The Effects of a Digital Digestive Care Management Program on Employee Absenteeism

We evaluated a digital gastrointestinal (GI) chronic care program provided as a health benefit. We found that employees over the age of 50 years who used a digital GI care management program had 17% fewer sick days than nonparticipants. Decreasing sick days was associated with significant employer savings.

Abstract

Objective

60 to 70 million people in the US have chronic gastrointestinal (GI) disorders. Employers bear a significant economic burden for GI conditions. The purpose of this study was to evaluate the effects of a digital digestive care program on absenteeism.

Methods

Using propensity score matching between participants and nonparticipants, we evaluated sick days of public school system employees who were offered a digital digestive care program as part of their health benefits.

Results

Baseline annual sick days were comparable for the 237 participants and 400 nonparticipants. Among employees aged 50 years and older, during the intervention period, participants had significantly fewer sick days than nonparticipants (1.7 days [SD, 4.2], P = 0.02), an approximate direct salary savings of $672/employee annually.

Conclusions

Digital digestive care is promising for decreasing sick days among older workers.

LEARNING OUTCOMES

• Chronic gastrointestinal (GI) conditions are common among working age individuals and are associated with considerable cost and absenteeism (about 9 missed workdays annually).

• In an analysis of HR payroll data, we found that employees over the age of 50 years who used a digital GI care management program had 17% fewer sick days than nonparticipants.

• Decreasing sick days is associated with significant employer savings ($672 salary savings per employee annually, not including nonsalary savings).

Chronic gastrointestinal (GI) conditions affect 60 to 70 million people in the US and account for $142 billion in total costs.¹ Depending on the source, between 40% and 70% of US adults report at least one digestive symptoms monthly, with abdominal pain, gas or bloating, heartburn, and diarrhea being the most common.² More than half of those with GI symptoms report that they are moderate or severe in intensity.³ Unfortunately, fewer than one in four patients with chronic GI symptoms receive care from a physician to evaluate and manage their symptoms.³ Telemedicine offers a potential for increased access to GI specialty care.

Employers bear a significant economic burden for GI conditions given that chronic GI symptoms peak in midlife, affecting working adults. In addition to employer spending on GI-related care, employers also cover costs associated with lost productivity, as individuals with GI symptoms are more likely to miss work, be less productive, and leave their employer than employees without GI symptoms.^4–9 Employer spending on their employees with GI disorders varies from about $14,000 annually for individuals with chronic functional GI disorders to over $62,000 for those with Crohn's disease.⁵

Whether a digital digestive care program aimed at decreasing GI symptoms can reduce workplace absenteeism has not been evaluated. Additionally, most studies of the health effects of chronic GI conditions on workplace productivity have relied on patient self-report rather than actual hours worked as reported through employer payroll systems.^5–7 The purpose of this study was to evaluate the effects of a digital digestive care program on employer verified sick days.

METHODS

Participants

We evaluated sick days of public school system employees with GI-related claims (Appendix Table 1, http://links.lww.com/JOM/B966). Adult employees (aged 18 to 64 years) were offered a digital digestive care program as part of their health benefits for 1 year (February 1, 2023, to January 31, 2024). Participants were included if they registered with the program within the first 3 months after its launch.

Control Group

We propensity score matched participants to nonparticipants within the same public school system based on age, gender, race, HHS-HCC risk score, CDC Social Vulnerability Index (SVI),¹⁰ wellness score, work days, baseline sick days, and occurrence of COVID or malignant neoplasms. For the claims analysis, all participants and nonparticipants had 24 months of data available in the baseline and intervention periods.

Data Sources

We collected demographic data from employer eligibility files and sick day data from the employer's payroll system. We computed HHS-HCC scores and determined clinical conditions from claims and SVI from CDC census data. The employer provided a wellness score, an overall marker of health that measures patients' health status, complexity, and adherence (on a scale from 200 to 800) as described previously.⁹ Participants used the app to track food, stool, and nine common GI symptoms (abdominal pain, bloating, diarrhea, constipation, reflux, gas, nausea, vomiting, and loss of bowel control) on a five-point scale: 0 (no symptoms), 1 (mild symptoms), 2 (moderate symptoms), 3 (severe symptoms), and 4 (very severe symptoms). We compared an overall digestive symptom score (0 to 36, computed as the sum of each of the scores) at baseline and at the last recorded symptom tracking. We collected engagement data from participants' interactions with their care team via the app.

