Patterns of dietary supplement use among United States patients with steatotic liver disease: Vitamins, minerals and botanicals

Abstract

BACKGROUND

Dietary supplement (DS) usage among United States adults has significantly increased. Patients with steatotic liver disease (SLD) may have unique motivations to take DS in light of their liver condition and co-morbidities.

AIM

To characterize DS use in SLD patients and explore motivations for their use.

METHODS

Adults ≥ 18 years old with complete transient elastography and dietary data from the National Health and Nutrition Examination Survey between January 2017 and March 2020 were studied. SLD was defined using consensus criteria, combining clinical indicators with elastography thresholds. The DS Questionnaire (DSQ) was used to record participants’ use of DSQ. Sample weights were applied to estimate national prevalence.

RESULTS

Of 2413 participants with SLD, 1058 reported using DS, for an estimated prevalence of 44.8% [standard error (SE) 2.4] with an average of 2.6 (SE 0.2) DS per person. Among SLD participants taking DSQ, 53.2% (SE 3.3) reported using non-vitamin/non-mineral ones, with an average of 1.8 (SE 0.1) such supplements per person. DS users were more likely to be female, have higher levels of education, and have greater food security (P < 0.02 for all). The most common motivations for using DS were to follow doctor’s advice (36.7%, SE 1.8), to improve overall health (22.1%, SE 2.0), and to maintain health (19.2%, SE 1.9).

CONCLUSION

Nearly half of individuals with SLD report taking DS. This study underscores the pressing need to deepen our understanding of DS use/motivations to develop tailored patient counseling strategies.

Key Words: Dietary supplement; Disparity; Steatotic liver disease; Botanical; National Health and Nutrition Examination Survey

Core Tip: Almost half of patients with steatotic liver disease (SLD) report taking dietary supplements (DS). Older patients, women, and patients with higher education/food security are more likely to use DS. Patients with SLD take DS to improve their health and by doctor’s advice. Hepatologists have an opportunity to address DS use during clinic visits.

INTRODUCTION

Dietary supplements (DS), defined by the United States Congress in the Dietary Supplement Health and Education Act, are products (excluding tobacco) designed to supplement the diet, containing one or more dietary ingredients like vitamins, minerals, botanicals, or other constituents[1]. Since the enactment of this legislation in 1994, the prevalence of supplement usage among adults in the United States has increased dramatically, with recent data indicating that over half of adults report taking one or more supplements[2,3]. Different reviews have demonstrated the efficacy of DSs in many conditions, including omega-6 fatty acids to reduce the risk of myocardial infarction in high-risk patients, psyllium for better blood pressure and glycemic control, St. John’s wort for major depression, glucosamine for osteoarthritis, and ginseng for cancer-related fatigue[4-10].

When assessing the use of DSs, individuals with steatotic liver disease (SLD) may be a unique group. Patients with SLD may feel especially compelled to take supplements to support their liver health and to reduce their liver inflammation, as there had been no medication approved for the indication of inflammation in the setting of metabolic dysfunction-associated SLD (MASLD) with the exception of the recent approval of resmetirom in 2024[11]. Furthermore, management of metabolic co-morbidities (e.g., obesity, diabetes) that accompany SLD may increase patients’ desire to consider DSs to support their metabolic health. On the other hand, the risks of DS-induced hepatotoxicity may be elevated for those with SLD and fibrosis above the general population, altering the risk-benefit ratio of DS use in this population. These considerations underscore the importance of characterizing DS use in patients with SLD.

While a few studies have explored the potential efficacy of certain DSs in patients with SLD, to our knowledge, the current patterns of DS use in this population is unknown[12-14]. Therefore, in this study, we aimed to estimate and characterize DS use among patients with SLD in the United States and explore motivations for their use. These data are an essential foundation for developing patient-centered approaches to managing SLD in today’s clinical practice.

