1.7%)		0.553
	Military/govt	0 (0%)		3 (1.7%)		0.553
	Hospital-Based	11 (13.8%)		23 (13.1%)		1.000
Location	Northeast	15 (16.0%)		30 (15.4%)		1.000
	Midwest	20 (21.3%)		29 (14.9%)		0.184
	South	26 (27.7%)		52 (26.7%)		0.888
	West	22 (23.4%)		46 (23.6%)		1.000
	Undisclosed	11 (11.7%)		38 (19.5%)		0.131
Patient Volume (number of EoE patients annually)	None	0 (0%)		1 (0.5%)		1.000
	1-5	5 (5.8%)		10 (5.1%)		0.945
	6-19	43 (49.4%)		36 (18.5%)		0.00
	20-50	25 (28.7%)		74 (38.0%)		0.057
	>50	13 (14.9%)		74 (38.0%)		0.00
Patient Population	Adults	62 (71.3%)		96 (49.2%)		0.007
	Children	16 (18.4%)		15 (7.7%)		0.024
	Both	8 (9.2%)		84 (43.1%)		0.00
Access to RD		58 (66.7%)		131 (67.3%)		0.359

While the majority of providers agreed that dietitians (85%) or physicians (72%) should be responsible for educating patients about EoE diets, access to a dietitian was not universal. Non-academic providers reported disproportionately less access and support from a licensed dietitian (56% vs 85% academic, p=0.011; Table 2) and were more likely to manage diet alone (41% vs 23% academic, p=0.033). The most commonly recommended diet was the empiric 6FED (29%), which did not vary between practice settings (p=0.057).

TABLE 2.

EoE dietary therapy practice patterns based on practice setting (Primary cohort)

	Academic (n=26)	Non-academic (n=54)	P- value
Access to RD	22 (84.6%)	30 (55.6%)	0.011
Diet Management
Manage diet alone	6 (23.1%)	21 (41.2%)	0.033
Refer to RD	18 (23.4%)	8 (15.7%)
Refer to allergist	2 (7.7%)	12 (23.5%)
Refer to RD + allergist	8 (30.8%)	10 (19.6%)
Recommended initial diet approach
Empiric 6FED	8 (30.8%)	15 (29.4%)	0.057
Empiric 4FED	4 (15.4%)	7 (13.7%)
Empiric 2FED	7 (26.9%)	12 (23.5%)
Empiric 1FED	6 (23.1%)	2 (3.9)
Allergy test directed diet	0 (0%)	10 (19.6%)
Elemental diet	1 (3.9%)	1 (2.0%)
Other	1 (3.9%)	4 (7.8%)
Resources used to initiate diet
Dietician referral	24 (32.9%)	26 (22.6%)	N/A
Verbal education	16 (21.9%)	39 (33.9%)
Printed materials I’ve collected	14 (19.2%)	18 (15.7%)
Handouts created by me or my institution	7 (9.6%)	6 (5.2%)
Websites and apps	7 (9.6%)	19 (16.5%)
Patient advocacy groups	5 (6.9%)	3 (2.6%)
None used	0 (0.0%)	2 (1.7%)

The greatest barrier to EoE diet therapy was the perception that patients were not interested in, unlikely to accept, or adhere (58%); only a few providers (14%) felt that their personal lack of knowledge and experience in the diet was a barrier (Figure 1). Other barriers were concern about multiple endoscopies (36%), provider’s concern that the diet was already excessively restricted (24%), and physician preference to use medication (20%). Given limited clinical resources on EoE dietary therapy, (79%) expressed desire to use a provider-focused tool to initiate diet therapy.

FIGURE 1.

Barriers to recommending or starting EoE dietary therapy

Findings that dietitians should be responsible for educating patients about diets, academic providers have more access to dietitian support compared to non-academic (94% vs 61%, p≤0.01), 6FED as the most commonly recommended diet (34%), provider-perceived barriers, and desire to use a provider-focused resource were observed in our validation cohort.

DISCUSSION

Our findings that providers view dietary therapy as the least effective for EoE, lack access to dietitian support in community settings and perceive low patient acceptance and adherence to diet suggest that they are not prone to support diet therapy in EoE. In contrast, a study of 42 EoE patients on maintenance diet therapy revealed high acceptance; 96% reported benefits outweighed the inconveniences and would it recommend to others.⁶ Our prior patient survey found that barriers to TCS included side effects and preferences for a “natural” approach, but that patients with EoE value but infrequently experience shared decision making (SDM) about treatments.⁷ Additionally, our analysis of patient experiences from online EoE communities demonstrated that individuals struggle with providers who lacked disease-related knowledge or did not support treatment choices.¹³ Taken together, this calls attention to a discordance between provider and patient preferences, incomplete communication, and potentially unrecognized provider knowledge gaps.

Limitations of our study include distribution through multiple listservs and inability to calculate an accurate response rate, but we estimate that the response rate was similar to prior EoE provider surveys (i.e. less than 25%). Our survey is subject to response bias, so our findings may underestimate how the average clinician with less EoE experience may view this treatment. A distribution of our primary and validation groups across practice settings, geographic location, and access to dietitians supported the generalizability of our findings. Our study included providers within the United States and may not reflect perspectives in other countries.

Our findings have implications for both clinicians and patients. For providers, recommending diets without dietitian guidance may reduce the chance of successful implementation and a poorly executed diet may be misinterpreted as nonadherence, nonacceptance, or refractory disease. Not only should providers equip themselves with evidence-based resources (including for less restrictive diets) to address knowledge gaps, but there is a critical need for the development and dissemination of educational tools to appropriately implement EoE dietary therapy.^14,15 For patients, our findings highlight opportunities for improved patient-provider communication to allow informed treatment decisions and better adherence to diet. Given the preference-sensitive nature of EoE treatment and high patient acceptance for dietary treatments, educating providers with evidence-based knowledge about treatment options, understanding patient preferences and engaging in SDM, and improving dietitian-led education are essential in providing high-quality EoE care. These findings are particularly salient in light of the recent FDA approval of dupilumab, offering another treatment option with tradeoffs (e.g. financial cost, self-administration of an injectable medication, adherence, side effects) to consider.