Digital health for functional gastrointestinal disorders

Abstract

Background:

Functional gastrointestinal disorders are a common but challenging set of conditions to treat. Gastroenterology practices often struggle to meet the needs of patients with functional disorders given the need for careful monitoring, frequent communication, and management of stressors that occur outside of the clinical setting. In recent years, applications in digital health have created a new set of tools that can improve the care of these patients, including through improved symptom tracking, physiologic monitoring, direct provision of care, and patient support.

Purpose:

The purpose of this review is to evaluate how digital applications are being used to manage functional gastrointestinal disorders today, with several examples of relevant technologies and organizations. It also the shortcomings of current treatment strategies and how they can be overcome.

Functional gastrointestinal disorders (FGIDs), or disorders of gut-brain interaction, are among the most common medical conditions in the world,¹ and greatly impact daily functioning and quality of life.

In recent years, evidence has emerged that counseling, customized diets, and other forms of high-touch care (ie, higher intensity care with more frequent interactions or customized planning) can significantly improve FGID treatment. While this level of support may stretch the personnel resources of a traditional gastroenterology (GI) practice, recent advances in digital health have created new opportunities to engage and manage these patients. Digital health is optimally suited to FGIDs as it enables more frequent provider-patient interactions and empowers patients to have greater agency over their care—a crucial advantage where FGIDs are part of a larger biopsychosocial construct in which innate coping mechanisms and external stressors outside the clinician’s office play an outsized role in symptoms.

GI digital health applications fall into four broad categories: (1) symptom tracking platforms; (2) digitally connected devices; (3) telemedicine; and (4) patient support groups. In this perspective, we review the current landscape of digital tools for FGIDs and future opportunities to improve care (Figure 1).

FIGURE 1.

Opportunities for digital health

1 |. SYMPTOM TR ACKING PL ATFORMS

One of the challenges of treating FGIDs is their innate symptom heterogeneity in the absence of a clear biomarker of disease activity. Thus, the clinician is reliant on patient histories that necessarily have gaps and biases. Physicians have historically asked patients to manually track symptoms and their relationship to foods, behaviors, and other stimuli, but rates of compliance are low, and information is non-comprehensive in paper trackers.² By contrast, web and mobile apps have demonstrated improved reporting compliance and prompt patients to fill in relevant data.³ Because many IBS patients associate trigger foods with symptom flares,⁴ trackers can identify culprit foods and associated symptoms. GI symptom trackers have proliferated over the last several years (Table 1).

TABLE 1.

Examples of digital health solutions for FGIDs

Application	Description	Website
Symptom trackers
Bowelle	App to track symptoms, food and water intake, mood, stress, sleep, activity, and bowel movements	https://bowelle.com/
mySymptoms	App to track symptoms, food and water intake, medication use, stress, sleep, activity, bowel movements, and other outcomes	https://www.mysymptoms.net/
Auggi	Developed and trained stool recognition algorithm to automatically classify patient-captured stool images according to the Bristol Stool Form Scale, which can then be correlated to a logged diet	https://www.auggi.ai/
Dieta	App that integrates analytical capabilities to predict triggers and recommend behavioral changes based on tracked symptom and activity data	https://dietahealth.com/
Digitally connected devices
AbStats	Audio biosensor worn on the abdomen to record bowel sounds and evaluate bowel function	http://gi-logic.com/products/abstats/
G-Tech Patch	Wearable electrode patches to detect electrical signals from the GI tract	http://www.gtechmedical.com/
Telemedicine
Parallel	3-month digital cognitive behavioral therapy program for patients with IBS	https://www.mahanatx.com/treatments/parallel
Cara Care	Combination of free symptom tracker and paid therapeutic program for IBS, including education, hypnosis, and behavioral therapy	https://cara.care/
Oshi Health	Virtual GI clinic with access to gastroenterologists, nutrition and mental health counseling, health coaching, and other clinic services	https://oshihealth.com/
Patient support
IBS Patient	Education and support community for patients with IBS	https://www.ibspatient.org/
IBS Self Help and Support Group	Education and support community for patients with IBS	https://www.ibsgroup.org/

Although symptom trackers have intuitive benefits, there are few studies demonstrating improved outcomes as a function of tracking. Zia, et al. ran a pilot in 11 IBS patients using a mobile app,³ but this small study did not show a significant improvement in bowel distress. Further studies are underway but tracking alone is unlikely to result in improved outcomes unless it impacts behavior and treatments. To that end, several apps have developed functionality to specifically draw correlations between stimuli and symptoms (Table 1).

