1. Introduction
Sedentary behavior (SB) refers to any activity performed while sitting, reclining, or lying down, resulting in an energy expenditure of less than or equal to 1.5 metabolic equivalents, and sitting time is a commonly considered outcome for SB measurement (Healy et al., 2012; Tremblay et al., 2017). With evidence showing a negative impact of high SB levels on health outcomes, including all-cause and cardiovascular disease mortality, cancer risk, musculoskeletal diseases, decreases in cognitive function, and depression, amongst others (Park et al., 2020; Saunders et al., 2020), several health guidelines (e.g., WHO guidelines) have been providing evidence-based recommendations to limit sedentary time or replace this behavior with physical activity of any intensity. Due to the increasing risk of all-cause mortality, when daily sitting time is higher than 9 hours, previous evidence has suggested that the optimal amount of daily sitting in adults should be under 7.5 hours (Ku et al., 2018; 2019). Nevertheless, there are currently no established threshold for SB (Bull et al., 2020; Dempsey, 2020).
Because office environments are places of exposure to high levels of sitting time, workers have been shown to spend most of their working hours in prolonged SB (Healy et al., 2012; Clemes et al., 2014; Hadgraft, 2016). According to Hadgraft and colleagues, office workers spend, on average, 79% of their working hours in a sitting position, mostly in prolonged bouts (≥30 minutes) (Hadgraft, 2016). Similarly, Daneshmandi et al. (2017) showed that, within an 8-hour workday, office workers spend 81% of their working time in sitting positions. While investigating the impact of SB on health outcomes, the authors also demonstrated that prolonged sitting was directly related to exhaustion during the workday, hypertension, and musculoskeletal disorders in the shoulders, lower back, thighs, and knees.
Particularly in office environments, sit-stand desks (SSD) – i.e., desks or work surfaces that can be adjusted to allow height adjustment for both sitting and standing work (Grunseit & Chau, 2013), represent a potential strategy to reduce sitting time effectively (Shrestha & Dunn, 2020; Zhou et al., 2023). While investigating the impact of a SSD intervention on the general health of 74 desk workers, a recent investigation found significant benefits in short-term decreased sitting time and subjective health scores (Ma et al., 2021).
Phase angle (PhA) is a clinical parameter, considered a potential marker of inflammation and oxidative stress, and suggested as an indicator of cellular health, that measures the angular shift between voltage and current sinusoidal waveforms, which is an index of cell membrane integrity and vitality, and it is measured through bioelectrical impedance analysis (BIA) (Vincenzo, 2021; Da Silva et al., 2023). A higher value of PhA (between 5 and 7 in healthy populations) indicates greater cellularity, cell membrane integrity, and cellular function (Vincenzo, 2021). However, cell mass and membrane integrity depend on age, sex, fluid distribution, and body mass index (BMI) (Gonzalez et al., 2016).
Prolonged physical inactivity negatively affects tissue electrical properties, resulting in prominent decreases in PhA values (Norman, 2012). Lin et al. (2024) has suggested that the negative association found between SB and PhA may be related to the negative effects of prolonged sitting time on oxidative stress and inflammation, which results in cellular structures damaging and impaired cellular functions, hence the variations observed in PhA values. Although few investigations have been conducted analyzing the relationship between SB and PhA (Asano et al., 2023; Ahmadi et al., 2024), none has evaluated the impact of an SSD-based intervention and reduction in sitting time on office workers’ PhA. Therefore, this study aims to investigate the alterations in PhA resulting from an SSD intervention for office workers compared to a waiting list control group.
2. Methods
2.1. Settings and participants
The methods used in the SUFHA (Standing Up for Healthy Aging) two-arm crossover randomized controlled trial (FAZER+/ILIND/CIDEFES/1/2022, https://doi.org/10.17605/OSF.IO/JHGPW) have been described in detail elsewhere (Júdice et al., 2023). The study was granted ethical approval by the University Lusófona committee (D0522) and was conducted following the Declaration of Helsinki for Human Studies. Briefly, office workers were recruited through university advertisements. This study involved a 6-month SSD intervention and a 3-month follow-up.
Eligible participants were full-time office workers over 20 years old and spent at least 70% of their working week on desk-related activities. Participants were then randomly assigned to an intervention or waiting-list control group using an online program (www.random.org) by a researcher not involved in recruitment or data collection. The randomization process was designed to ensure that participants in both groups were matched in crucial variables, such as time spent sitting, number of sit-to-stand transitions, and BMI. An initial session with both groups was conducted to explain the benefits of reducing sitting time and interrupting it with standing time. The participants from the intervention group were also given verbal and written instructions on how to use the SSD, information on maintaining a correct ergonomic posture, and individualized guidance on increasing standing time progressively. Additionally, motivational prompts were emailed and available on the SUFHA website during the intervention to support relatedness, perceived competence, and autonomy in using the desk. The prompts were varied weekly in the first month, fortnightly in the second and third months, and monthly until the end of the intervention.
