Feasibility study of a behavioral lifestyle intervention for Hispanic/Latino patients with metabolic dysfunction-associated steatotic liver disease

Abstract

Introduction

Metabolic dysfunction-associated steatotic liver disease (MASLD) can lead to longer-term adverse health outcomes if left unmanaged, including liver failure and liver cancer. Hispanic adults have the highest rates of MASLD of any U.S. racial/ethnic subgroup. The most effective treatment is weight loss, through healthy dietary changes and increased physical activity. This study conducted 2022–2023 aimed to assess the feasibility and acceptability (primary), as well as exploratory effectiveness outcomes, of a behavioral lifestyle intervention for Hispanic patients with MASLD.

Methods

Participants were recruited from a federally qualified health center. Self-administered questionnaires assessed demographic, psychosocial and behavioral variables. Staff measured height and weight. Individuals were randomized into intervention or wait-list control groups. The intervention, Healthy Liver/Hígado Sano, is adapted from the Diabetes Prevention Program and involves 16 weekly, group-delivered sessions. We assessed feasibility and acceptability of delivering the program in partnership with a clinic, as well as summary statistics at baseline and follow-up. We calculated between-group differences in change between the two groups for exploratory effectiveness outcomes using two-sample t-tests and chi-square tests, as appropriate.

Results

Of the 60 individuals screened, 45 were eligible (75%). Of these, 33 consented (73%) and 27 completed baseline data collection (60%) and were randomized, 13 to the wait-list control (usual care) and 14 to intervention condition. At follow-up, 19 of the 27 randomized (70%) were retained. Average age was 48.5 and 46.6 years in control and intervention groups, respectively. Most were female (84.6% control, 85.7% intervention) and had low acculturation (92–93%). Median satisfaction was 26 (of potential range 8–32), suggesting moderate acceptability among participants who completed follow-up. For exploratory effectiveness outcomes, while not statistically significant, intervention participants showed greater improvements in MASLD illness perception, treatment efficacy, social support, healthy eating index, and walking MET-minutes as compared to the control group.

Conclusions

Recruitment was feasible and participants found the intervention acceptable, though improvements are needed for retention. Findings will inform future trial implementation and refinement.

Trial registration

This study was registered at clinicaltrials.gov (NCT05200585) on January 6, 2022.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-026-27906-z.

Introduction

Hispanic/Latino (hereafter “Hispanic”) adults in the U.S. experience a variety of liver-related disparities, including increased incidence of hepatocellular carcinoma (HCC) [1, 2], HCC diagnosis and treatment [3], liver transplantation [4], and chronic liver disease/cirrhosis mortality [5]. Metabolic dysfunction-associated steatotic liver disease (MASLD) [6], formerly Non-alcoholic Fatty Liver Disease (NAFLD), is an increasingly concerning and prevalent form of chronic liver disease, especially in Hispanic adults [7]. A diagnosis of MASLD involves excessive fat accumulation in the liver (hepatic steatosis), along with the presence of at least one cardiometabolic risk factor [6, 8]. Metabolic dysfunction-associated Steatohepatitis (MASH), MASLD’s progressive subtype, is now the leading indication for liver transplantation in patients with HCC in the U.S [9]. The prevalence of MASLD is high in the U.S. population and is even higher in the Hispanic adult population [10].

Hispanic individuals have a higher MASLD prevalence for various reasons, including the higher prevalence of genetic polymorphisms associated with susceptibility to heptaic steatosis and fibrosis [11], higher prevalence of co-morbid conditions such as metabolic syndrome and type 2 diabetes [12], a higher likelihood of living in an environment that hinders healthy lifestyle behaviors [13], and thus a lower prevalence of meeting lifestyle behavior guidelines, such as physical activity [14]. Because of MASLD’s role as a determinant for advanced liver disease, such as cirrhosis, and because the prevalence of MASLD has nearly doubled in the last three decades [15], an impending increase in HCC incidence, cirrhosis, and liver transplants can be expected among Hispanic adults. Fortunately, the progression to advanced liver disease can be prevented through lifestyle interventions.

