https://orcid.org/0000-0001-5273-4522
Figures
Abstract
Background
Visceral obesity (VATob) increases the risk for many diseases. Central obesity has been associated with an augmented prescription use; however, there is a paucity of research focused on VATob. Here, the aim was to evaluate the association between VATob and prescription use.
Methods
Data was collected from the NHANES dataset (2011–2018). Visceral adipose tissue was measured using dual x-ray absorptiometry, and VATob was defined as ≥100 cm2. Prescription use was collected from the RXQ_RX files and classified according to Vademecum. Association between VATob and prescription use was determined with logistic regression and reported as odds ratios (ORs) with 95% confidence intervals (95%CIs).
Results
10,952 participants (weighted: 121,090,702) were included, in which 41.8% were VATob and 52.0% of them had ≥1 prescription. Overall, VATob demonstrated an augmented rate of prescription use when compared to non-VATob (52.0% versus 36.7%, p<0.001), specifically with metabolic (4.5-fold increase), cardiovascular (3.9-fold increase), gastrointestinal (2.5-fold increase), and hematopoietic agents (2.3-fold increase). This was associated with increased the risk for overall prescription use (ORoverall = 1.9, 95%CI: 1.7–2.1, p<0.001). Similar results were observed with metabolic and cardiovascular agents. However, when stratified by BMI, normal weight participants (ORmetabolic = 10.4; 95%CI: 6.5–16.6 & ORcardiovascular = 7.0; 95%CI: 4.4–11.1, p<0.001) had a greater risk than the overweight (ORmetabolic = 4.1; 95%CI: 3.0–5.6 & ORcardiovascular = 3.4; 95%CI: 2.5–4.7, p<0.001) or obese participants (ORmetabolic = 3.5; 95%CI: 2.3–5.3 & ORcardiovascular = 3.5; 95%CI: 2.5–4.9, p<0.001).
Citation: Acevedo-Fernández M, Ochoa Precoma R, Porchia LM, Posadas VM, Torres-Rasgado E, Gonzalez-Mejia ME, et al. (2025) Visceral obesity augments prescription use: An analysis of the cross-sectional study of NHANES 2011–2018. PLoS ONE 20(2): e0318413. https://doi.org/10.1371/journal.pone.0318413
Editor: Neftali Eduardo Antonio-Villa, Instituto Nacional de Cardiologia Ignacio Chavez, MEXICO
Received: July 7, 2024; Accepted: January 15, 2025; Published: February 3, 2025
Copyright: © 2025 Acevedo-Fernández et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All data generated or analyzed during this study are available at the NHANES website (https://wwwn.cdc.gov/nchs/nhanes/default.aspx). The raw data for this specific subset used for this analysis is available at Harvard Dataverse website (https://doi.org/10.7910/DVN/HRULLS).
Funding: This study was supported by grants from the Programa para el Desarrollo Profesional Docente to MEGM and ETR (CA-160 FACMED) and the Vicerrectorıa de Investigacion, Benemerita Universidad Autonoma de Puebla, Mexico to ETR (100493499-viep2023). The funders had no role in the study’s design, data collection or analysis, decision to publish, or preparation of the manuscript.
Competing interests: Authors declare no competing interest.
Introduction
Prescription use has increased dramatically over the past 50 years. In the United States of America (USA), from 2009 to 2022, the annual number of prescriptions rose from 4.0 billion to 6.7 billion [1], equating to about 19 prescriptions per citizen. Per capita spending increased from around $200 USD in the early 1980s to between $1000 and $1200 USD by 2015 and has remained constant. Since 2009, the cost of brand-name prescription drugs has doubled [2]. Improper dosage (overdosing and underdosing), inappropriate prescribing (targeting and selection), and inadequate monitoring have contributed to an increase in adverse drug effects and drug interactions [3]. Additionally, some clinicians prescribe more medications to counter these adverse effects. Hoel et al. suggest that polypharmacy (5 or more medications) is associated with increased hospital admissions, adverse drug events, non-adherence, and higher healthcare costs [4]. Consequently, prescription use has become a significant health concern, necessitating efforts to address its root causes.
Several factors contribute to the rise in prescription use, including an aging population, the prioritization of medications as the first-line of treatment (specially for chronic non-communicable diseases), medicalization, prolong medication use without discontinuation, and the increasing prevalence of modifiable co-factors, such as obesity. Among these, obesity is one of the most modifiable contributors to increased prescription use.
Over the years, augmented prescription use has been linked to obesity. Kantor et al. associated this trend with the rise in obesity prevalence [5]. Obesity—a complex, multifactorial disease—is characterized by the growth of bioactive adipose tissues that promote chronic diseases and reduce life expectancy [6]. Since 1980, global obesity prevalence has doubled, with almost 30% of the population now classified as overweight or obese [7]. If current trends persist, it is projected that 60% of the population will be overweight or obese by 2030 [8], likely driving further increases in prescription use.
In clinical practice, obesity is commonly assessed using the body mass index (BMI), which reflects general obesity, comprising both visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT). Importantly, different types of SAT have varying effects on disease risk. For instance, abdominal SAT is linked to metabolic diseases, whereas femoral-gluteal SAT has a neutral effect [9]. Thus, BMI may be a weak predictor of certain pathologies. Studies suggest that abdominal adipose tissue correlates more strongly with metabolic-associated steatotic liver disease (MASLD), insulin resistance, and other metabolic diseases than BMI [10–12]. Consequently, abdominal adipose tissue might serve as a better indicator for evaluating association with prescription use [7, 13].
