Stand Up for Your Health: Rationale and Design for a Randomized Controlled Trial

Abstract

Background:

Sedentary behavior is an independent risk factor for cardiometabolic disease. With the rise of desk-based occupations and work-from-home culture, the workplace offers a promising setting to employ interventions to reduce sedentary time.

Objective:

The goal of this randomized controlled trial is to determine if reducing sedentary time with sit-stand desks improves cardiometabolic health outcomes in overweight and obese individuals with sedentary jobs who are at risk of developing diabetes.

Methods:

Approximately 198 eligible participants (includes estimated 25% drop-out) with pre-diabetes or at-risk for type 2 diabetes will be randomized in a 1:1:1 ratio into three groups: a control group (no intervention), a 2-hour group (instructed to stand at their sit-stand desks for at least 2 hours daily), and a 3-hour group (instructed to stand for at least 3 hours daily). Each participant will complete three study visits over a 6-month period. The primary outcome is insulin resistance.

Discussion:

The potential impact of this study is significant, given that over 70% of the U.S. population is overweight or obese, and more than 80% of jobs are sedentary. Affirmative findings from our study would advance the field by demonstrating that an easily adoptable intervention can reduce cardiometabolic risk, providing justification for widespread implementation of standing desks in the workplace.

Conclusion:

This randomized, control study will determine if reducing sedentary behavior at work, with an adjustable sit-stand desk, improves insulin resistance in individuals at risk for diabetes. This study may help inform public health guidelines benefitting a large population of sedentary workers.

Introduction

Type 2 diabetes mellitus (T2DM) and cardiovascular disease (CVD) continue to be pressing public health issues. According to the Centers for Disease Control and Prevention, about 10% of U.S. adults have T2DM, while another 98 million - around one in three - have prediabetes, a condition that often progresses to T2DM within five years if left untreated.^1-4 If the trend continues, it is predicted that as many as one in every three to five adults in the United States could have diabetes by 2050.⁵ Likewise, CVD remain the leading cause of death in the United States and globally.^6,7 While physical inactivity and elevated body mass index (BMI) are well-established risk factors for diabetes and CVD, recent research suggests sedentary behavior to be an independent risk factor for cardiometabolic diseases.^7-9 This emerging evidence highlights the potential for interventions aimed at reducing sedentary time to improve cardiometabolic health, independent of physical activity levels.

Sedentary behavior is defined as any activity involving low levels of energy expenditure (≤1.5 metabolic equivalents [METs]).¹⁰ This includes behaviors such as sitting, watching TV, reading, and driving. It is distinct from physical inactivity, which refers to not meeting national guideline recommendations for physical activity– defined as at least 150 minutes of moderate physical activity or 75 minutes of vigorous intensity physical activity per week, or an equivalent combination.^10-12 The independent role of sedentary behavior as a risk factor for CVD and diabetes has gained considerable attention.⁹ The long-term health consequences related to sedentary behavior are distinct from and additive to lack of physical activity. In a report from the Women’s health initiative, the hazard ratio for CVD was 1.18 (95% CI 1.09, 1.29) (P for trend < 0.001) in women who reported sitting for more than ten hours per day compared to women who sat less than five hours per day, independent of overall physical activity levels.¹³ Experimental interventions administering five days of bed rest in otherwise 20 healthy participants led to increases in insulin resistance and cholesterol levels, further supporting a link between sedentary time and cardiometabolic adverse effects.¹⁴ A prior study showed that the negative impact of six hours of sitting on fitness was similar to the beneficial effect of one hour of moderate intensity exercise suggesting that one can still suffer the health consequences of sedentary behavior despite meeting the physical activity guidelines.¹⁵