Excess body fat and metabolic dysregulation are central mechanisms underlying MASLD pathogenesis [16]. Clinical practice guidelines for MASLD highlight the central role of lifestyle behavior change (including reduced dietary calorie and sugar intake and increased physical activity) leading to weight loss [8]. Even with the new availability of antifibrotics and weight loss medications [17], lifestyle behavior change will continue to be essential to long-term MASLD management. Positive outcomes, including reductions in hepatic steatosis can be seen with any amount of weight loss [18], though greater weight loss is likely needed to reverse fibrosis [19]. Even in the absence of weight loss, dietary modifications and increased physical activity are each associated with reductions in liver fat and improvements in insulin resistance [20–23]. Thus, the focus for patients with MASLD should be on achievable behaviors that can be sustained over time, with the goal of eventual weight loss and improvements in liver fat and/or stiffness.

While existing lifestyle interventions for adult patients with MASLD have shown promise [19, 20, 24–26], it is unclear if those results generalize to Hispanic patients with MASLD. There is currently a lack of tailored interventions for Hispanic adults with MASLD [27]. Our interviews with Hispanic patients (all of Mexican/Central American descent) with MASLD revealed that they were concerned about their diagnosis, wanted additional education on MASLD, and wanted to have conversations about MASLD management with other patients like them [28]. These findings were similar to those found in other studies that suggest MASLD diagnosis motivates change [29–31]. To address patient needs for additional MASLD-specific education and support, we developed Hígado Sano/Healthy Liver [32], a behavioral lifestyle intervention for Hispanic patients with MASLD to help them achieve the weight loss needed for improved liver health. This intervention was intended to fill an existing gap in behavioral interventions tailored to disease-specific needs and cultural preferences.

This study aimed to test the feasibility and acceptability (primary outcomes) and exploratory effectiveness outcomes of a behavioral lifestyle intervention for Hispanic/Latino patients with MASLD using a randomized controlled design.

Methods

Study design

We conducted a feasibility study using a randomized controlled design with a waitlist control group from 2022 to 2023. We partnered with a local federally qualified health center serving a predominantly Hispanic population in Houston. We identified a physician champion who spearheaded the administrative effort and served as a liaison to help navigate internal approvals and processes. After recruitment and baseline data collection, participants were randomized to study conditions in a 1:1 ratio using the REDCap randomization module [33]. Intervention participants attended 16 weekly classes, while wait-list control participants continued to receive usual care (visits with a physician every 3–6 months). Study staff collected follow-up data at the conclusion of the intervention period. After follow-up data collection, control participants were invited to receive the intervention. This study was reviewed and approved by the UTHealth Houston IRB (HSC-SPH-21-1008) and was registered at clinicaltrials.gov (NCT05200585). Reporting adheres to the CONSORT guidelines.