VAT is particularly metabolically active compared to SAT, producing adipokines that augment metabolic and cardiovascular risk [14]. VAT is the adipose tissue that is posited around the organs in the peritoneal cavity and visceral obesity (VATob), also known as “Hidden Obesity”, is a threshold that is associated with an increased risk for certain diseases [15]. Numerous studies have demonstrated that VATob increases the risk of dyslipidemia, Type 2 Diabetes, hypertension, cardiovascular disease, and other metabolic abnormalities [16–18], potentially increasing prescription use. Interestingly, VATob in individuals with normal BMI (18.5–24.9 kg/m2) has been linked to a higher risk of MASLD than in obese individuals with VATob (BMI range: 30.0–39.9 kg/m2) [19]. However, measuring VAT is not routine in clinical practice, leaving many VATob cases undiagnosed across BMI categories.
Since most obesity-related complications require pharmaceutical intervention as part of the treatment [20], it is posited that the association between VAT and prescription use parallels that observed with general obesity [21].
Here, the aim was to evaluate the association between VATob and prescription use using data from the National Health and Nutrition Examination Survey (NHANES).
Materials and methods
Data source
This study was conducted following the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines (S1 Table). Data were obtained from the publicly available NHANES dataset, administered and managed by the National Center for Health Statistics (NCHS) [22]. The data was downloaded and compiled on March 2, 2023. Due to the NCHS policies, all participant identities remained anonymous to the authors. To have a nationally representative survey of the health and nutrition status of the USA, the NCHS uses a stratified multistage probability model. The NHANES received its ethical approval for all study protocols from the NCHS Research Ethics Review Board [23]. All participants provided written consent by signing a comprehensive informed consent [23]. All procedures followed by the NHANES were in accordance with the Declaration of Helsinki of 1975.
This study utilized data from 2011 to 2018. To be part of this study, the participants had to 1) have values of VAT, as measured by whole-body dual-energy x-ray absorptiometry (available for individuals aged 8 to 59 years) and 2) be ≥18 years old. The participants were excluded if they had the following conditions: 1) missing BMI values or were <18.5 kg/m2, or ≥40 kg/m2; 2) were pregnant; 3) suffered from viral diseases or potential infections (HIV, Hepatitis B, B/D, or C); or 4) had cancer. These exclusion criteria were chosen because the listed conditions could independently increase prescription use through mechanisms unrelated to adipose tissue.
Measurement methods and instrumentation
Data was collected according to the standardized protocols for the NHANES [24]. Three classes of variables were collected: anthropometric measurements, demographic data, and laboratory tests. The anthropometric variables obtained were height (cm), weight (kg), BMI (kg/m2), waist circumference (cm), systolic and diastolic blood pressure (SBP and DBP, mmHg), and VAT (cm2). VAT was categorized using a cutoff value of 100 cm2 into VATob or non-VATob [25]. According to the World Health Organization, BMI was classified into normal weight (18.5–24.9 kg/m2), overweight (25.0–29.9 kg/m2), obese class 1 (30.0–34.9 kg/m2), or obese class 2 (35.0–39.9 kg/m2). Hypertension was defined as SBP ≥ 130 mmHg or DBP ≥90 mmHg.
Demographic data included biological sex (male and female), age (years), and ethnicity (Non-Hispanic Black, Non-Hispanic White, Mexican American, other Hispanics, or other races), education level [<9th grade, 9-11th grade (including 12th grade with no diploma), High School graduate/GED or equivalent, some college or associate degree, college graduate or above], marital status (Single, Married, Divorced/Separated, Widowed), income (<$19.9k, $20-$74.9k, $75-$99.9k, ≥$100k) and physical activity. For physical activity, the metabolic equivalent (MET) was calculated using data from the Global Physical Activity Questionnaire [26]. This questionnaire uses 2 components: types of daily activities (work, recreational, walking/biking) and intensity (moderate and vigorous). The data collected was the frequency of the activity and the duration spent per month or week. To calculate the MET-minutes/week, vigorous-intensity activities were assigned as 8.0 MET, whereas moderate-intensity activities were assigned as 4.0 MET. For each participant, each MET category was calculated by multiplying the number of days by the average time by the corresponding MET modifier. Total MET score was estimated by summing up 1) vigorous work activity/vigorous recreational activities, 2) moderate work activity/moderate recreational activities, and 3) walking/bicycling. MET scores was classified as low (<600 MET-minutes/week), moderate (600–2999 MET-minutes/week), and high (≥3000 MET-minutes/week). Insurance data was also obtained: 1) if the participants were covered by insurance; 2) if they had private insurance, and 3) if their plans covered medications.
During the interview, the survey participants were asked if they are currently using prescription medications. Those who answered affirmatively provided information concerning all prescription medications, in which the containers were examined and medication names were recorded, duration, and reasons for use. Prescription data were obtained from RXQ_RX files and classified using Vademecum [27] into 1) cardiovascular, 2) metabolic agents, 3) gastrointestinal, 4) central nervous system, 5) psychotherapeutic, 6) respiratory, 7) hematopoietic, 8) topical nasal, 9) antineoplastic, 10) anti-infective, 11) hormones/hormone modifiers, 12) topical, 13) ophthalmic, or 14) otic agents.
The laboratory variables collected were total cholesterol (mg/dL), high-density lipoprotein cholesterol (mg/dL), and glycosylated hemoglobin (HbA1c, %). According to the American Diabetes Association criteria, subjects were classified as either normal glucose tolerance (HbA1c: <5.6%), prediabetics (HbA1c: 5.6–6.4%) and Type 2 Diabetics (HbA1c: ≥6.5%). The laboratory variables were only reported if the missing data for a variable was <5%.