MATERIALS AND METHODS

Data source and study population

We analyzed data from the National Health and Nutrition Examination Survey (NHANES) survey from January 2017 to March 2020. NHANES is a nationally representative, cross-sectional survey employing a complex, stratified, multistage probability cluster sampling design to sample non-institutionalized United States residents[15]. This survey is collected by the National Center for Health Statistics (NCHS) in two-year stages. The protocol includes an in-person interview and a health examination in a mobile center for each participant. Written informed consent is obtained from all participants or proxies and NHANES survey protocol is approved by the Research Ethics Review Board at the Centers for Disease Control and Prevention, NCHS.

We included patients aged 18 years or older with available controlled attenuation parameters (CAP) and liver stiffness measures (LSMs) obtained by vibration-controlled transient elastography (n = 7768, 50%). We excluded patients with missing data on DS usage (n = 410, 5%) and alcohol consumption (n = 400, 5%). Pregnant and lactating women were also excluded from the analysis.

SLD population

We identified a subgroup of patients with SLD from the entire participant pool in the survey, based on validated criteria for identifying SLD patients in the NHANES database[16]. This SLD population included patients meeting the definition for MASLD, MetALD, ALD, or specific etiology/cryptogenic SLD as follows: (1) MASLD = steatosis (CAP of 288 dB) + 1 metabolic risk factor; (2) MetALD = MASLD + average daily alcohol intake of 20-50 g (women)/30-60 g (men) in the past 12 months; (3) ALD = steatosis (CAP of 288 dB) + average daily alcohol intake > 50 g (women)/> 60 g (men) + 1 metabolic risk factor OR steatosis (CAP of 288 dB) + average daily alcohol intake 20 g (women)/30 g (men) + no metabolic risk factors; and (4) Specific-etiology or cryptogenic SLD = steatosis + no metabolic risk factors + average daily alcohol intake < 20 g (women)/< 30 g (men). Specific etiologies include drug-induced liver injury, monogenic diseases (e.g., Wilson disease, inborn errors of metabolism), and miscellaneous causes (e.g., hepatitis C virus and celiac disease), while cryptogenic reflects no identifiable cause.

The above-mentioned metabolic risk factors were defined according to the new consensus nomenclature and followed Lee et al’s methodology[16]: Systolic/diastolic blood pressure of 130/85 mmHg or taking hypertension medications[16]; body mass index 25 kg/m² ( 23 kg/m² if Asian) or waist circumference > 80 cm (women) or > 94 cm (men); fasting plasma glucose level of 100 mg/dL or hemoglobin A1c 5.7% or diabetes mellitus medications; serum level or triglycerides of 150 mg/dL or lipid-lowering agent; high density lipoprotein cholesterol level of < 50 mg/dL (women) or < 40 mg/dL (men) or lipid-lowering agent[16]. Missing values were excluded from the analysis.

Sociodemographic characteristics

All questionnaire data used for this analysis, including demographic and lifestyle data such as age, race, sex, marital status, medical conditions, education level, and alcohol use were collected from participants using the computer-assisted personal interview system during in-home interviews.

Poverty income ratio (PIR), Supplemental Nutrition Program (SNAP) participation status in the last 12 months, and food security were also assessed during the household interview. The PIR is a measure that represents the ratio of household income to the poverty threshold after adjustments for inflammation and family size. ¹⁷Three PIR categories were constructed: Less than 130%, 130% to 185%, and more than 185%. A PIR 130% is the cutoff to determine financial eligibility for SNAP, an extensive nutrition assistance program that provides vouchers for food purchases to improve the health of low-income families. Food security was assessed using 10 questions in the USDA’s Food Security Survey Module and further classified into 4 categories: Full food security (0 affirmative responses), marginal food security (1-2 affirmative responses), low food security (3-5 affirmative responses) or very low food security (6-10 affirmatives responses)[17,18].

DS usage

The DS Questionnaire (DSQ) was used to collect detailed data on the participant’s use of vitamins, minerals, botanicals, and other supplements over the past month.19 Trained NHANES interviewers inquired about supplement usage and asked participants to present containers of all products taken during this period. Interviewers recorded complete label information for each reported DS. If containers were unavailable, participants were asked to recall product details as accurately as possible. The ingredients of each supplement were identified and counted. Each specific ingredient was further classified into 5 categories, namely vitamins, minerals, amino acids, botanicals, and other ingredients. Motivations for supplement use were assessed using a categorical question, allowing participants to select multiple reasons from a provided list or offer their own.