Intervention

The Hígado Sano/Healthy Liver intervention is based on the Diabetes Prevention Program (DPP), the gold standard in behavioral lifestyle interventions [34]. The DPP has been widely disseminated and is one of the most successful and scalable lifestyle interventions [35–37]. DPP adaptations have successfully achieved ~ 5% weight loss in participants [38–40], including adaptations to who delivers the program, to whom it is delivered, the format of the sessions, and the delivery setting. Moreover, DPP was recently shown to improve hepatic steatosis [41]. We used IM Adapt (Intervention Mapping for Adaptation) to systematically adapt the CDC’s Prevent T2, a version of the DPP (full details available elsewhere) [32]. The resulting program, Hígado Sano/Healthy Liver, adds background information on MASLD to increase participants’ understanding of their condition and provides opportunities to discuss their concerns with peers. Hígado Sano/Healthy Liver consists of 16 in-peron weekly sessions (about 60 min), delivered by a Lifestyle Coach (Master’s student with a dietetic background who received the official CDC Lifestyle Coach training) in a group setting (~ 10 participants) with weigh-ins, review of self-monitoring records, presentation of new topics each week, and development of an action plan to address personal barriers. Determinants targeted by the intervention include knowledge, attitudes, subjective norms, perceived severity, perceived susceptibility, skills, self-efficacy, outcome expectations, and overcoming barriers [32]. Content each week varied, with sessions covering background information on the disease, basic components of a healthy meal (based on MyPlate), importance of physical activity and how to slowly increase physical activity, setting goals and finding solutions to overcome barriers, and tracking physical activity, diet, and weight, among other similar topics. The targeted outcomes of the program include eating healthier (e.g., lower-calorie foods, less sugar-sweetened beverages and foods with added sugars, etc.), working up to 150 min of moderate-intensity physical activity per week, avoiding alcohol, and achieving clinically significant weight loss (≥ 5%), with weight loss being the primary goal. Both Spanish and English versions were modified to include culturally appropriate food, types of physical activity, familiar names, and relevant pictures. Participants received text and/or call reminders for sessions, and sessions were scheduled to accommodate participants’ schedules. In certain situations, the Lifestyle Coach made a synchronous online option available to accommodate transportation or childcare barriers.

Recruitment

We employed multiple recruitment methods. We distributed flyers in Spanish and English at the clinic, advertising the study and providing the study contact information. Additionally, physicians at the clinic identified potentially eligible participants as they presented for their patient visits. The physicians mentioned the study and provided the study contact information to the patient. Alternatively, with the patient’s consent, the physician securely transmitted the participant’s name and contact information to the study staff. Once the participant connected with a study staff member, the staff member explained the study to the participant. Those interested in participating were screened for eligibility. Potential participants provided oral consent to be screened. Eligibility criteria included the following: adults 18 years of age or older, self-identifying as Hispanic/Latino, self-reporting a diagnosis of MASLD, speaking Spanish or English, having a working telephone number, and having access to the internet, either on their phone, at home, or at some other location convenient to them. While on the phone with an eligible participant, study personnel provided a detailed description of the study, answered questions, and, for those who remained interested, obtained written informed consent via REDCap [33]. Figure 1 shows a CONSORT diagram for the study.

Fig. 1.

CONSORT diagram

Data collection

Participants responded to self-administered baseline and follow-up questionnaires using REDCap [33]. While the questionnaire could be completed online before the in-person visit, most individuals completed the questionnaire during the baseline anthropometric visit on provided devices (computers or large tablets). This visit was held at the partner clinic site, in an available spare room during open clinic hours. Participants received a $25 gift card for completing the assessments at each time point. Moreover, they received $25 at both baseline and follow-up for wearing and returning an accelerometer (data not reported here).

Measures

The main outcomes were related to the program’s feasibility and acceptability, as determined by recruitment and retention, and satisfaction. We also collected behavioral and psychosocial variables for exploratory effectiveness outcomes. Sociodemographic characteristics collected included sex, age, education, annual family income, and other comorbidities. We assessed language acculturation using the Brief Acculturation Scale for Hispanics [42].

Feasibility and acceptability

Using REDCap, we kept detailed documentation on recruitment, retention and adherence. Metrics of success were set a priori and were ≥ 70% of eligible participants enrolled (recruitment feasibility), ≥ 80% completion of follow-up assessments (feasible to retain participants), and ≥ 70% attendance to 60% (10 of 16 classes) [43] of the intervention (adherence). We collected information on the number of participants interested, eligible, ineligible (and reasons for ineligibility), and enrolled during the recruitment process. We kept detailed records of the completion of baseline and follow-up assessments. We used the Client Satisfaction Questionnaire (CSQ) [44] to assess program satisfaction in the intervention group at follow-up only. The CSQ has 8 items and has been validated with Hispanic samples and in Spanish [45, 46]. CSQ response options are from 1 to 4, with a resulting scale range of 8 to 32, with higher numbers indicating greater satisfaction. We set a prior benchmark of an average score of 26 [47].