Statistical analysis
All statistical analyses were performed using the Statistical Package for the Social Sciences software version 26.0 (SPSS, IBM Corp., Armonk, NY, USA). The complex samples design option was used to account for NHANES’s stratified multistage probability sampling design. For quantitative and categorical variables, the mean or the percentage, respectively, with standard errors were calculated. For differences between the groups, the Rao Scott-Chi2 test was used for categorical variables, whereas the designed-adjusted T-test was used for continuous variables. The association between VAT and prescription use was determined with univariate and multivariate logistic regression and reported as odds ratios (ORs) with 95% confidence intervals (95%CIs). P-values were considered significant if they were <0.05 (two-tailed). The participants were classified in to 4 groups based on 2 characteristics: 1) VATob and 2) Prescription use. VATob was defined as VAT ≥100 cm2. Positive prescription use was defined as participants taking one or more medications. The 4 groups are 1) the control group: without VATob and without prescription use (non-VATob/RX-), 2) the prescription only group (non-VATob/RX+), 3) the VATob only group (VATob/RX-), and 4) the VATob who are taking a prescription group (VATob/RX+). For the complete cohort, participants were included if data was present for VAT and prescription use. Other variables were shown only if <5% of the data was missing. When selecting variables to be adjusted, if the sample size decreased by >10% as well as the portions between the independent and dependent variables changed by >1%, then the variable would not be included.
Results
Selection of participants
Of the 39,156 potential participants, 49.2% had acceptable DEXA scans; however, 8,308 records were removed due to age, being underweight or morbidly obese, or having a condition that could affect adiposity (Fig 1). This resulted in 10,952 participants (weighted: 121,090,702) included in this study. The characteristics of the cohort are shown in Table 1.
Overall, the VATob rate was 41.8±1.0%, with 47.4±1.3% of males and 35.5±1.2% of females having VATob. Interestingly, 7.3±0.8% of normal-weight participants had VATob, compared to 43.2±1.4%, 71.2±1.4%, and 83.4±1.5% of overweight, obese class 1, and obese class 2 participants, respectively. When stratified by VATob and prescription use, 36.8% of the cohort was classified as non-VATob/RX-, whereas 20.1% of the cohort was VATob/RX-, 21.3% were non-VATob/RX+, and 21.8% were VATob/RX+. Significant differences were observed among the four group with respect to race, marital status, education level, income, and physical activity. The principal ethnicity for all groups was Non-Hispanic Whites. Non-Hispanic Blacks were the second most prevalent for the non-VATob groups, while Mexican Americans were more prevalent in the VATob groups. As expected, the VATob groups were older with elevated weight, waist circumference, BMI (specifically the prevalence of overweight and obese classes 1 and 2), SAT, systolic and diastolic blood pressure, HbA1c, and total cholesterol levels. Additionally, the prevalence of hypertension, prediabetes, and Type 2 Diabetes was higher among the VATob groups. For each group, >67% had some form of health insurance, with private insurance being the most common form. Of participants with insurance, >90% of the insurance plans had prescription coverage.
VATob effect on prescription use
The number of prescriptions among participants ranged from none to 17, with approximately 0.6% reporting the use of 10 or more prescriptions. A significant positive correlation was observed between VAT and the number of prescriptions (r = 0.281, p<0.001, Fig 2A). Concerning medicine classes, 1 person was taking 10 different kinds of medication, none were taking 9 different classes, 7 were taking 8 different classes, 20 were taking 7 different classes, 52 were taking 6 different classes, and 114 were taking 5 different classes. Due to the low reporting rate, ≥5 was grouped together. VAT also significantly correlated with the number of medication classes (r = 0.263, p<0.001, Fig 2B).
Means (dot) and standard error (bars) plots were constructed the determine the relationship between VAT and number of prescriptions (A) and the number of medication classes (B).
When VATob was considered, participants with VATob had a higher rate of prescription use compared to those without VATob (52.0±1.0% vs. 36.7±0.9%, p<0.001, Table 2). When stratified by BMI class, the proportion of VATob users taking at least one medication was significantly higher in the VATob group than in the non-VATob group for normal-weight (53.1±4.9% versus 37.5±1.2%, respectively, p = 0.002), for overweight (47.6±1.7% versus 35.9±1.8%, respectively, p<0.001), for obese class 1 (54.4±1.6% versus 33.7±2.7%, respectively, p<0.001), for obese class 2 (54.8±2.1% versus 41.6±3.5%, respectively, p = 0.002). When stratified by the number of prescriptions, the non-VATob group was less likely to be taking ≥2 medications than the VATob group (17.1±0.7% versus 34.5±1.0%, respectively, p<0.001), and similarly, they were less likely to be taking ≥5 medications (3.0±0.3% versus 9.7±0.6%, respectively, p<0.001).
Further stratification by general obesity revealed a significant difference in the distribution of the number of prescriptions used by VATob participants according to BMI class (p<0.001). For normal weight participants, VATob was associated with increased use of 3 and 4 prescriptions (p<0.05, Fig 3) and showed a near-significant association with the use of ≥5 medications (p = 0.073). For overweight participants, VATob was associated with increased use of 3, 4, and ≥5 prescriptions (p<0.05). For obese class 1 participants, VATob was associated with increased use of 2, 3, 4, and ≥5 prescriptions (p<0.05). For obese class 2 participants, VATob was associated with increased use of 4 and ≥5 prescriptions (p<0.05). No effect was observed for the use of only 1 prescription.