This brings into consideration whether replacing sedentary time with low-intensity activities of daily living, such as standing, can lead to meaningful improvements in cardiometabolic health. A meta-analysis including 24 studies and 3,169 participants did not demonstrate significant changes in metabolic outcomes, including blood pressure and total cholesterol though there was improvement in high-density lipoprotein cholesterol.¹⁶ The included studies did not evaluate glucoregulatory outcomes. More recently, the RESET-BP and Rise for health studies implemented multipronged interventions (coaching, wrist-worn activity prompter, and a sit-stand desk) to reduce sedentary behavior in office workers.^17,18 While there was no significant change in blood pressure in RESET-BP, there was a small improvement in diastolic blood pressure in Rise for Health which enrolled older (post-menopausal) subjects compared to younger subjects in RESET-BP. In exploratory analyses, increased sit-stand transitions reduced fasting insulin levels by 20% only among those with prediabetes or diabetes. A major limitation of these studies is the multipronged intervention that includes individual coaching, behavior monitoring, goal setting, and personalized feedback – all of which would be costly to implement on a larger scale.¹⁹ Therefore, our goal was to focus on a simple workplace intervention, sit-stand desks, that if effective, could more easily be implemented on a larger scale.

In a pilot study conducted by our team involving 15 overweight or obese individuals, we observed improvements in insulin resistance, triglyceride levels, and superficial femoral artery (SFA) vascular endothelial function following 6 months of standing desk use, which was independent of changes in physical activity levels or BMI.²⁰ Here, we propose a fully-powered clinical trial and hypothesize that adults at-risk for diabetes will demonstrate improvements in insulin sensitivity, metabolic and vascular endothelial health with a sit-stand desk intervention at work. The primary aim of our randomized controlled clinical trial is to assess the impact of reducing workplace sedentary time on insulin resistance and ascertain if a dose-response relationship exists. Secondary outcomes include metabolic syndrome and severity scores and measures of vascular endothelial function. Direct measurement of vascular endothelial function, assessed by flow-mediated dilation (FMD), is a powerful tool in translational science. FMD strongly predicts cardiovascular events in patients with and without CVD,²¹ and endothelial dysfunction can be reversed by interventions known to reduce CVD risk.^22,23 Therefore, FMD will be an important surrogate endpoint for this study.

Methods:

Subject Recruitment:

All subject related activities are performed at the Medical College of Wisconsin (MCW) with approval by the MCW Institutional Review Board. The study is currently in the recruitment phase and is expected to complete enrollment by early 2027. To recruit potential participants, we use both printed and electronic advertisements. Flyers are posted around the offices and cubicle areas at the three main recruitment sites: MCW, Froedtert Hospital, and Marquette University. Advertisements are also sent out electronically on MCW’s intranet webpage, bulletins, and newsletters, and in emails to employees at all three recruitment sites. If needed, we will use Facebook advertisements to enroll other eligible individuals outside of these three workplaces.

Inclusion and Exclusion Criteria

The inclusion criteria for the study are as follows: age 18-79; overweight or obese (BMI of ≥ 25 or BMI ≥23 for participants identifying as Asians; including Japanese, Chinese, Filipino, Vietnamese, Korean, and South Asian); and a sedentary job which we defined as spending at least 75% of their workday sitting at a desk, working at least 4 days a week. Additionally, participants must be at risk of developing diabetes either due to having (1) a diagnosis of prediabetes (glycosylated hemoglobin [HbA1c] of 5.7% to 6.4% or a fasting glucose of 100 to 125 mg/dL) or (2) one or more additional risk factors (beyond elevated BMI) as defined by the National Institute of Diabetes and Digestive and Kidney Diseases ²⁴ including age 45 years or older, family history of T2DM in a first degree relative, no structured exercise activity, personal history of gestational diabetes, hypertension (HTN), or dyslipidemia, or belonging to one of the following races or ethnicities: African American, Alaska Native, American Indian, Asian American, Hispanic/Latino, Native Hawaiian, Pacific Islander.