Illness perceptions

The brief 8-item Illness Perception Questionnaire, adapted with language specifically referring to MASLD, was used to measure participants’ assessment of the consequences, timeline, personal control, treatment control, identity, concern, understanding and emotional response to their condition [48]. We summed the items to create a scale (Cronbach’s alpha = 0.77), with lower numbers indicating adaptive perceptions associated with better coping, self-management, and outcomes.

Perceived treatment efficacy

Perceived treatment efficacy was measured with 8 items, adapted from Glasgow et al. [49] that assess the importance of four protective factors (weight loss, eating healthy, avoiding sweets, and physical activity) and the likelihood of each of those factors to prevent future complications from MASLD. Response options are from 1 (not likely) to 5 (extremely likely). We summed the items to create a scale (Cronbach’s alpha = 0.81), with higher numbers indicating greater perceptions of treatment efficacy.

Social support

We collected social support for physical activity and diet using Sallis and colleagues’ Social Support for Exercise and Diet scales, focusing on the friends-related items, given the context of the group-delivered intervention [50]. Social support for physical activity included 10 items (Cronbach’s alpha = 0.91) and social support for diet from friends included 5 items (Cronbach’s alpha = 0.93), with responses 0 (none) to 4 (very often). Per scoring recommendations [50], we summed the items to create a scale, with higher numbers indicating greater perceived social support.

Alcohol use

We assessed alcohol use using the brief 3-item alcohol use disorders identification test (AUDIT-C) [51], given that evidence suggests that even light-to-moderate drinking in those with MASLD can cause worsening of their liver condition [8, 52]. Items asked about the frequency of drinking, the number of drinks on a typical day, and the occasions on which 6 or more drinks were consumed. We calculated the AUDIT-C score based on standard scoring [51].

Anthropometrics

During the visit, trained study staff measured height with a stadiometer and weight with a Tanita scale. Two measurements were taken and averaged. We calculated the percentage weight change.

Diet

We used a brief 22-item food frequency questionnaire tailored to a 4th -grade reading level and modified to include common traditional Mexican foods, which we previously validated [53]. Items include fatty, fried and lean meats and fish, eggs, white and whole wheat bread/grains, vegetables, beans, fruit, fruit juice, sugar-sweetened drinks and sodas, frozen desserts, sweets and candy. Individuals reported how many times they consumed the foods/drinks yesterday, ranging from 0 (none) to 5 (5 or more). From the 10 items focused on healthy foods (e.g., whole wheat, vegetables), we calculated the Healthy Food Index.

Physical activity and sitting time

We collected self-reported physical activity using the International Physical Activity Questionnaire [54]. Given the curriculum focused on walking more and also discusses reducing sitting, we calculated Total MET minutes from walking as well as weekend sitting time in minutes (weekend vs. weekday expected to be more amenable to change when focused on leisure-time activities). Participants also received and were asked to wear a hip-worn ActiGraph GT3X accelerometer for 7 days at each assessment time point. Participants were given a prepaid envelope to return the device.

Statistical analyses

Baseline demographic characteristics were summarized using sample mean, standard deviation, median, minimum, and maximum for continuous variables, and frequency and percentage for categorical variables, stratified by study arms. For key outcome measures, including illness perceptions, perceived treatment efficacy, social support, alcohol use, anthropometrics, diet items, physical activity, and sitting time, we reported summary statistics at baseline and follow-up. We also reported change scores from baseline to follow-up for the sample with data at both time-points. Between-group differences in baseline-to-follow-up change scores for continuous variables in the intervention and control groups were assessed using two-sample t tests, and categorical variables were compared using chi-square tests. All analyses were performed using R version 4.0 [55].

Results

Demographic characteristics

Demographic and other baseline characteristics are presented in Table 1. The average age was 48.5 years in the control group and 46.6 years in the intervention group, with no statistical difference. Most of the sample was female (84.6% in the control, 85.7% in the intervention) and had low levels of acculturation (92–93% in both groups). Randomization resulted in similar demographic characteristics across the two groups. However, the intervention group had more individuals with high cholesterol compared to the control group.