Bar charts were constructed for the percentage who were taking 0 (A), 1 (B), 2 (C), 3 (D), 4 (E), or ≥5 (F) prescriptions. For each panel, the rate of use for the complete cohort (Total, white bar) or when stratified into non-VATob (checkered bar) or VATob (black bar) was determined for the total cohort (Total) or when stratified by BMI class: normal weight (NW), overweight (OW), obese class 1 (OB1), and obese class 2 (OB2). * Indicates a significant difference (p<0.05) between non-VATob and VATob group.
For the number of medication classes, participants in the non-VATob group were less likely to take ≥2 different classes of medications compared to those in the VATob group (14.4±0.6% versus 31.1±1.0%, respectively, p<0.001). Similarly, the non-VATob group was less likely to take ≥5 different classes of medications (1.1±0.2% versus 3.1±0.3%, respectively, p<0.001). When stratified by general obesity, there was significant difference in the distribution medicine classes between VATob and non-VATob participants based on BMI class (p<0.001). For normal-weight and overweight participants, VATob was associated with an increased prevalence of taking medications from 2, 3, and 4 different classes (p<0.05, Fig 4). For obese class 1 participants, VATob was associated with an increased prevalence of taking medications from 2, 3, 4 and ≥5 different classes (p<0.05). For obese class 2 participants, VATob was associated with an increased prevalence of taking medications from 3 and ≥5 different classes (p<0.05). No effect was observed for 1 medication class.
Bar charts were constructed for the percentage who were taking 0 (A), 1 (B), 2 (C), 3 (D), 4 (E), or ≥5 (F) different medication classes. For each panel, the rate of use for the complete cohort (Total, white bar) or when stratified into non-VATob (checkered bar) or VATob (black bar) was determined for the total cohort (Total) or when stratified by BMI class: normal weight (NW), overweight (OW), obese class 1 (OB1), and obese class 2 (OB2). * Indicates a significant difference (p<0.05) between non-VATob and VATob group.
With respect to the type of medication, the most commonly prescribe overall were cardiovascular agents (13.3±0.5%), metabolic agents (10.4±0.4%), central nervous system agents (11.8±0.5%), gastrointestinal agents (6.5±0.4%), anti-infective agents (12.8±0.6%), and psychotherapeutics agents (11.4±0.5%). When VATob status was considered, significant associations were observed with several medication classes. VATob was strongly associated with the use of metabolic agents (4.5-fold), cardiovascular agents (3.9-fold), gastrointestinal agents (2.5-fold), hematopoietic agents (2.3-fold), topical nasal agents (1.6-fold), central nervous system agents (1.5-fold), psychotherapeutic agents (1.4-fold), and respiratory agents (1.3-fold, Table 2).
VATob increases the risk of prescription use
Potential confounding variables associated with prescription use were evaluated (Table 3). Univariate logistic regression was employed to identify variables that showed an association with prescription use, and multivariate logistic regression was used to confirm these associations while controlling for potential confounders. Age, sex, race, education, marital status, insurance coverage, and BMI were all significantly associated with augmented prescription use. To account for these confounders, all analyses will control for sex and age (Model 1) as well as race, education, marital status, and insurance coverage (Model 2). The impact of general obesity (BMI) will be further examined through stratification.
The effect VATob has on prescription use is shown in Table 4. In the crude analysis, VATob was associated with increase overall prescription use, particularly for cardiovascular, metabolic, gastrointestinal, central nervous system, psychotherapeutic, respiratory, hematopoietic, and topical nasal agents. After controlling for sex, age, race, marital status, education, income, and insurance coverage (Model 2), significant associations persisted for overall prescription use, and the use of cardiovascular, metabolic, gastrointestinal, and psychotherapeutic agents. The interaction between VATob and general obesity status was also evaluated. Significant interactions were observed for cardiovascular (pinteraction = 0.026), metabolic (pinteraction = 0.003), psychotherapeutic (pinteraction = 0.050), and hematopoietic agents (pinteraction<0.001). When the cohort was stratified by BMI class, a decreasing trend in odds ratios was noted for cardiovascular and metabolic agents, with normal weight participants showing the highest odds ratio. However, after adjustment using Model 2, significant interactions between VATob and general obesity status remained only for metabolic (pinteraction = 0.020) and hematopoietic agents (pinteraction<0.001).
Physical activity influences the association between VATob and prescription use
Given the role of physical activity in reducing VAT, its potential mitigating effects on VATob’s association with prescription use were assessed (Table 5). A significant interaction was observed for overall prescription use in the Crude analysis as well as in Model 1 and Model 2. Participants who were highly active exhibited a lower risk of overall prescription use compared to those with lower levels of physical activity. For specific medication classes, however, no significant differences were detected among participants with low, moderate, or high physical activity levels.
Discussion
Here, VATob was associated with increased prescription use. Among normal-weight participants, VATob increased the likelihood of using 3 or 4 prescriptions. In contrast, among obese participants, VATob increased the likelihood of using of 2 or more prescriptions. There was also a strong association between VATob and the use of cardiovascular and metabolic agents, especially among normal-weight participants.
VATob is an underestimated pathology, with 27–46% of the population affected [28]. Additionally, many individuals are unaware of the VATob status, particularly normal-weight individuals, where the prevalence is 10–25% [28–30]. In our cohort, 41.8% of participants had VATob, including 7.3% of the normal-weight participants. Regarding biological sex, VATob prevalence varies significantly depending on age and BMI, ranging from 22.8% to 96.6% in men and from 3.5% to 61.5% in women, findings that closely support the results of our study [28]. Therefore, our cohort was similar to other published studies with respect to the rate of VATob by biological sex and BMI class.