The elastrography measurements were obtained in the NHANES Mobile Examination Center using the Fibroscan^® model 502 V2 Touch equipped with a medium (M) or extra-large (XL) probe. Advanced fibrosis was defined as a LSMs 11.7kPa, per previously defined threshold[19,20].

Statistical analysis

All analyses were performed using the STATA survey procedures package to accommodate the cluster-stratified complex design of NHANES (Version 18, College Station, TX). Sampling weights provided by NHANES were applied to derive nationally representative estimates as described[21] Weighted means and percentages were calculated for demographic factors and clinical characteristics. In exploratory analyses, distinctions between the SLD population using DSs and those not using DSs were evaluated using the Wald score and χ² test. Frequencies of DS ingredients and reported motivations were weighted and presented in terms of the top 10 most frequent occurrences. A two-sided P-value of < 0.05 was considered statistically significant.

RESULTS

Baseline characteristics of individuals with SLD

From 2017 to 2020, out of the 6958 patients surveyed in the NHANES study, 2413 individuals (34.7%) met the predefined criteria for a diagnosis of SLD. Among patients with SLD, 44.8% (n = 1058, SE 2.4) reported using DSs in the past month, averaging 2.6 (SE 0.2) supplements per patient. Of those using supplements, 53.2% (SE 3.3) utilized non-vitamin/non-mineral products, averaging 1.8 (SE 0.1) botanical or other similar supplements per patient. Notably, 9.2% (SE 1.1) of SLD supplement users only used non-vitamin and non-mineral supplements. Baseline characteristics of patients with SLD are presented in Table 1. DS users and non-users were clinically similar in age (although the difference between the groups was statistically different). Compared to non-users, supplement users were more likely to be women (49% vs 39%), have attained a high school degree or higher (93% vs 87%), and report full food security (86% vs 74%). Supplement users were more likely to have a Poverty-Income Ratio greater than 185% compared to non-users (79% vs 64%), suggesting that supplement users generally have a higher income level relative to the poverty line. Patients who reported using DSs had a higher prevalence of diabetes (36% vs 25%) and hypertension (62% vs 40%). While MASLD cases predominated in both DS usage and non-usage groups, the DSs usage group demonstrated a higher proportion of MetALD and a lower proportion of ALD compared to the non-usage group (8% vs 3% and 0.6 vs 1.7%, respectively). However, the presence of advanced liver fibrosis was similar between the two groups (8% vs 6%).

Table 1 Baseline characteristics of patients with steatotic liver disease with and without dietary supplement use, % (SE).

Variable	No dietary supplement use (n = 1355)	Dietary supplement use (n = 1058)	P value
Age			< 0.001
18-39 years old	40.1 (3.0)	13.5 (2.4)
40-59 years old	37.8 (3.0)	36.9 (3.2)
≥ 60 years old	22.0 (3.6)	49.6 (3.9)
Female	39.3 (2.6)	49.2 (3.1)	0.02
Race/ethnicity			0.001
Mexican American	20.0 (3.5)	5.4 (1.2)
Other Hispanic	7.5 (1.0)	4.9 (1.2)
Non-Hispanic White	53.5 (3.8)	76.7 (2.7)
Non-Hispanic Black	9.0 (1.4)	6.6 (1.5)
Non-Hispanic Asian	3.8 (0.7)	4.6 (1.1)
Other	6.5 (1.8)	1.8 (0.6)
Marital status			0.01
Married/living with partner	66.1 (4.3)	72.8 (3.4)
Widowed/divorced/separated	14.7 (2.7)	17.9 (1.5)
Never married	19.2 (3.1)	9.3 (2.6)
High school graduate/GED	86.5 (1.2)	93.0 (0.7)	< 0.001
Full food security	73.5 (2.0)	85.9 (1.6)	< 0.001
SNAP participation1	50.5 (4.3)	47.5 (6.5)	0.62
Obesity	73.1 (3.0)	67.1 (3.1)	0.12
Poverty income ratio > 185%	63.8 (3.0)	79.4 (1.6)	0.001
Diabetes	25.3 (2.1)	35.9 (4.0)	0.03
Current smoker	34.5 (4.2)	25.7 (4.3)	0.14
Hypertension	39.6 (3.7)	62.1 (4.3)	< 0.001
Dyslipidemia	72.5 (3.3)	62.7 (3.7)	0.07
Advanced fibrosis	6.1 (1.2)	8.3 (2.4)	0.41
Liver disease etiology			0.02
MASLD	94.1 (0.7)	91.0 (1.9)
MetALD	3.9 (0.6)	8.4 (1.9)
ALD	1.7 (0.4)	0.6 (0.4)
Cryptogenic/specific	0.3 (0.2)	0 (0)
Aware of liver disease diagnosis	0.03 (0.03)	0.04 (0.01)	0.81