Intervention

The digital digestive health program has four key components: symptom tracking, personalized medical nutrition therapy, health coaching, and targeted education. It has been described in detail elsewhere,¹¹ but briefly, the app enables users to easily monitor their symptoms over time. Registered dieticians provide 40-minute, evidence-based, one-on-one medical nutrition therapy. They provide education on nutrition monitoring and management of common GI conditions. All participants have access to one-on-one health coaching to help with goal setting, development of self-efficacy and self-advocacy, and use of GI disease management resources. The program provides participants with personalized care plans that include targeted education on their symptoms, conditions, and relevant lifestyle interventions. Users of this program have been demonstrated to have significant reductions in symptoms and symptom severity.¹¹

Analysis

We utilized a case-control study design deploying 2:1 propensity score matching and difference-in-difference methodology to quantify the change in the number of sick days from the baseline period (12-month period preprogram launch) to the intervention period (12-month period postprogram launch) for participants compared with nonparticipants. We used a test to compare the change in symptoms and engagement with the program for participants younger than 50 years and those 50 years and older.

This study was performed under Vanderbilt IRB number 241271. The analysis and presentation of data conform to the STROBE guidelines (Appendix: STROBE checklist, http://links.lww.com/JOM/B967).

RESULTS

Demographics

The study included 237 participants and 400 well-matched nonparticipants (Table 1). Overall, their baseline mean age was 44 years (SD, 10 years), 89% identified as female, 76% as White/Caucasian, 10% had diabetes, and their mean annual healthcare spending was $9987/person. Overall, the most prevalent baseline GI diagnostic categories were abdominal pain (15.7%), esophageal disorders (5.0%), other GI disorders (3.9%), neoplasms (3.6%), and benign neoplasms (2.8%) (Appendix Table 1, http://links.lww.com/JOM/B966).

TABLE 1.

Baseline Characteristics of Participants and Matched Controls

Characteristic	Participants	Matched Controls	Absolute Standardized Difference*
N	237	400	n/a
Age, mean (SD), y	43.3 (9.6)	43.7 (7.5)	0.04
% Female	90%	90%	0 (Exact matching)
% White/Caucasian	77%	77%	0 (Exact matching)
Baseline health spend, mean (SD), $	$10,052 ($15,760)	$9367 ($13,338)	0.05
HCC risk score, mean (SD)	0.81 (2.1)	0.76 (1.5)	0.03
SVI, mean (SD)	0.4 (0.3)	0.4 (0.2)	0.02

*An absolute standardized difference between participants and matched controls of less than 0.10 is a strong indication of balance and suggests no selection bias.

Engagement

Older participants were more engaged (albeit not statistically significantly) with the program than younger participants (17.8 vs 16.6 total program interactions per participants, P = 0.78) including scheduling more care team appointments, more chats with the care team, and more stool and symptom tracking.

Symptom Improvement

Among younger participants, the average baseline total symptom score was 22.3, average change in symptoms at the end of the intervention was −9.3, representing a 42% improvement. Among older participants, the average baseline total symptom score was 17.3, average change in symptoms at the end of the intervention was −6.5, representing a 38% improvement. There was no significant difference in symptom improvement between younger and older participants (P = 0.7).

Absenteeism

Baseline annual sick days were comparable for participants and nonparticipants (9.1 days [SD, 5.1] vs 9.2 days [SD, 4.4], P = 0.76). During the intervention period, sick days among participants were 8.8 days [SD, 5.2] and among nonparticipants were 9.1 days [SD, 4.1]. Overall, participants had 0.2 fewer sick days than nonparticipants (not a statistically significant difference, P = 0.74).

Among employees aged 50 years and older, baseline sick days were comparable for participants and nonparticipants (9.1 days [SD, 5.0] vs 9.1 days [SD, 4.2], P = 0.98). However, during the intervention period, these older participants had significantly fewer sick days (8.5 days [SD, 5.1]) compared with nonparticipants (10.2 days [SD, 4.4]) (Fig. 1). On average, they had 1.7 fewer sick days [SD, 4.2], a 17% annual decrease in sick days compared to nonparticipants (P = 0.02). Compared with younger employees, the over age 50 years population was significantly more likely to be male (although still only represented 17% of the population), White/Caucasian, sicker, have higher baseline medical spending, and lower social vulnerability (Table 2).