Prescription use patterns in our cohort showed that 38% of participants were taking at least one medication, lower than the 64% reported by Barrett et al. [31]. Their study, which used BMI to define obesity, found that 64% of obese participants were taking at least 1 medication [31]. Here, 48–53% of obese participants were taking at least 1 medication. Even though both studies used the NHANES with overlapping collection periods, the major difference is the age range. The Barrett et al. study used BMI to determine overall obesity, which was collected for all ages. Here, visceral obesity was collected by DEXA, which was collected for participants aged between 8 and 59 years. The exclusion of older participants likely skewed prescription use rates downward. Nonetheless, there are no studies in which the VATob and prescription use are evaluated.
Obesity significantly affects prescription use. The Counterweight Project demonstrated that as BMI categories increased, there was a corresponding rise in prescription use. Both the total number of prescriptions and the variety of medication classes were significantly higher among individuals in the obese category [21]. A study conducted in the United Kingdom reported a 16% increase in prescription use among obese individuals [21], whereas a study in the USA found an 11.3% increase [31]. These findings highlight the influence of obesity on prescription use despite the differing healthcare systems in these two countries. Here, VATob was associated with a 15.3% increase in the overall prescription use. However, stratifying by BMI class demonstrated no observable trend or differences between each class (Table 4), suggesting that VATob has no promoting effect when compared to general obesity.
When analyzing specific medication classes, VATob was associated with augment used of cardiovascular (17.4%), metabolic (14.8%), gastrointestinal (6.1%), central nervous system (4.8%), psychotherapeutic (4.0%), respiratory (1.4%), hematopoietic (0.9%), and topical nasal (0.9%) agents (Table 2). Interaction analysis showed that BMI class influenced the association of VATob with cardiovascular, metabolic, psychotherapeutic, and hematopoietic agents (Table 4) Specifically, normal-weight participants with VATob had a higher risk of being prescribe a cardiovascular, a metabolic, or a hematopoietic agent. Therefore, this study posits that VAT should be measured in normal-weight individuals.
Here, for normal-weight participants, the risk of receiving cardiovascular, metabolic, or hematopoietic agents was enhanced by VATob, more than overweight and obese participants. Therefore, it is posited that VATob is more detrimental in the absence of substantial SAT. Recently, a condition has been described in which subjects with normal BMIs (18.5 to 24.9 kg/m2) can have elevated VAT, resulting in the presence of metabolic complications. These “hidden obese” subjects could potentially be missed for treatments [32]. This potentially means that the distribution of body fat could be a greater risk factor than either weight gain or BMI [33]. The American Heart Association stated that the level of misdiagnoses of cardiometabolic disease in individual with “normal-weight obesity”—normal weight individuals with metabolic and inflammation profiles similar to obese individuals—is higher. They even suggest using waist circumference, an index more associated with VAT than SAT, could be more prognostic for cardiovascular disease [34]. This could explain the higher odds ratios for cardiovascular and metabolic agent use among normal-weight participants. The “obesity paradox” provides support of this possibility. The “obesity paradox” is when obese patients have better outcomes than other BMI classifications [35]. Several explanations had been posited; however, the counter-active effects of SAT to VAT are very plausible. Peripheral adiposity confers cardiovascular protection due to the secretion of adiponectin, which has an anti-inflammatory effect [36], in contrast to VAT, which secretes interleukin-6, interleukin-8, angiotensinogen, vascular endothelial growth factor, monocyte chemoattractant protein-1, promoting more chronic inflammation and dyslipidemia [14].
Environmental and lifestyle factors, such as high-calorie diets, sedentary behaviors, psychological stress, and aging, contribute to VAT accumulation. VAT serves as an active lipid storage depot with mesenteric fat fostering a high lipogenic environment due to continuous lipid flux [15, 37]. Its proximity to organs allows VAT to exert localized metabolic effects, such as promoting atherosclerosis and cardiovascular disease [18]. VAT plays a central role in driving systemic inflammation and metabolic dysfunction through its dysregulated secretion of pro-inflammatory adipokines, including leptin, interleukin-6, resistin, tumor necrosis factor-α, and visfatin [38]. This pro-inflammatory milieu activates pathways such as the NF-κB pathway, perpetuating chronic inflammation that underlies insulin resistance, metabolic syndrome, MASLD, and cardiovascular disease [3, 39–41]. The metabolic disturbances associated with VAT accumulation are linked to an increased need for medications targeting these conditions, such as lipid regulators, antidiabetics, and cardiovascular agents [42]. Furthermore, VAT’s influence extends to pharmacokinetics, as its expansion alters drug metabolism by affecting hepatic cytochrome activities (e.g., CYP3A4, CYP2C19) and renal clearance [43], impacting the efficacy of medications like statins [43, 44]. Weight gain and VAT accumulation exacerbate cardiometabolic risk via localized inflammation, immune cell activation, and cytokine release, contributing to higher prescription use [45]. Additionally, medications themselves, such as psychotropics and antidiabetics [46, 47], can disrupt metabolic homeostasis and promote VAT accumulation, creating a feedback loop of inflammation and metabolic dysfunction. Collectively, VAT’s unique inflammatory and metabolic profile provides a stronger predictor of medication needs than BMI alone, underscoring its pivotal role in the pathophysiology of chronic disease management.