2.2. Control group
The waiting-list control group attended the initial psychoeducational session about the independent benefits of reducing and interrupting sitting time with standing, like the intervention group, but had no contextual change across the 6-month intervention (i.e., no access to SSD or prompts). After the intervention, this group was given access to the SSD.
2.3. Assessments
Baseline (pre-intervention) and 6-month (post-intervention) assessments were conducted. The assessment procedures were equal in both moments and are described in detail in a previous publication (Júdice et al., 2023).
2.4. Demographical data
Participants provided self-reported information on their age, sex, country of birth, education, financial status, occupation, duties, working hours, years of work, and the existence of chronic diseases or conditions through an online questionnaire.
2.5. Bioelectrical impedance analysis
Primary resistance (R) and reactance (Xc) BIA data were analyzed to assess PhA using a phase-sensitive single-frequency bioimpedance analyzer (BIA 101 BIVAPRO, Akern, Florence, Italy) (Campa et al., 2021; Júdice et al., 2023). A frequency of 50 KHz and an alternating current of 250 mA were passed through the distal electrode of each pair, and the voltage drop across the body was measured using the proximal electrode (Campa et al., 2021). The BIA analyzer measurements were calibrated before each test using a precision circuit, and the calibration was deemed successful if the R-value was 383 Ω and Xc was 46 Ω (Francisco et al., 2023; Matias et al., 2016). The BIA analysis was performed with participants in an 8-hour fasting condition, with an empty bladder, lying supine for a minimum of 5 minutes, with arms and legs abducted, and without metallic objects. Four electrodes were placed on the right side of the participant’s body in standardized points (Ballarin et al., 2024; Genton et al., 2018).
2.6. Body composition
The participants’ BMI was assessed by weighing them to the nearest 0.01 kg on an electronic scale (Tanita BC-601), wearing minimal clothes and without shoes. Height was measured using a stadiometer (Seca, Hamburg, Germany) to the nearest 0.1 cm. BMI was calculated as body mass (kg)/height (m2) (Júdice et al., 2023). Using the raw BIA data, the fat-free mass was estimated using previously validated equations (Segal et al., 1988) (i.e., specific for women and men). Then, body fat mass was calculated as total body mass minus fat-free mass.
2.7. Physical behaviors
Physical behaviors such as sitting and standing time were assessed using an inclinometer/accelerometer (ActivPAL4; PAL Technologies, Ltd., Glasgow, UK) and the CREA (v1.3) classification algorithm, which allows determining non-wear, primary lying, secondary lying, sitting, standing, stepping, cycling, and sit/stand transitions (Júdice et al., 2015). Participants wore the device on their right thigh 24 hours a day for seven days without taking it off while maintaining their daily activities (Júdice et al., 2023).
2.8. Statistical analysis
2.8.1. Power calculations
Cluster size calculations were performed with the assumption that participants within the same workplace would be independent, so an intra-cluster correlation of 0.01 was used. Considering sitting time (main outcome), an effect size of 0.80, a power of 0.80, and a significance level of 0.05 (two-tailed), the total calculated sample size was 34 participants.
2.8.2. Data analysis
The data were analyzed using IBM SPSS software (version 28.0, 2024). First, descriptive analysis was conducted on all continuous variables and presented in mean ± standard deviation. Frequency analysis was performed for non-continuous variables. A repeated measure analysis of variance adjusting for covariates (ANCOVA) was used to analyze differences between groups from baseline to 6 months and group-by-time interaction, controlling for potential confounders (clustering, age, sex, and BMI). Paired-sample T-tests were performed to check for time differences within each group, and independent sample T-tests were used to assess differences between groups at baseline and 6 months. Statistical significance was set with a p-value less than 0.05 for all tests.
3. Results
3.1. Participant characteristics
The CONSORT diagram depicting the recruitment and participant process was published elsewhere (Júdice et al., 2024). Initially, 39 office workers were randomly assigned to the intervention or control group. One participant was removed from the analysis due to pregnancy, resulting in 38 office workers being included. The sample consisted of 29 women and nine men with an average age of 43.8 ± 8.0 years. All participants worked full-time and, on average, spent 474.0 ± 78.8 min/day at work. Based on the independent sample T-test, no significant differences were found between groups for any variable at baseline, including age, weight, and BMI. More information about the sample can be found in Table 1.