¹Supplemental Nutrition Program participation status in the last 12 months.

SNAP: Supplemental Nutrition Program; GED: General Education Department, MASLD: Metabolic Associated Steatotic Liver Disease; MetALD: Metabolic Alcohol-Associated Liver Disease; ALD: Alcohol-Associated Liver Disease.

Reasons for taking DSs

Adults with SLD predominantly cited two reasons for using DSs: To “follow doctor’s advice” (36.7%, SE 1.8) and to “improve overall health” (22.1%, SE 2.0) (Table 2). While motivations related to specific organ health were less common, other notable reasons included improving bone and heart health, enhancing immunity, and supplementing the diet. These motivations were consistent among those exclusively using non-vitamin and non-mineral DSs (Table 3). Improving liver health was also a motivation to use DSs among a small percentage of adults using any DSs (1.2%, SE 0.5) and those using only non-vitamin and non-mineral DSs (3.6%, SE 1.5).

Table 2 Top 10 reported motivations for dietary supplement use among patients with steatotic liver disease (All product types = 3792).

Motivation	% (SE)
Doctor’s advice	36.7 (1.8)
To improve overall health	22.1 (2.0)
To maintain health	19.2 (1.9)
To prevent health problems	9.9 (1.2)
For bone health	9.7 (0.9)
To supplement diet	9.5 (1.1)
To boost immunity	7.6 (1.2)
For heart health	7.7 (1.5)
For healthy joints	5.4 (1.0)
For healthy hair/nails	5.1 (1.1)

Table 3 Top 10 reported motivations of patients with steatotic liver disease with exclusive non-vitamin and non-mineral dietary supplement use.

Motivation	% (SE)
To improve overall health	41.1 (9.4)
Doctor’s advice	15.2 (7.0)
For bowel health	7.6 (5.7)
To maintain health	10.8 (3.3)
To prevent health problems	7.2 (3.0)
For stress	6.9 (4.1)
For blood sugar	6.1 (2.8)
To boost immunity	6.0 (3.3)
For heart health	5.2 (1.6)
For healthy joints	4.8 (2.1)

Intake of specific DS ingredients

Table 4 provides a breakdown of the most frequently encountered nutrients in the DSs reported by patients with SLD. NHANES respondents with SLD reported taking 649 unique vitamin or mineral DSs. In the vitamin subcategory, vitamin B complex was the most common DS, accounting for 16.2% (SE 0.6) of DSs taken by the SLD population with vitamin D (6.7%, SE 0.4) and vitamin C (4.2%, SE 0.3) following. Among minerals, calcium (4.1%, SE 0.2), magnesium (3.0%, SE 0.2), and zinc (2.2%, SE 0.2) were the most frequently used supplements. Taurine (2.6%, SE 0.1) and cysteine (1.6%, SE 0.1) were the two most commonly reported amino acids used.

Table 4 Top 10 dietary supplements per category among patients with steatotic liver disease.