Healthcare studies have showed an augmented cost associated with obesity, particularly in patients with Type 2 Diabetes, hypertension, depression, and other chronic conditions [48]. Although no studies specifically address the costs associated with VATob, similar expenses are likely. Harrison et al. observed that higher BMI correlated with an increased prevalence of chronic health conditions, which amplifies healthcare costs [49]. A systematic review confirmed that obesity augments healthcare costs, specifically in the USA, obesity-related costs raised from $2.7 billion USD in 2005 to $6.9 billion USD in 2011 [50]. Therefore, it is posited that VATob will augment the cost as well. For that reason, clinicians should focus on reducing VATob as much as weight control. Increasing physical activity is one potential intervention. Our findings demonstrated that higher physical activity levels reduced the likelihood of VATob-associated prescription use. However, this effect was not evident when examining specific medication classes, suggesting the reduction was driven by fewer medications or fewer medication classes overall. Future studies should further investigate this relationship.
Here, 24.4% and 5.8% of the cohort was taking ≥2 or ≥5 prescriptions, respectively. Polypharmacy (use of ≥5 medications) is a critical concern due to its association with adverse outcomes, including increased hospitalizations and mortality [51]. VATob was associated with a greater risk of polypharmacy, particularly among participants classified as overweight or obese class I. While this effect was less evident in the normal-weight and class II obese groups, the limited sample sizes suggest the need for larger studies to confirm these findings. Reducing VAT may help mitigate the risks of polypharmacy and even facilitate discontinuation of certain medications. VAT reduction has been associated with improvements in inflammatory cytokine profiles, insulin levels, and plasminogen activator inhibitor-1 levels [52, 53], as well as promoted longevity [54]. Recently, a study demonstrated that a personalized intervention that included Plant-Based Whole Foods, Time Restricted Eating, and Fractionized Exercise led to a 67% decrease in insulin doses and a 27% reduction in the number of medication categories [29]. However, no studies have specifically assessed the impact of VAT reduction on prescription needs, highlighting an important area for future research.
This study has a few limitations. First, while key factors affecting prescription use were adjusted, other VAT-associated factors, such as smoking and triglyceride levels, were missing. Second, insurance data does not describe the kind of prescription coverage. Different insurances, even from the same provider, can have different types of prescription coverage. Third, smokers were included as well as other lifestyle conditions that affect prescription use [55]. Fourth, the classification of the prescriptions was made according to Vademecum. Other systems could modify the classification, which could move a few medications between categories. Moreover, the reason for the prescription was not considered. Off-script prescription use was not taken into consideration. For example, Ozempic® (semaglutide) can be used for weight loss instead of its glucose-lowering effect. Fifth, prescription use was reported; however, it should be noted that medication compliance and efficacy were not considered. Sixth, VATob was defined using a 100 cm2 threshold. Participants with 100 cm2 or with 400 cm2 were considered the same. Kredel et al. showed that increased VAT levels are associated with more inflammation [14]. When stratified by overall obesity, it is possible that in normal-weight-VATob participants had VAT values closer to 100 cm2, whereas obese-VATob participants had VAT values closer to 400 cm2. Finally, as a cross-sectional study, causal associations cannot be established. Future longitudinal studies should explore how changes in VAT influence medication needs.
Conclusion
VATob is associated with higher overall prescription use, but mainly with cardiovascular and metabolic agents. This association was higher for normal-weight participants than for overweight and obese participants.
Supporting information
S1 Table. STROBE statement—checklist of items that should be included in reports of case-control studies.
https://doi.org/10.1371/journal.pone.0318413.s001
(PDF)
Acknowledgments
The authors would like to express their gratitude to Mtro. Alfredo Avendaño Arenaza, Director de la Biblioteca Central Universitaria, and Mtro. Ricardo Villegas Tovar, Coordinator of Scientific Production and International Visibility, at the Benemérita Universidad Autónoma de Puebla. The authors would also like to thank the participants and staff of the National Health and Nutrition Examination Survey for their valuable contributions.
References
- 1.
Statista. Total number of medical prescriptions dispensed in the U.S. from 2009 to 2022 2022 Available from: https://www.statista.com/statistics/238702/us-total-medical-prescriptions-issued/#statisticContainer. [Accessed 12 Jun. 2023].
- 2.
Foundation PGP. How Have Prescription Drug Prices Changed Over Time? 2022 Available from: https://www.pgpf.org/blog/2022/04/how-have-prescription-drug-prices-changed-over-time. [Accessed 12 Jun. 2023].
- 3. Ho JY. Life Course Patterns of Prescription Drug Use in the United States. Demography. 2023; 60(5):1549–79. pmid:37728437.
- 4. Hoel RW, Giddings Connolly RM, Takahashi PY. Polypharmacy Management in Older Patients. Mayo Clin Proc. 2021; 96(1):242–56. pmid:33413822.
- 5. Kantor ED, Rehm CD, Haas JS, Chan AT, Giovannucci EL. Trends in Prescription Drug Use Among Adults in the United States From 1999–2012. JAMA. 2015; 314(17):1818–31. pmid:26529160.
- 6. Bluher M. Metabolically Healthy Obesity. Endocr Rev. 2020; 41(3):bnaa004. pmid:32128581.
- 7. Chooi YC, Ding C, Magkos F. The epidemiology of obesity. Metabolism. 2019; 92:6–10. pmid:30253139.
- 8. Hanley MJ, Abernethy DR, Greenblatt DJ. Effect of obesity on the pharmacokinetics of drugs in humans. Clin Pharmacokinet. 2010; 49(2):71–87. pmid:20067334.
- 9. Alser M, Naja K, Elrayess MA. Mechanisms of body fat distribution and gluteal-femoral fat protection against metabolic disorders. Front Nutr. 2024; 11:1368966. pmid:38590830.
- 10. Huang H, Zheng X, Wen X, Zhong J, Zhou Y, Xu L. Visceral fat correlates with insulin secretion and sensitivity independent of BMI and subcutaneous fat in Chinese with type 2 diabetes. Front Endocrinol. 2023; 14:1144834. pmid:36909323.