Exclusion criteria include known history of clinical CVD including myocardial infarction (MI), coronary stent, coronary artery bypass grafting (CABG), cardiac transplant, angina, congestive heart failure (CHF) or cardiomyopathy, and peripheral vascular disease. Additional exclusion criteria include: diabetes mellitus, chronic musculoskeletal disorders involving lower extremities such as arthritis of the knees or hips that impact the ability to use a standing desk, regular use of ambulatory assistive devices, neuropathy of any origin, positional syncope, orthostasis, tobacco use within previous 12 months, current use of a standing or sit-stand desk at work, excessive alcohol use (defined as more than 14/drinks per week for women and more than 28 drinks per week for men), steroid use for more than 21 days per year, fasting triglyceride level ≥500, thyroid disease (unless thyroid medication dose has been stable for the last 12 months), uncontrolled HTN (>150/90 mm Hg), pregnant or lactating (or plans to become pregnant in the next 6 months), low density lipoprotein (LDL) cholesterol ≥190, currently enrolled in or plans to enroll in a diet or weight loss program, and current use or plans to use weight loss medications.

General Study Design and Power Analysis

We are enrolling 198 overweight or obese participants at-risk for T2DM, in a randomized controlled study of 6 months duration. Participants are randomized to one of three groups (control, 2 hours, and 3 hours of standing during work hours) in a 1:1:1 allocation. Participants in the 2-hour arm are instructed to use the desk in the standing position for a minimum of 2 hours/day while those in the 3-hour arm subjects are instructed to use the desk in the standing position for at least 3 hours/day. The control arm receives no intervention. We selected a target of at least two hours per day based on our preliminary pilot data, which showed substantial improvements in insulin resistance and FMD among participants who reduced sitting time at work by an average of 90 minutes per day.²⁰ The 2-hour target is also supported by an expert consensus statement from United Kingdom, which recommends initially accumulating two hours per day of standing and light activity during working hours, eventually progressing to a total of four hours per day.²⁵

Sex-based randomization schemes are generated by the PLAN procedure in SAS^® 9.4 (SAS Institute Inc., Cary NC). Our primary outcome measurement is insulin resistance, which precedes abnormal glucose tolerance (prediabetes) and T2DM.²⁶ We will assess insulin resistance using Homeostatic Model Assessment for Insulin Resistance (HOMA-IR) based on fasting insulin and glucose measurements. Targeting this early stage of insulin resistance provides an opportunity to prevent the development of diabetes and its associated complications. Secondary outcomes of interest include fasting glucose, fasting insulin, HbA1c, fasting triglyceride levels, metabolic syndrome severity scores ²⁷, and FMD of the brachial artery (BA) and SFA. Changes in fasting serum free fatty acids (FFAs) and their correlation with changes in endothelial function are exploratory outcomes.

Accounting for at least a 25% dropout rate, the sample size of 198 randomized participants (132 randomized to intervention vs. 66 randomized to control) provides greater than 80% power for the primary efficacy endpoint for a medium effect size (Cohen's d = 0.5) to detect a 0.75 (log transformed value: −0.2877) absolute mean difference²⁰. This calculation is based on data from our pilot study.²⁰ Meanwhile, a sample size of 66 each for the 2h- and 3h-standing group is sufficient for statistical analysis. The power for the primary endpoint is calculated based on a two-sided t-test with a significance level of 5%.

Screening & Enrollment

Potential participants contact study coordinators using the contact information provided in the advertisement. The study team then emails them a summary of the study protocol for their review. If they indicate a willingness to participate, the consent form is emailed to them, and a screening phone call is scheduled. During this call, the subject is assessed for eligibility by reviewing the inclusion/exclusion criteria. If determined eligible, the screener completes a medical history questionnaire by asking the potential participants specific questions to elicit their demographics, detailed medical history, current medications, and social history. The consent form is also discussed in detail during this call, and any questions or concerns are addressed prior to enrollment. If the participants remain interested, the first (baseline) of the three study visits is scheduled. Since the study visits and the intervention occur during working hours, participants are required to have permission from their supervisor to participate in the study.

Blinding

To ensure rigor of the study, all study personnel involved in data assessment remain fully blinded, including the Principal Investigator and biostatisticians. The study participants remain unblinded due to the nature of the intervention. The study coordinator handling randomization and overseeing study visits need to be unblinded. The unblinded study coordinator will have no role in data analysis, collection, or outcomes assessments.