Table 1.

Baseline characteristics of study sample (N = 27)

	Control	Intervention	P-value
	(N = 13)	(N = 14)
Age
Mean (SD)	48.5 (10.6)	46.6 (10.4)	0.66
Median [Min, Max]	51.0 [29.0, 63.0]	46.5 [20.0, 63.0]
Sex
Female	11 (84.6%)	12 (85.7%)	1
Male	2 (15.4%)	2 (14.3%)
Language acculturation
High	1 (7.7%)	1 (7.1%)	1
Low	12 (92.3%)	13 (92.9%)
Highest Level of Education
Middle school or less	5 (38.4%)	6 (42.9%)	0.31
Some high school/ high school graduate	4 (30.8%)	5 (35.7%)
Some college or higher	3 (23.1%)	2 (14.3%)
Refuse	1 (7.7%)	1 (7.1%)
Annual Household Income
Less than $14,999	6 (46.2%)	6 (42.9%)	0.70
$15,000-$24,999	4 (30.8%)	2 (14.3%)
$25,000-$54,999	1 (7.7%)	3 (21.4%)
Refuse/Don’t Know	2 (15.4%)	3 (21.4%)
Type 2 diabetes
No	7 (53.8%)	6 (42.9%)	0.85
Yes	6 (46.2%)	8 (57.1%)
Heart disease
No	12 (92.3%)	13 (92.9%)	1
Yes	1 (7.7%)	1 (7.1%)
Hypertension
No	11 (84.6%)	9 (64.3%)	0.44
Yes	2 (15.4%)	5 (35.7%)
High cholesterol
No	12 (92.3%)	6 (42.9%)	0.02
Yes	1 (7.7%)	8 (57.1%)

Primary outcomes: feasibility and acceptability

A total of 151 individuals were initially interested in the study (Fig. 1). We were able to speak with 126 of the initially interested individuals to provide an in-depth description of the study (we were unable to reach the rest), and of those, 60 expressed interest in being screened for eligibility. Of those 60, 45 were eligible (75%). Of the 45 eligible, 33 consented (73%). We lost 6 individuals in the time between consent and baseline data collection, resulting in 27 individuals completing baseline data collection (60% of those eligible) and subsequently being randomized: 13 to the wait-list control and 14 to the intervention group. At follow-up, 7 and 12 were retained, respectively (70% of those randomized). For intervention participants, 9/14 participants (64%) completed at least 10 of the 16 classes (62.5% of the intervention), mean CSQ at follow-up was 26.2 (SD = 3.6), and median CSQ was 26, indicating the intervention was perceived as acceptable.

Exploratory effectiveness outcomes: psychosocial, behavioral, and anthropometric variables

Mean and median change scores from baseline to follow-up for MASLD Illness perceptions decreased more in the intervention (-12.5 and -9.0, respectively) compared to the control group (-2.9 and -2.0, respectively; Table 2). Lower numbers on this measure are better, as they are associated with constructive coping, emotional resilience, treatment adherence, and long-term health outcomes. Mean and median change scores from baseline to follow-up of perceived treatment efficacy increased more in the intervention (1.4 and 4.0, respectively) compared to the control group (-0.6 and -1.0, respectively). Higher numbers are better, indicating an individual believes that diet, physical activity and/or weight loss can help them treat and prevent future complications from MASLD. Social support for diet increased in the intervention group (mean change 2.3, median change 1.0) compared with the control group, in which it remained stable from baseline to follow-up (mean change 0.1, median change 0). Social support for physical activity followed a similar pattern, though the increase in the intervention group was less pronounced.

Table 2.