- 11. Ko YH, Wong TC, Hsu YY, Kuo KL, Yang SH. The Correlation Between Body Fat, Visceral Fat, and Nonalcoholic Fatty Liver Disease. Metab Syndr Relat Disord. 2017; 15(6):304–11. pmid:28481662.
- 12. Kurniawan LB, Bahrun U, Hatta M, Arif M. Body Mass, Total Body Fat Percentage, and Visceral Fat Level Predict Insulin Resistance Better Than Waist Circumference and Body Mass Index in Healthy Young Male Adults in Indonesia. J Clin Med. 2018; 7(5):96. pmid:29723977.
- 13. Sam S. Differential effect of subcutaneous abdominal and visceral adipose tissue on cardiometabolic risk. Horm Mol Biol Clin Investig. 2018; 33(1):20180014. pmid:29522417.
- 14. Mirza MS. Obesity, Visceral Fat, and NAFLD: Querying the Role of Adipokines in the Progression of Nonalcoholic Fatty Liver Disease. ISRN Gastroenterol. 2011; 2011:592404. pmid:21991518.
- 15. Tchernof A, Despres JP. Pathophysiology of human visceral obesity: an update. Physiol Rev. 2013; 93(1):359–404. pmid:23303913.
- 16. Abramowitz MK, Sharma D, Folkert VW. Hidden Obesity in Dialysis Patients: Clinical Implications. Semin Dial. 2016; 29(5):391–5. pmid:27144783.
- 17. Jensen MD. Visceral Fat: Culprit or Canary? Endocrinol Metab Clin North Am. 2020; 49(2):229–37. pmid:32418586.
- 18. Neeland IJ, Ross R, Després JP, Matsuzawa Y, Yamashita S, Shai I, et al. Visceral and ectopic fat, atherosclerosis, and cardiometabolic disease: a position statement. Lancet Diabetes Endocrinol. 2019; 7(9):715–25. pmid:31301983.
- 19. Elguezabal Rodelo RG, Porchia LM, Torres‐Rasgado E, López-Bayghen E, Gonzalez-Mejia ME. Visceral and subcutaneous abdominal fat is associated with non-alcoholic fatty liver disease while augmenting Metabolic Syndrome’s effect on non-alcoholic fatty liver disease: A cross-sectional study of NHANES 2017–2018. PLOS ONE. 2024; 19(2):e0298662. pmid:38394065.
- 20. Bray GA. Why do we need drugs to treat the patient with obesity? Obesity (Silver Spring). 2013; 21(5):893–9. pmid:23520198.
- 21.
Counterweight Project Team. The impact of obesity on drug prescribing in primary care. Br J Gen Pract. 2005; 55(519):743–9. PMID: 16212848.
- 22.
Centers for Disease Control and Prevention. National Health and Nutrition Examination Survey 1999–2022 (NHANES): U.S. Department of Health & Human Services; 2022 Available from: https://wwwn.cdc.gov/nchs/nhanes/continuousnhanes/default.aspx?BeginYear=2017. [Accessed 02 Mar. 2023].
- 23.
Centers for Disease Control and Prevention. NHANES NCHS Ethics Review Board Approval 1999–2022: U.S. Department of Health & Human Services; 2022 Available from: https://www.cdc.gov/nchs/nhanes/irba98.htm. [Accessed 02 Mar. 2023].
- 24.
Centers for Disease Control and Prevention. NHANES Procedure Manuals 2017–2018: U.S. Department of Health & Human Services; 2022 Available from: https://wwwn.cdc.gov/nchs/nhanes/continuousnhanes/manuals.aspx?Cycle=2017-2018. [Accessed 02 Mar. 2023].
- 25. Kim JA, Choi CJ, Yum KS. Cut-off values of visceral fat area and waist circumference: diagnostic criteria for abdominal obesity in a Korean population. J Korean Med Sci. 2006; 21(6):1048–53. pmid:17179685.
- 26. Keating XD, Zhou K, Liu X, Hodges M, Liu J, Guan J, et al. Reliability and Concurrent Validity of Global Physical Activity Questionnaire (GPAQ): A Systematic Review. Int J Environ Res Public Health. 2019; 16(21). pmid:31717742.
- 27.
Vidal Vademecum Spain. Vademecum: Vidal Group Drug Information Systeme,; 2010 Available from: https://www.vademecum.es/. [Accessed 01 May. 2023].
- 28. Kong M, Xu M, Zhou Y, Geng N, Lin N, Song W, et al. Assessing Visceral Obesity and Abdominal Adipose Tissue Distribution in Healthy Populations Based on Computed Tomography: A Large Multicenter Cross-Sectional Study. Front Nutr. 2022; 9:871697. pmid:35548570.
- 29. Franco LP, Morais CC, Cominetti C. Normal-weight obesity syndrome: diagnosis, prevalence, and clinical implications. Nutrition Reviews. 2016; 74(9):558–70. pmid:27473199.
- 30. Tatsumi Y, Nakao YM, Masuda I, Higashiyama A, Takegami M, Nishimura K, et al. Risk for metabolic diseases in normal weight individuals with visceral fat accumulation: a cross-sectional study in Japan. BMJ Open. 2017; 7(1):e013831. pmid:28093438.
- 31. Barrett LA, Xing A, Sheffler J, Steidley E, Adam TJ, Zhang R, et al. Assessing the use of prescription drugs and dietary supplements in obese respondents in the National Health and Nutrition Examination Survey. PLoS One. 2022; 17(6):e0269241. pmid:35657782.