These represent nascent efforts to organize symptom data. However, food and bowel diaries can overwhelm clinicians with a high volume of marginally useful information. Thus, the unrealized opportunity for digital tracking is to identify and report relevant metrics. Foodprint takes a step in this direction, as it incorporates “pre-visit notes” in which the patients synthesize their tracking data and questions ahead of appointments in order to make the visit more effective.⁵ Future programs can build on this by linking symptoms more accurately to objective data such as stool form and validated symptom severity scores, synthesizing this information for efficient provider review.

2 |. DIGITALLY CONNECTED DEVICES

Digital devices in GI are intended to complement subjective symptom tracking data with objective physiologic findings. Consumer wearables have been used for this purpose, linking symptom tracking apps to a health metric-enabled smartphone and correlating information such as step counts, sleep, or vital signs to symptom triggers. Using data from Fitbit (Google) in patients with constipation, one study found fewer steps, longer periods of inactivity, and reduced total sleep time were associated with higher rates of constipation and irregular bowel movement days.⁶ Similarly, a study in Japan used a LifeCorder (Suzuken Kenz) pedometer and found that higher step counts were associated with improved IBS symptoms.⁷ Additional details on wearable-connected technologies are found in Table 1.

Although promising, most technologies remain under investigation for FGID-specific use cases. A key question is how these abstract measurements translate into addressable physiology. Prior research using wireless motility capsule technology suggests patients with slow transit constipation have impaired colonic wake responses after sleeping.⁸ Correlating symptoms, wearable measurements, and physiologic studies can create annotated physiologic models, providing a more precise assessment of how a patient’s eating behaviors, activities, and sleep/wake cycle affect GI function.

3 |. TELEMEDICINE

Treatment of FGIDs requires extensive contact between patients and caregivers. IBS alone accounts for 2.4–3.5 million physician office visits per year.⁹ FGID patients can benefit from multidisciplinary care including dietary counseling and psychological counseling. This level of interaction is both inconvenient for the patient and resource-intensive for healthcare providers.

Remote care delivery can fill this gap. While symptom tracking provides patients with an ability to recognize and self-correct dietary triggers and other behaviors associated with symptoms, a logical next step is to offer the ability to access medical services through telemedicine. This takes many forms including virtual visits, patient education, care coordination, and bowel preparation support. Virtual treatments for FGIDs cover many services that have limited in-person availability including cognitive behavioral therapy (CBT), gut hypnotherapy, pelvic floor physical therapy, and nutritional services. These approaches can improve access to care while simultaneously reducing the load on providers. A home-based model was found to be at least as effective as standard CBT for IBS,¹⁰ while Skype-based hypnotherapy had a robust 65% responder rate although it was slightly less effective than face-to-face therapy.¹¹

A range of companies has started to commercialize telemedicine solutions (Table 1). These care models have generated encouraging, preliminary data. A study on Mahana’s digital CBT program found that over 60% of patients had a clinically meaningful improvement in IBS symptoms, outperforming usual care.¹² Cara Care ran a study on patients in Germany showing that overall symptoms improved 45.4% in the first 100 days in the program, and 78.8% of patients showed clinically significant improvements in symptoms.¹³ Oshi ran a 10-week study on 31 patients with IBS, IBD, and GERD, and found that 91% reported improved symptom control.¹⁴

4 |. PATIENT SUPPORT

Functional gastrointestinal disorders frequently cause isolation or embarrassment, leading to underreporting of symptoms or avoiding care.¹⁵ These feelings can be exacerbated by stigmatization from friends, family, or caregivers that FGIDs are not “real” medical conditions, with the patient to blame. As symptoms can flare with increased stressors, such inadequate social support can create a cycle of worsening symptoms.

Peer-to-peer support networks and group education programs have been shown to improve FGID symptoms.¹⁶ Patient feedback on such interventions highlights the importance of a safe community to discuss difficult topics. Most studies on patient support to date have focused on in-person interventions, but these sessions take time and energy to organize/administer. They also place the burden on the patient for enrollment and physical attendance for already-stigmatized conditions. Digital health offers a way to provide patient support across time and space through social media and other networks. The impact of these networks on symptoms has not been well studied, and importantly, they also carry the risk of spreading medically inaccurate information. This highlights the need for effective content curation to help digital support networks realize their potential.