Summary statistics by intervention groups and time-point

	Baseline		Followup		Baseline to Follow-up Change		p-value
	Control	Intervention	Control	Intervention	Control	Intervention
	(N = 13)	(N = 14)	(N = 7)	(N = 12)	(N = 7)	(N = 12)
Illness Perceptions
Mean (SD)	45.5 (15.7)	46.9 (17.3)	41.3 (13.4)	37.2 (9.4)	-2.9 (12.2)	-12.5 (19.3)	0.20
Median [Min, Max]	43.0 [17.0, 76.0]	48.5 [17.0, 69.0]	41.0 [17.0, 57.0]	36.0 [20.0, 51.0]	-2.0 [-20.0, 17.0]	-9.0 [-36.0, 16.0]
Perceived Treatment Efficacy
Mean (SD)	32.2 (5.4)	32.6 (4.2)	34.6 (3.4)	34.2 (6.1)	-0.6 (3.2)	1.4 (7.8)	0.45
Median [Min, Max]	34.0 [22.0, 40.0]	33.5 [24.0, 40.0]	34.0 [30.0, 40.0]	35.5 [24.0, 40.0]	-1.0 [-4.0, 5.0]	4.0 [-16.0, 9.0]
Social support for physical activity
Mean (SD)	9.7 (7.5)	10.7 (6.8)	6.9 (5.9)	10.0 (9.6)	-0.4 (8.7)	0.3 (6.9)	0.85
Median [Min, Max]	5.5 [4.0, 22.0]	9.0 [2.0, 24.0]	4.0 [4.0, 20.0]	4.5 [4.0, 28.0]	0 [-16.0, 14.0]	0.5 [-13.0, 10.0]
Social support for diet
Mean (SD)	5.7 (6.47)	3.3 (5.7)	5.6 (5.7)	4.8 (6.4)	0.1 (4.2)	2.3 (5.0)	0.32
Median [Min, Max]	2.5 [0, 19.0]	1.0 [0, 16.0]	2.0 [0, 15.0]	3.0 [0, 19.0]	0 [-4.0, 8.0]	1.0 [-4.0, 15.0]
Weight (kg)
Mean (SD)	78.3 (12.8)	84.0 (19.2)	78.7 (16.3)	82.6 (19.2)	0 (1.3)	0.2 (3.0)	0.9
Median [Min, Max]	78.2 [52.6, 105]	81.2 [62.5, 120]	78.1 [52.5, 108]	77.6 [64.3, 116]	-0.1 [-2.1, 2.2]	0.6 [-5.7, 5.0]
Percent weight change
Mean (SD)					-0.1 (1.5)	0.4 (3.7)	0.66
Median [Min, Max]					-0.1 [-2.8, 2.1]	0.9 [-6.8, 7.3]
Healthy Eating Index
Mean (SD)	7.8 (4.5)	5.8 (2.2)	7.1 (3.0)	6.3 (2.3)	-1.3 (5.7)	0.2 (1.9)	0.54
Median [Min, Max]	7.5 [2.0, 20.0]	5.5 [2.0, 9.0]	7.0 [3.0, 11.0]	6.0 [3.0, 12.0]	1.0 [-13.0, 4.0]	0 [-4.0, 3.0]
Total MET mins from walking
Mean (SD)	1980 (2790)	1560 (1820)	1960 (2680)	2230 (2600)	956 (2818)	650 (2604)	0.82
Median [Min, Max]	446 [0, 8320]	1190 [0, 6530]	218 [0, 6930]	1100 [0, 7030]	0 [-1485, 6930]	396 [-3564, 5841]
Weekend Sitting time (minutes)
Mean (SD)	350 (377)	259 (140)	201 (172)	180 (105)	-54 (109)	-72 (118)	0.75
Median [Min, Max]	235 [72, 1440]	240 [45, 480]	180 [5, 480]	180 [60, 360]	-60 [-180, 120]	-79 [-300, 180]
AUDIT score
Mean (SD)	1.2 (2.5)	1.0 (1.5)	0.6 (0.8)	1.0 (1.0)	0 (0)	0.3 (0.9)	0.34
Median [Min, Max]	0 [0, 9.0]	0 [0, 5.0]	0 [0, 2.0]	1.0 [0, 3.0]	0 [0, 0]	0 [-1.0, 2.0]