Study Visits

The participants undergo three in-person study visits over the duration of 6 months: a baseline visit, a 3-month visit, and a 6-month visit, Figure 1. All the in-person study visits are conducted at the Adult Translational Unit (ATRU) of the Clinical and Translational Science Institute (CTSI) of Southeast Wisconsin. Participants are directed to fast for at least eight hours prior to each of the three visits and instructed to withhold their morning doses of vasoactive medications since these can affect the FMD measurements.²¹

Figure 1. General flow chart of study activities and randomization scheme.

* Deliver/install sit-stand desk and anti-fatigue mat at place of work; provide instructions on use

IPAQ: International Physical Activity Questionnaire

OSPAQ: Occupational Sitting and Physical Activity Questionnaire

FMD: flow-mediated dilation

DXA: dual-energy X-ray absorptiometry

Written informed consent is obtained during the first visit prior to initiating any research related activities. Once the informed consent is signed; participants’ height and weight are checked to calculate their BMI. Participants with a BMI below the cut-off are excluded, while those above the cut-off are randomized.

The following metrics are obtained during all three visits: vital signs (i.e. heart rate, blood pressure, and oxygen saturation), height, weight, BMI, waist circumference measured 1 cm above the iliac crest, and hip circumference measured at the widest circumference of the buttocks at the level of the greater trochanter. Blood is drawn in fasting state at each visit to measure glucose, HbA1c, insulin, lipid panel, and free fatty acids. HOMA-IR is calculated using fasting glucose and insulin levels. An oral glucose tolerance test (OGTT) is administered with 1-hour and 2-hour fingerstick glucose checks after a 75-gram glucose drink. BA and SFA FMDs are performed before and after (BA only) completion of OGTT. Additionally, a urine pregnancy test is administered to all premenopausal female participants of potential childbearing age at each visit unless they have not had menses for more than 12 months without alternative medical cause or are surgically sterile.

A dual-energy X-ray absorptiometry (DXA) scan using the iDXA (General Electric, Lunar Medical Systems, Madison, Wisconsin) is completed during the first and the 6-month visits to measure the total body composition. The DXA scan is analyzed using General Electric’s enCORE software to calculate the total and regional fat mass, adiposity distribution, and lean mass.

Participants are asked to complete four surveys during each of the three visits. Occupational Sitting and Physical Activity Questionnaire (OSPAQ) is a well validated tool to assess percentage distribution of occupational sitting, standing, walking and heavy labor time.²⁸ International Physical Activity Questionnaire (IPAQ) is used to assess the types of intensity of physical activity and sitting time of participants.²⁹ Dietary assessment questionnaire is administered through ATRU bio-nutrition core lab and consists of a validated 24-hour dietary recall assessment that includes 2 weekdays and 1 weekend day. Participants also report their weekly average of steps per day for each of the preceding 12 weeks.

During the 6-month visit, the participants who were randomized to the intervention group complete an additional desk acceptability questionnaire, consisting of a 10-item five-point Likert scale (1 strongly disagree, 2 disagree, 3 neutral, 4 agree, 5 strongly agree) with questions on ease of use, comfort, perceived work productivity, and musculoskeletal discomforts related to use of the sit-stand desk.

For all participants, at the end of each visit, an accelerometer device (activPal) enclosed in a nitrile sleeve is attached to the participant’s anterior thigh using a Tegaderm. The device is worn for subsequent seven days during a regular work week and provides objective assessment of activity levels and sedentary times. Participants are allowed to shower with the dressing but are instructed to avoid bathing or swimming. The control participants receive a complementary adjustable sit-stand desk at the end of the 6-month study participation.

Flow Mediated Dilation

FMD assessments are completed through the MCW ATRU Vascular Core lab and performed by trained personnel as previously described.²⁰ For a given study, the same sonographer performs the FMD for any subject at all study visits to minimize intertester variability. Endothelial function is assessed with BA and SFA FMD before and after (BA only) the 2-hour OGTT during each visit. FMD is used to non-invasively assess macrovascular endothelial function using high-resolution ultrasound technology (7.5-13 MHz probe).²¹ The arterial diameter is measured at baseline and then again after a 5-minute period of arterial occlusion mediated by constriction from a blood pressure cuff (>200 mm Hg systolic or at least 60 mm Hg above resting systolic pressure). Upon rapid deflation of the cuff after that period of arterial occlusion, the transient hyperemia caused by rapid blood flow stimulates nitric oxide production which results in arterial dilation.³⁰ FMD is measured as the change in pre- to post-diameter and is expressed as a percentage of the baseline diameter (FMD%). The protocol has been previously described in detail before.³⁰ To minimize inter-tester variability, the same sonographer performs the FMD across all three study visits for each participant. During each FMD, participants are asked to rate their pain/discomfort on a scale of 1-10 to account for any pain that may affect FMD measurements.