5 |. FUTURE DIRECTIONS

Digital health offers tremendous promise in the treatment of FGIDs. Even so, it has seen minimal adoption in GI practice. Prior to the COVID-19 pandemic, surveys found that GI had among the lowest telemedicine rate of adoption of any specialty,^17,18 and the tools identified above continue to have limited uptake in practice.

We believe three major steps are required for wider adoption of digital health for FGIDs. First, rigorous clinical studies are needed to evaluate the impact of these interventions on health outcomes. For now, the quality of evidence to support the widespread use of digital interventions is lower for FGIDs than other chronic diseases such as heart disease and diabetes.^19,20 High-quality trials would build confidence in digital tools among prescribers, health systems, and patients.

This raises a question as to what constitutes a high-quality trial. We have identified four common shortcomings in digital health studies on FGIDs. First, they frequently take the form of small, single-center, uncontrolled pilots. Trials going forward require greater scale and control groups to draw meaningful conclusions. Second, they often focus on digitally native patients who are excited about using technology. Although this captures some FGID patients, it misses many more. Studies must enroll representative populations. Third, trials tend to have short durations of follow-up, in many cases, less than one year. The promise of digital tool for FGIDs rests in their ability to improve chronic symptoms, making longer term findings crucial. And fourth, studies sometimes use a shotgun approach to data collection and analysis. Digital technology can collect immense amounts of input and output data. This creates a temptation to retrospectively parse it looking for correlations, even if they may be incidental. Studying targeted questions with objective findings is particularly important for FGIDs, where symptoms are often poorly defined and better physiologic models are needed. Alongside improved clinical data, business models need to come into sharper focus. At present, companies commercializing digital interventions use a wide range of strategies including ad-supported free products, selling services directly to the patient, charging the provider with the expectation that the provider will seek reimbursement, directly seeking reimbursement from payers, physician-industry partnerships, and contracting with employers. The most important initial question that remains unanswered: Who is the customer? With this question left unanswered, many business models suffer from a lack of sustainability. The absence of reliable and replicable models creates barriers to access, particularly for those with limited means. It also means that many businesses in the space are not sustainable. Clinicians, understandably, are hesitant to use the services of a company that may not exist in a year. There has been some progress on payment models, including the release of remote patient monitoring CPT codes in 2019 and 2020, but more clarity is needed on how to pay for digital services. Effective business models will vary by country and health system; for example, what proves effective in the United States may not apply to other models such as the United Kingdom’s National Health Service.

Finally, service providers are needed that combine multiple applications under one roof. Although symptom trackers, digital devices, and other point offerings can all be useful, they are most effective when offered together. Gastroenterologists would ideally be able to place FGID patients in a unified program that can customize use of tracking platforms, wearables, virtual care, and patient support based on the patient’s specific needs. Importantly, such programs need to fit into provider workflows both with regards to reporting clinically relevant metrics, and integrating with electronic medical records and other technology. Digital tools cannot function in isolation; they need to complement existing care pathways.

These platforms are starting to come online; the American College of Gastroenterology recently partnered with GastroGirl^™ to offer GI OnDEMAND, an integrated virtual care and support platform.²¹ Availability of broad scale offerings ensures that both patients and physicians are well supported in digital care.

As digital care models mature, it is also important to consider their limitations. One central concern is that they can contribute to widening healthcare disparities, as patients in lower socioeconomic strata may have less access to digital care. Minimizing this inequity should be considered in the design of tools (eg, building them on widely available technology like smartphones rather than using customized platforms), their evaluation (eg, ensuring trials enroll and demonstrate success across socioeconomic strata), and their implementation (eg, focusing on business models that provide equitable access). In addition, physicians and FGID patients must remember that digital health is not a panacea; in-person counseling, intervention, and psychosocial support will remain vital components of care.

There are few conditions for which digital health is better suited than FGIDs. Careful monitoring, frequent caregiver interactions, and robust peer support are all hallmarks of effective treatment but are difficult to implement in traditional care models. Digital applications have the ability to fill that gap and represent the next frontier in FGID care.

DATA AVAILABILITY STATEMENT

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.