FIGURE 1.

Sick day utilization among participants and nonparticipants aged 50 years and older. This figure shows matched observation-weighted differences in sick-day utilization between digital digestive health program participants and nonparticipants between the baseline period (light gray) and intervention period (black).

TABLE 2.

Baseline Demographics: Over 50 Population Versus Under 50 Population

Characteristic	Older Adults (≥50 y)		Younger Adults (<50 y)		P Value Difference Between Older and Younger Participants
	Participants	Nonparticipants	Participants	Nonparticipants
N	64	121	173	279	n/a
Age, mean (SD), y	56 (3.8)	56 (2.9)	39 (6.2)	39 (5.3)	<0.0001
% Female	83%	83%	92%	92%	0.04
% White/Caucasian	83%	83%	75%	75%	0.18
Baseline health spend, mean (SD), $	$13,527 ($19,952)	$12,133 ($18,370)	$8766 ($13,741)	$8343 ($10,031)	0.0387
HCC risk score, mean (SD)	1.7 (3.6)	1.4 (2.5)	0.48 (0.8)	0.51 (0.6)	<0.0001
SVI, mean (SD)	0.30 (0.3)	0.31 (0.2)	0.46 (0.3)	0.45 (0.2)	0.0004

Based on the employer's average salary of $49.50/hour for employees aged 50 years or older, a 1.7 sick day reduction saves approximately $672/employee salary costs annually—this does not include other direct medical costs for GI-related care or other indirect costs such as the cost of workplace disruptions due to absenteeism or presenteeism.

DISCUSSION

Despite the common prevalence of digestive disorders among working-age adults, poor access to GI specialty care persists. Although the direct and indirect costs to employers of digestive disorders have been described,^4–9 the effects of clinical interventions on these outcomes have not been well characterized. This, the first study to evaluate a digital digestive health program's effect on absenteeism by assessing employer-reported sick days, had three key findings. First, that public school system employees with GI conditions had an average of 9 sick days as reported via their employer's payroll system. This is higher than what has been reported previously via employee self-report for patients with inflammatory bowel disease.^4,5 Prior authors have noted that greater self-reported absenteeism is associated with higher disease severity for individuals with dyspepsia.⁷ Because of the nature of their work, GI symptoms can make it particularly troublesome for teachers to remain in the classroom, which may have contributed to this higher rate of absenteeism (many teachers report chronically fluid restricting due to challenges leaving their classrooms unattended while taking a restroom break). Because our analysis included participants with a spectrum of GI conditions (who could also have had other clinical conditions), further analysis of absenteeism as reported through employer-verified data stratified by GI condition and severity is warranted.

Second, although the use of a digital digestive care program was not associated with a significant change in sick days for younger participants, it was associated with a 17% reduction in sick days for participants aged 50 years and older compared to controls. Both younger and older participants reported symptom improvement, and older participants had higher engagement with the program but not statistically significantly so. Given that older employees were sicker overall and had higher medical spending, they may have had higher disease acuity and benefited more from the disease management intervention. Some authors have described that younger workers are more likely to take sick days even at lower symptom acuity.¹² Thus, this interesting age-related finding warrants further evaluation to explore the association of symptom severity and work place absenteeism among younger and older workers.

Third, based on a conservative calculation that only included hourly wages, the reduction in sick days among older participants resulted in a statistically significant annual savings of $672 in direct salary. This promising finding warrants a more comprehensive economic analysis that includes employer spending on participants' GI care compared to controls. Notably, these estimates only include the direct payroll impacts of workplace absences. They do not include other important features, such as increased medical spending or personal costs of workplace absences, such as childcare or family disruptions. For these reasons, these estimates may represent a lower-bound.

This study had three key limitations: it evaluated predominantly female employees (notably, several common digestive disorders such as irritable bowel syndrome is more commonly reported among women), was based on a single employer, and did not limit the assessment to sick days for GI-specific conditions. Further research is warranted in other populations, with larger samples of older employees, to assess the generalizability of our findings.

We conclude that employees with GI conditions have significant employer-verified absenteeism. Moreover, a digital digestive care management program designed to reduce GI symptoms and symptom severity is associated with reductions in these productivity losses among older employees. These promising findings suggest that employer-provided productivity data should be more routinely included in evaluations of employer-sponsored health benefits.