Sit-Stand Desk Intervention

A non-electric, pre-assembled, height adjustable sit-stand desk (Vari^®) is delivered to the participants in the 2-hour and 3-hour groups within 2 weeks of randomization. Two separate desks are provided if the participant has a hybrid work schedule. The workstation is installed on top of the regular desks, is easy to move up and down, and can hold up to 2 monitors, a full-sized keyboard, and mouse. An anti-fatigue mat is provided to minimize back and leg discomfort that could result from prolonged standing. Participants in the control group also receive a desk after they complete their 6 months of participation.

We send weekly emails to participants in both the 2-hour and 3-hour intervention groups reminding them to use the desk in the standing position for at least 2 or 3 cumulative hours per day, depending on their intervention group. Participants do not have to work in the standing position continuously for 2 or 3 hours and are allowed to distribute their assigned hours throughout their workday.

Accelerometer

We use the activPal^™ accelerometer (PAL Technologies^™, Glasgow, UK) to quantify sitting and standing time, sedentary bouts, daily steps, cadence (walking speed in steps/minute), up/down transitions (“interrupted” sitting) and exercise activity – and distinguish between activities occurring at work verses outside of work. The activPAL is a research-grade accelerometer which has been validated in studies to accurately distinguish between sedentary and standing postures.^31,32 The device is enclosed in a nitrile glove and attached to the anterior thigh using a Tegaderm. The accelerometer is worn for 7 days during a typical work week (not a vacation week). Participants are instructed to avoid bathing, swimming or saunas while wearing the accelerometer, but they are allowed to take a shower. A prepaid envelope is provided to the participants to return the device.

Statistical Analysis Plans

All the primary, secondary, and exploratory analyses will be based on the intent-to-treat analysis set (ITT) which comprises all randomized participants who will be analyzed according to the assigned group. The per-protocol set (PP), consisting of all participants from the ITT without significant protocol violations, will be used for sensitivity analyses for the primary endpoint.

All analyses will be performed using SAS^® 9.4. All statistical tests will be two-sided at the 0.05 level of significance except where noted.

The descriptive statistics will be provided for each time point (baseline, 3-/6-month), which will be conducted for 1) all outcomes, including HOMA-IR, HbA1c, fasting glucose, fasting insulin, metabolic syndrome (MetS) criteria, MetS severity z-score, SFA and BA FMD, fasting triglycerides, very low density lipoproteins (VLDL), and free fatty acids; 2) mean sedentary time and standing time at work, daily sedentary bouts; 3) demographics and occupation, weight, BMI, systolic and diastolic blood pressures, waist and hip circumferences; 4) dietary intake, including unsaturated and saturated and trans fats, refined and unrefined carbohydrates, protein, omega-3 fatty acids, alcohol, and sugars; 5) physical activity data, including mean daily steps, moderate-to-vigorous intensity physical activity (MVPA), and number of up/down transitions (“interrupted” sitting). For all the continuous variables, such as age and BMI, the mean and standard deviation will be presented. For all the categorical variables, such as sex and race, the frequency and percentage will be provided.

In the primary analysis, the null hypothesis of no difference in the log-transformed HOMA-IR between the intervention and control arm at 6 months will be tested using a two-sample t-test. We will employ the longitudinal general linear mixed-effects model approach for each primary and secondary outcome at baseline and 3-/6-month, which includes time (baseline, 3-/6-month) and intervention (sit-stand desk vs. control) as fixed effects and a random intercept for individual variation if needed. To test the hypotheses of whether the intervention significantly improves each outcome, we will use the likelihood ratio test for the intervention effect. Additionally, the general linear mixed-effects model approach will be utilized to detect the possible dose-response relationship of insulin resistance as measured by HOMA-IR change over time to compare these changes between three arms.