Although not statistically significant, the healthy eating index improved more in the intervention group (mean change: 0.2) than in the control group (-1.3) between baseline and follow-up. There was a non-significant increase in mean total MET minutes per week from walking in both intervention and control groups, though the intervention group had greater improvements in median total MET minutes from walking compared to the control (396 MET minutes per week vs. 0). There was also both a greater mean and median decrease in weekend sitting time in the intervention as compared to the control group. Almost all participants reported no alcohol use at baseline (e.g., AUDIT-C score of 0), and thus the median change score from baseline to follow-up was 0 in both groups. Accelerometer-measured physical activity follow-up data were missing for 48% of the sample and were not assessed further. We did not observe a weight decrease from baseline to follow-up in the intervention group compared with the control group.

Discussion

In the present study, we implemented a 16-session behavioral lifestyle intervention in Hispanic patients with MASLD. This was a feasibility study, thus, the intention was to assess the recruitment, retention, and participant satisfaction with the program [56, 57]. We determined that it was feasible to recruit (≥ 70% of eligible participants initially enrolled) participants into the study and that participants were generally satisfied with the program (average CSQ of 26 out of 32). We fell short of our goals of ≥ 80% completion of follow-up assessments and ≥ 70% of participants completing at least 10 of the 16 sessions, indicating only partial feasibility; our current protocol will need to be further refined to focus on retention. We also employed a randomized controlled design and thus compared intervention and control groups on key variables that will serve as future primary and secondary outcomes in a full-scale trial. While no differences in change scores between the intervention and control groups reached statistical significance, we observed trends in the expected direction for most psychosocial and behavioral variables we expected to change with the intervention.

This feasibility study was useful for determining the necessary changes to the implementation protocol. For example, a few individuals withdrew between consent and baseline data collection. This was due to a variety of reasons (no time, lost contact, no transportation, new job), but a main underlying factor was the need to wait until enough individuals were recruited to begin a group class. Given ongoing issues with the COVID-19 pandemic, this may have made it harder to recruit individuals interested in participating in an in-person behavioral lifestyle intervention. Moreover, we fell short of our goals of having at least 70% of intervention participants complete 10 of the 16 classes and having at least 80% of the overall enrolled sample complete follow-up assessments. Some of these retention issues may be due to the in-person nature of the classes, which require transportation and additional time. The decision to deliver classes in person was made based on the results of a needs assessment with the target population [32], which was also completed during the height of the COVID-19 pandemic. However, we did provide a synchronous online attendance option on a case-by-case basis if needed. Regardless of the cause, this was an important lesson related to feasibility; strategies to be implemented to improve retention in future trials include reducing the delay between enrollment and intervention initiation, offering additional hybrid or remote delivery options, and providing transportation or scheduling support.

While we employed two methods of data collection for physical activity (self-report and accelerometry), given the number of issues with accelerometers experienced in this study (lost by the participant, lost in the mail, not worn for the minimum duration, or broken), the use of the accelerometer was not feasible as currently implemented. For future evaluations, procedures will be altered, including additional in-person training on wearing the accelerometer, additional reminders to wear the device, additional reminders about the incentives linked to the wear and return of the device, and return of the device in person rather than via mail. This will ensure more complete data and allow the pairing and validation of self-reported data with accelerometry in this sample. We will also implement enhanced follow-up reminders to improve adherence to wearing accelerometers.

As mentioned, we had variability in intervention attendance, which likely impacted individuals’ success in the program, including their weight loss. Future iterations of the program will need to focus on increasing attendance in the intervention group to ensure that they receive a sufficient dose of the intervention. Moreover, the intervention’s impact on weight, one of the key outcomes in an eventual full-scale trial, will need to be carefully tracked. This intervention was delivered in 16 weekly sessions due to staffing constraints, with follow-up 4–5 months after program initiation. Moving to a weekly model for the first 8 sessions, followed by a bi-weekly model for the last 8 sessions (in line with the DPP), might give individuals more time to lose weight at a 6-month follow-up assessment.