Table 1
Demographic variables, phase angle, and ActivPAL outcomes at baseline.
| DEMOGRAPHICS | CONTROL (n = 19) | INTERVENTION (n = 19) |
|---|---|---|
| Age (years) | 42.3 ± 9.5 | 45.3 ± 6.0 |
| % Female participants | 73.3% (n = 14) | 78.9% (n = 15) |
| Full-time workers | 19 | 19 |
| Years of service | 14.6 ± 11.5 | 15.2 ± 10.1 |
| Daily working (min/day) | 467.4 ± 79.0 | 480.0 ± 80.3 |
| % of participants with chronic disease | 21.1% (n = 4) | 20.0% (n = 4) |
| Biometric assessments | ||
| Height (m) | 1.64 ± 0.1 | 1.63 ± 0.1 |
| Body mass (kg) | 82.3 ± 21.1 | 72.0 ± 16.5 |
| Body mass index (kg/m2) | 30.5 ± 7.4 | 27.2 ± 6.1 |
| Body fat (%) | 38.3 ± 8.6 | 36.5 + 6.9 |
| Phase angle (º) | 5.3 ± 0.7 | 5.3 ± 0.7 |
| Resistance (R) | 544.2 ± 88.6 | 553.4 ± 72.1 |
| Reactance (Xc) | 50.2 ± 8.8 | 51.0 ± 7.2 |
| ActivPAL outcomes | ||
| Sitting time (min/day) | 475.9 ± 130.4 | 469.0 ± 70.5 |
| Prolonged sitting (min/day)* | 266.7 ± 118.7 | 276.1 ± 93.8 |
| Standing time (min/day) | 248.4 ± 82.2 | 274.2 ± 92.5 |
| Stepping time (min/day) | 81.5 ± 23.3 | 99.9 ± 27.3 |
| Number of sit to stand transitions | 45.1 ± 12.2 | 44.9 ± 11.2 |
| Number of valid days | 6.0 ± 0.0 | 6.0 ± 0.0 |
[i] Notes: *Prolonged sitting, ≥ 30 min.
3.2. Bioelectrical impedance analysis
At baseline, both groups had an average PhA of 5.3 ± 0.7. After adjusting for clustering, age, sex, and BMI, both groups marginally increased PhA from baseline to 6 months (Δ ≈ 0.15). No group-by-time interaction (p = 0.729) or differences between groups at baseline (p = 0.964) and 6 months (p = 0.816) were found. Additionally, no group-by-time interaction was found for R (p = 0.204) or Xc (p = 0.559), neither differences between groups at baseline (p = 0.727) and 6 months (p = 0.404) for R, nor differences between groups for Xc values at baseline (p = 0.748) and 6 months (p = 0.470). Figure 1 displays the PhA, R, and Xc levels at baseline and post-intervention.
Figure 1
Change in Phase Angle from baseline to 6 months in both groups.
Also, participants from both groups showed high inter-individual variability in R, Xc values, and PhA scores (Figure 2).
Figure 2
Inter-individual changes among group participants on resistance (R), reactance (Xc), and Phase Angle (PhA).
3.3. Body composition
Although both groups experienced a significant reduction (p < 0.05) in fat mass percentage between baseline and post-intervention assessments (intervention, 36.5 ± 7.2 vs. 32.9 ± 6.3, 3.6% fat mass reduction; control, 38.3 ± 8.6 vs. 36.7 ± 8.8, 1.6% reduction), no group-by-time interaction was found (p = 0.089). Additionally, no differences between groups were found at baseline (p = 0.503) or post-intervention (p = 0.135).
3.4. Physical behaviors (ActivPAL)
Adherence to ActivPAL was 100%, and there were no data losses in the data collected. The intervention group significantly decreased prolonged sitting, and the control group did not (–25.5 vs –1.26 min; p < 0.05), although no time*group interactions were found (p = 0.236). Non-significant improvements were found in all the other physical behaviors favoring the intervention group.
4. Discussion
This study analyzed the impact of a SSD-based intervention coupled with motivational prompts on the PhA of office workers using data from SUFHA (Júdice et al., 2023). It is the first randomized controlled trial investigating the effects of changes in sitting time on PhA among office workers. PhA is an essential indicator of cellular health, and higher scores are associated with better integrity and functionality of the cell membrane (Norman, 2012; Vincenzo, 2021). PhA values can help identify musculoskeletal issues such as skeletal injury, sarcopenia, frailty, and weight loss (Lukaski & Garcia-Almeida, 2023). It is also valuable in clinical settings (Lukaski & Garcia-Almeida, 2023). Studies have shown that lower PhA levels are associated with increased inflammation and oxidative damage (Da Silva et al., 2023). A survey conducted by Sebastian et al. (2022) found that SB was associated with higher inflammatory and oxidative marker levels in 44 participants. These findings suggest that monitoring PhA through BIA could be a practical, non-invasive way to assess the potential benefits of reducing SB and improving overall health.