To account for the possible compliance variance of the actual standing desk time, we will conduct an exploratory analysis by performing the general linear mixed-effects model for the primary outcome HOMA-IR, with time, actual standing time and intervention as fixed effects. To explore possible impacts from confounding factors, we will fit a multivariate general linear mixed-effects model for each outcome.

Discussion

This study aims to investigate the effects of reducing workplace sedentary time on cardiometabolic health and vascular endothelial function in overweight or obese individuals at risk of developing diabetes, while also exploring the underlying mechanisms. Average medical expenditure among diabetics is more than twice as high as for people without diabetes accounting for about $413 billion in 2022.³³ Targeting individuals at risk for diabetes, rather than those with established diabetes, provides a critical window for prevention, helping to avoid adverse outcomes and reduce the associated healthcare burden. Additionally, as reported by Katzmarzyk et al., the adverse effects of prolonged sitting on mortality are more pronounced in overweight or obese individuals, suggesting that this population may derive the greatest benefit from interventions aimed at reducing sedentary time.³⁴ The potential impact of our study in the general population is significant, given that over 70% of the U.S. population is overweight or obese, and more than 80% of jobs are sedentary.^35,36 Affirmative findings from our study would advance the field by demonstrating that a simple, easily adoptable workplace intervention aimed at reducing sitting time can favorably reduce cardiometabolic risk providing justification for the widespread implementation of standing desks in the workplace.

Our study addresses several critical knowledge gaps in the field. First, although there is robust evidence supporting the role of MVPA in preventing cardiometabolic diseases and most public health guidelines provide specific recommendations regarding its type and duration, there is lack of comparable evidence to inform such recommendations related to sedentary behavior.⁹ By exploring the dose–response relationship between sedentary behavior and health outcomes, our study could contribute valuable evidence toward the development of such public health guidelines. Second, as noted in American Heart Association Scientific Advisory, further research is needed to understand the mechanisms linking sedentary behavior to adverse cardiometabolic outcomes.⁹ Meta-analyses show that sedentary behavior is a mediator of T2DM.³⁷ For example, reductions in sedentary time, even with light-intensity walking or standing, stimulate skeletal muscle contraction. This triggers the translocation of glucose transporter type 4 (GLUT4), a glucose transporter, to the muscle cell surface, increasing glucose uptake thereby improving glucose metabolism and insulin sensitivity.³⁸ Mendelian randomization studies support a causal relationship between triglyceride-rich proteins and coronary heart disease with lower triglycerides levels associated with regression of coronary artery plaques.³⁹ Our pilot study demonstrated improvements in insulin resistance, triglyceride levels and SFA FMD, offering preliminary insight into the pathways through which sedentary behavior may influence cardiometabolic health.²⁰ The findings from the proposed, fully-powered clinical trial could contribute valuable evidence toward understanding theses mechanisms.

The strengths of our study include a large sample size and the inclusion of both 2-hour and 3-hour intervention groups, which will allow us to evaluate dose response relationships. Additionally, given the potential discrepancy between self-reported and actual activity levels, the use of accelerometers in our study will enable more accurate assessment of participant’s physical activity levels and compliance with the desk use.^40,41 A key limitation of our study is the inability to blind participants, which is not possible due to the nature of the intervention. However, we will ensure that all study personnel involved in data assessment will remain fully blinded. Another limitation of our study is that we did not include a behavioral intervention, which has been shown to enhance the effects of physical interventions in reducing sitting time.⁴² However, a single, simplified intervention would be effortless to implement on a large scale across the different employers comprising the 80% of U.S. workers that have sedentary occupations.

Conclusion

This randomized, control study will determine if reducing sedentary behavior at work, with an adjustable sit-stand desk, improves insulin resistance in individuals with pre-diabetes or at risk for diabetes. By providing quantitative measure of sit-stand desk usage, this study may help inform public health guidelines benefitting a large population of sedentary workers.