Lastly, there are considerations for changes to intervention content. While it is critical to minimize alcohol use in patients with MASLD, the present sample had little to no alcohol use. This finding aligns with other research that indicates that low-acculturated or Spanish language-dominant Hispanic women, who comprised most of our sample, have some of the lowest levels of alcohol consumption [58, 59]. Thus, while interventions for Hispanic patients with MASLD should certainly address alcohol use, given its synergistic effect in those with the condition,⁵² if baseline alcohol use is low, it may not have to be a major topic of discussion. However, resources and referrals should be available as needed. Additionally, since the implementation and evaluation of this intervention, there is now newer guidance on diets that could improve outcomes for patients with MASLD [8]. Specifically, any future iterations of this intervention should modify the diet-related education to include education and skill-building related to the Mediterranean-style diet [60]. These changes could potentially enhance the intervention and its effects moving forward.

There are several limitations in the present study. Given the number of issues we experienced with the wear and return of accelerometers, we were unable to evaluate the accelerometry data. We have identified potential solutions to improve accelerometer data collection in future tests of this intervention, including additional reminders for wear and return, and incentives for accelerometer return. We also did not have the capacity, given funding, to collect liver biomarkers in the present study. This limits our ability to interpret the impact of this intervention on MASLD. However, this is a feasibility study, with only exploratory effectiveness outcomes, and a future trial will collect relevant liver biomarkers and be powered to detect differences in those biomarkers. Despite these limitations, this study details the implementation and evaluation of one of the first feasibility or pilot studies testing a behavioral lifestyle intervention specifically tailored for Hispanic patients with MASLD.

Conclusion

We found that a behavioral lifestyle intervention adapted for Hispanic patients with MASLD was acceptable to participants and feasible to recruit for, though improvements will be needed to ensure better attendance at intervention sessions and retention in the study, including reducing the delay between enrollment and intervention initiation and offering hybrid delivery options. We found non-statistically significant improvements in some exploratory effectiveness variables in the intervention as compared to the control group between baseline and follow-up. However, additional enhancements to the intervention will be needed to ensure that participants improve their target behaviors of physical activity and diet and eventually achieve their weight loss goals.

Abbreviations

HCC

Hepatocellular carcinoma

MASLD

Metabolic dysfunction–associated steatotic liver disease

NAFLD

Non–alcoholic fatty liver disease

MASH

Metabolic dysfunction–associated steatohepatitis

U.S.

United States

IRB

Institutional Review Board

CONSORT

Consolidated Standards of Reporting Trials

REDCap

Research Electronic Data Capture

IM Adapt

Intervention Mapping for Adaptation

DPP

Diabetes Prevention Program

CDC

Centers for Disease Control and Prevention

CSQ

Client Satisfaction Questionnaire

AUDIT

C–Alcohol Use Disorders Identification Test

MET

Metabolic Equivalent of Task

SD

Standard Deviation

NCT

National Clinical Trial

Authors' contributions

NH was responsible for the conception and design of the work, the acquisition and interpretation of the data, and wrote the main manuscript text. JP, MF, CB, and LM provided substantial revisions of the manuscript text and provided supervision and oversight on the design of the study and analyses of the data. JL analyzed the data and helped write the statistical analyses section, as well as prepare the tables. All author have reviewd and approved the manuscript.

Funding

This study was funded by the Prevent Cancer Foundation. NH’s time was partially funded by NIH/NIMHD (K01MD019149).

Data availability

Data is available upon reasonable request to the corresponding author.

Declarations

Ethics approval and consent to participate

Approval was obtained from the Institutional Review Board of The University of Texas Health Science Center at Houston (HSC-SPH-21-1008). The procedures used in this study adhere to the tenets of the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.