Dietary supplement	% (SE)
Vitamins
Vitamin B complex	16.2 (0.6)
Vitamin D	6.7 (0.4)
Vitamin C	4.2 (0.3)
Folate and derivates	2.9 (0.2)
Vitamin E and derivates	2.8 (0.2)
Vitamin A	2.2 (0.1)
Vitamin K and derivates	1.2 (0.2)
Choline	0.6 (0.1)
Minerals
Calcium	4.1 (0.2)
Magnesium	3.0 (0.2)
Zinc	2.2 (0.2)
Copper	1.8 (0.1)
Selenium	1.6 (0.1)
Manganese	1.6 (0.2)
Chromium	1.6 (0.1)
Iron	1.5 (0.2)
Iodine	1.4 (0.1)
Molybdenum	1.2 (0.1)
Botanicals
Turmeric	0.5 (0.1)
Saw palmetto	0.3 (0.1)
Cinnamon	0.3 (0.1)
Gingko biloba	0.3 (0.1)
Olive	0.2 (0.04)
Cranberry	0.2 (0.1)
Grape	0.2 (0.1)
Green tea	0.1 (0.1)
Garlic	0.1 (0.03)
Ginger	0.1 (0.04)
Amino acids
Taurine	2.6 (0.1)
Cysteine and derivatives	1.6 (0.1)
Glycine	0.2 (0.1)
Lysine	0.2 (0.1)
Glutamic acid and derivates	0.1 (0.1)
Alanine	0.1 (0.1)
Tyrosine	0.1 (0.1)
Arginine and derivates	0.1 (0.1)
Carnitine	0.1 (0.04)
Leucine	0.1 (0.1)
Other supplements
Omega 3 supplements	3.1 (0.2)
Fish oil	1.2 (0.1)
Lutein	0.8 (0.1)
Inositol	0.7 (0.2)
Lycopene	0.7 (0.1)
Glucosamine	0.6 (0.1)
Coenzyme Q10	0.5 (0.1)
Fiber	0.5 (0.1)
Chloride	0.4 (0.1)
Melatonin	0.3 (0.1)

In total, NHANES respondents reported taking 751 unique non-vitamin and non-mineral supplements (e.g., botanicals, amino acids, fatty acids). Turmeric was the most prevalent botanical supplement ingredient reported, found in 0.5% (SE 0.1) of DSs. Palmetto (0.3%, SE 0.1) and cinnamon (0.3%, SE 0.1) were also commonly used. Omega-3/fish oil supplements were present in 4.3% (SE 0.3) of the DSs. Additionally, lutein, an antioxidant, was found in 0.8% (SE 0.1) of DSs taken by patients with SLD.

Liver-health-specific supplements, including milk thistle, n-acetylcysteine, and curcumin, were each found in only 0.1% (SE 0.1, SE 0.1, and SE 0.01 respectively) of DSs. Notably, 3.5% (SE 0.6) of supplements contained calories, 2.1% (SE 0.5) included additional fats, and 0.9% (SE 0.2) of supplements had “additional sugars” in their nutritional details. Potassium and sodium were present in 1.5% (SE 0.2) and 0.8% (SE 0.1) of supplements, respectively.

DISCUSSION

In this study of patients meeting criteria for SLD in the NHANES database, nearly half reported using DSs in the past month. Hundreds of types of supplements were identified, primarily taken to enhance overall health or as per medical recommendations. With the rise in prevalence of SLD patients in hepatology clinics and high prevalence of DS use, this study underscores the importance of understanding DSs use among patients with SLD. Hepatologists should engage in open conversation with a holistic, collaborative, and non-judgmental approach regarding management of supplement use.

Certain demographic factors were strongly associated with DS usage. Specifically, individuals who were women, older than 60 years, had higher educational attainment, and/or reported food security were more likely to report using DSs. This data is consistent with prior studies in the general population[22,23]. Furthermore, the types of supplements consumed by patients with SLD mirror those commonly reported in the general adult population, with a high prevalence of supplements such as vitamin D, vitamin B complex, and calcium[23,24].

It is essential to acknowledge why individuals with SLD report taking DSs. Over one-third of individual reported taking supplements based on doctor’s advice, which likely reflects recommendations for taking vitamins and minerals. The other “top 10” reasons for supplement use reflected a desire by individuals to promote positive health benefits, particularly in areas where conventional medicine leaves a gap (i.e., health promotion, healthy hair). These motivations represent an opportunity for clinicians to have discussions with their patients who are “primed” to engage in proactive steps to improve their overall health and well-being. In other words, discussing these motivations can open further conversation regarding recommendations for other health-promoting activities, including dietary and physical exercise.