- 32. Kalra S, Arora S, Kapoor N. Hidden Obesity. J Pak Med Assoc. 2023; 73(4):937–8. pmid:37052022.
- 33. Aparisi Gomez MP, Ponti F, Mercatelli D, Gasperini C, Napoli A, Battista G, et al. Correlation between DXA and laboratory parameters in normal weight, overweight, and obese patients. Nutrition. 2019; 61:143–50. pmid:30711863.
- 34. Powell-Wiley TM, Poirier P, Burke LE, Despres JP, Gordon-Larsen P, Lavie CJ, et al. Obesity and Cardiovascular Disease: A Scientific Statement From the American Heart Association. Circulation. 2021; 143(21):e984–e1010. pmid:33882682.
- 35. Bosello O, Vanzo A. Obesity paradox and aging. Eat Weight Disord. 2021; 26(1):27–35. pmid:31865598.
- 36. Ouchi N, Shibata R, Walsh K. Cardioprotection by adiponectin. Trends Cardiovasc Med. 2006; 16(5):141–6. pmid:16781946.
- 37. Dhawan D, Sharma S. Abdominal Obesity, Adipokines and Non-communicable Diseases. J Steroid Biochem Mol Biol. 2020; 203:105737. pmid:32818561.
- 38. Adamczak M, Wiecek A. The adipose tissue as an endocrine organ. Semin Nephrol. 2013; 33(1):2–13. pmid:23374889.
- 39. Cupido AJ, Asselbergs FW, Natarajan P, Group CIW, Ridker PM, Hovingh GK, et al. Dissecting the IL-6 pathway in cardiometabolic disease: A Mendelian randomization study on both IL6 and IL6R. Br J Clin Pharmacol. 2022; 88(6):2875–84. pmid:34931349.
- 40. Yao C, Narumiya S. Prostaglandin-cytokine crosstalk in chronic inflammation. Br J Pharmacol. 2019; 176(3):337–54. pmid:30381825.
- 41. Park HS, Park JY, Yu R. Relationship of obesity and visceral adiposity with serum concentrations of CRP, TNF-alpha and IL-6. Diabetes Res Clin Pract. 2005; 69(1):29–35. pmid:15955385.
- 42. Aitken M, Berndt ER, Cutler DM. Prescription drug spending trends in the United States: looking beyond the turning point. Health Aff (Millwood). 2009; 28(1):w151–60. pmid:19088102.
- 43. Gouju J, Legeay S. Pharmacokinetics of obese adults: Not only an increase in weight. Biomed Pharmacother. 2023; 166:115281. pmid:37573660.
- 44. Climent E, Benaiges D, Pedro-Botet J. Hydrophilic or Lipophilic Statins? Front Cardiovasc Med. 2021; 8:687585. pmid:34095267.
- 45. Koenen M, Hill MA, Cohen P, Sowers JR. Obesity, Adipose Tissue and Vascular Dysfunction. Circ Res. 2021; 128(7):951–68. pmid:33793327.
- 46. Thaete FL, Fuhrman CR, Oliver JH, Britton CA, Campbell WL, Feist JH, et al. Digital radiography and conventional imaging of the chest: a comparison of observer performance. AJR Am J Roentgenol. 1994; 162(3):575–81. pmid:8109499.
- 47. Domecq JP, Prutsky G, Leppin A, Sonbol MB, Altayar O, Undavalli C, et al. Clinical review: Drugs commonly associated with weight change: a systematic review and meta-analysis. J Clin Endocrinol Metab. 2015; 100(2):363–70. pmid:25590213.
- 48. Padula WV, Allen RR, Nair KV. Determining the cost of obesity and its common comorbidities from a commercial claims database. Clin Obes. 2014; 4(1):53–8. pmid:25425133.
- 49. Harrison S, Dixon P, Jones HE, Davies AR, Howe LD, Davies NM. Long-term cost-effectiveness of interventions for obesity: A mendelian randomisation study. PLoS Med. 2021; 18(8):e1003725. pmid:34449774.
- 50. Tremmel M, Gerdtham UG, Nilsson PM, Saha S. Economic Burden of Obesity: A Systematic Literature Review. Int J Environ Res Public Health. 2017; 14(4). pmid:28422077.
- 51. Chae J, Cho HJ, Yoon SH, Kim DS. The association between continuous polypharmacy and hospitalization, emergency department visits, and death in older adults: a nationwide large cohort study. Front Pharmacol. 2024; 15:1382990. pmid:39144630.
- 52. Castro-Barquero S, Casas R, Rimm EB, Tresserra-Rimbau A, Romaguera D, Martinez JA, et al. Loss of Visceral Fat is Associated with a Reduction in Inflammatory Status in Patients with Metabolic Syndrome. Mol Nutr Food Res. 2023; 67(4):e2200264. pmid:36416291.
- 53. Lira FS, Rosa JC, Dos Santos RV, Venancio DP, Carnier J, Sanches Pde L, et al. Visceral fat decreased by long-term interdisciplinary lifestyle therapy correlated positively with interleukin-6 and tumor necrosis factor-alpha and negatively with adiponectin levels in obese adolescents. Metabolism. 2011; 60(3):359–65. pmid:20359719.
- 54. Finelli C, Sommella L, Gioia S, La Sala N, Tarantino G. Should visceral fat be reduced to increase longevity? Ageing Res Rev. 2013; 12(4):996–1004. pmid:23764746.
- 55. Li H, Shi Q. Drugs and Diseases Interacting with Cigarette Smoking in US Prescription Drug Labelling. Clin Pharmacokinet. 2015; 54(5):493–501. pmid:25701380.