Equally important to creating an environment in which individuals feel comfortable discussing DSs with their practitioner is the potential for drug-induced liver injury, which may be of particular importance for patients with SLD. While this is a relatively rare occurrence-a previous study found that 3% of patients with liver disease reported consuming potentially hepatotoxic DSs-it is a critical consideration[25]. We also observed that patients with SLD consumed DSs containing calories, fat, and sugars, likely without realizing the potential for these added macronutrients to worsen their metabolic dysfunction. This further emphasizes the importance of detailed evaluation of non-prescribed substances for potential supplement-drug interactions and adverse effects on patient health.

Past research has identified that almost a third of DS users do not disclose their supplement intake during clinical visits[26]. Various factors contribute to this reluctance, including patients’ beliefs that discussing supplements is unnecessary and concerns that their healthcare provider may consider the supplements unsafe[2]. Nevertheless, the most common reason for nondisclosure has been lack of inquiry by medical providers[2]. Studies regarding physician perception of DSs revealed concerns of inadequate regulation of these agents and potential interactions and adverse effects[27,28]. Nonetheless, the high prevalence of DS users among SLD patients highlights the necessity for open communication between patients and providers in the clinic setting.

Exploration of the potential benefits that certain DSs may offer in improving liver health is also essential. While various medical treatments have emerged to manage and slow the progression of SLD, there is a growing interest in exploring the potential liver benefits of specific dietary nutrients[29]. For instance, DSs such as curcumin, milk thistle, and N-acetyl cysteine have been demonstrated to attenuate hepatic lipid accumulation and prevent fibrosis progression in animal models[30]. Targeting these DSs has been suggested as a potentially beneficial therapeutic approach for patients with SLD[31]. Despite these known benefits, our study reveals that only a small proportion of the DSs taken by patients contained these three aforementioned liver-specific ingredients. This may be because a portion of the study population was unaware of their underlying liver condition and, as a result, may not have actively sought DS to improve their liver health. Nonetheless, future efforts should further explore the potential benefits of these liver-specific nutrients.

We acknowledge the following limitations. First, there is potential self-selection bias inherent to the NHANES survey, as more health-conscious individuals may be more inclined to participate. NHANES is a cross-sectional survey, and despite efforts to ensure national representativeness to account for nonresponse and noncoverage through sample weights, it is impossible to eliminate selection bias[32]. Those with missing DS survey data were excluded from the study, which may also lead to an underestimation of the number of individuals using DSs particularly if some are hesitant to disclose usage. The DSQ allows participants to recall product details from memory if containers were unavailable and not independently verified by source documents. This may limit the accuracy of some data; however, this method is also used in clinic settings. This type of self-reported data is subject to additional types of bias including recall bias and non-response bias. Additionally, individuals with missing metabolic factors were not included in the study, which can further lead to selection bias. Furthermore, the estimates provided in this study encompass patients with SLD who are unaware of their condition. The limited sample size precluded our ability to compare patients who were and were not aware of their liver condition. Given this limitation and the lack of liver-specific health outcomes available in the NHANES database, the interactions between different DS components and their effect on the health of patients with SLD were not explored.

CONCLUSION

Nevertheless, this study offers a comprehensive overview of DS usage among patients with SLD, shedding light on the widespread utilization of these supplements among a significant proportion of individuals with this liver condition. The findings provide detailed insights into the specific types of supplements taken and the motivations behind their usage, contributing to a better understanding of the DS habits of individuals with SLD. Future studies should consider developing multicenter cohorts to better study DS use in patients with SLD and to identify liver-specific health outcomes. Exploring interactions between different DSs and their effect on the health of SLD patients may further enhance clinical guidance. This study provides evidence for the need to closely examine the safety and efficacy of hepatology-related supplements and to develop clinical guidelines regarding the use of DSs within this patient population.