No differences between the two groups nor group-by-time effects were found in PhA, R, or Xc values. The non-existence of significant differences between groups may be mainly related to the lack of significant post-intervention changes in body composition between groups (Cancello et al., 2023). The marginal increases in PhA post-intervention may be related to the improvements in the participant’s body composition, specifically the reduction in fat mass, which led to a decrease in the R values (Martins et al., 2021). When observing BIA-derived R, Xc, and PhA values at an individual level, high interindividual variability was found among participants in both groups. However, it is possible to find a more consistent pattern of improvement in the PhA of the intervention group, while the control group presented a more considerable individual heterogeneity. These findings are relevant to highlight that averaging the values of the entire group does not allow for the seeing of these differences. Thus, future studies on PhA should be aware of this individual heterogeneity in response to an intervention.
Delving deeper into the potential mechanisms of SB effects on PhA, it is plausible to suggest that reducing SB may be as important as increasing physical activity levels, considering that prolonged SB can potentially modulate muscular cell activity, contributing to muscle atrophy and loss of function, accompanied by a reduced energy production due to decreases of mitochondrial function (Chen et al., 2024; Martins et al., 2021; Mundstock et al., 2019; Raffin et al., 2023). Thus, longer SB may result in more extensive functional deterioration (Raffin et al., 2023). Also, SB is negatively associated with cardiorespiratory fitness, and prolonged sitting time reduces overall skeletal muscle activity (Prince et al., 2024). Since PhA scores are related to cardiorespiratory fitness and muscle strength, reducing SB may mediate PhA scores (Martins et al., 2021; Prince et al., 2024). Considering the ecological approach in our intervention (i.e., no daily prompts or weekly reminders as performed in previous interventions) and the lower impact of such intervention, another factor that may explain the lack of significant changes in PhA can be related to the length (i.e., 6 months), which may not be enough to impact cell integrity and therefore PhA. Future interventions may consider longer durations to explore if reducing SB may impact PhA in SSD interventions.
Another potential mechanism is related to body composition. It has been previously shown that inflammation and oxidative stress, from which PhA is a potential marker, are linked to obesity (Cancello et al., 2023; Da Silva et al., 2023). Since it has been suggested that higher SB may lead to higher adiposity, reducing it may lead to improved body composition and better PhA scores, but only if replaced by physical activity, given that reducing weight or fat mass just by itself may not have the desired effects on PhA, if lean mass is not preserved or even increased (Cancello et al., 2023; Edwardson et al., 2022; Schnurr et al., 2021). Nonetheless, since PhA is directly related to muscle strength and cardiorespiratory fitness, which depend on augmented levels of physical activity and not necessarily reduced SB, it suggests that reducing sitting time and replacing it with physical activity may be important (Prince et al., 2024; Raffin et al., 2023). Lastly, sleep quality is another mechanism to be further investigated that may explain the influence of SB in PhA scores.
Our findings suggest that simply reducing SB may not improve PhA scores. Exercise and physical activity have been found to impact PhA positively, so it may be beneficial to substitute SB with active behaviors other than standing in interventions aiming to improve office workers’ PhA scores (Mundstock et al., 2019; Sardinha & Rosa, 2023). These suggestions are supported by Lin et al. (2024) that reported better PhA scores in older adults with higher activity levels compared to those with lower. One possible reason for the lack of effect of the SSD intervention conducted in this study on PhA scores, may be explained by the fact that, although SSD coupled with motivational prompts are effective interventions to reduce SB, they do not necessarily influence physical activity levels. Still, given the potential mechanisms of SB identified, it is plausible to assume that reducing sitting time may have a protective effect on cellular health and contribute to reduce the decline of PhA scores. Future studies analyzing if longer or more intense SB reduction interventions, focusing on promoting higher alteration in sitting time may be conducted to further explore the effects on PhA to confirm or contradict this possibility. Also, given the benefits of replacing SB for light intensity physical activity, investigating the effects of SSD interventions compared with walking desks on PhA scores may be interesting to further analyze this theory (Buffey et al., 2022). Additionally, considering the findings in these studies, future investigations using interventions to improve the PhA of office workers may add strategies to raise activity levels during working hours while decreasing SB. Applying portable pedal machines or treadmill desks may increase total workday physical activity while reducing SB (Nasir et al., 2024; Thompson et al., 2014). Additionally, it may be relevant to combine SB reduction interventions with nutritional interventions since the pilot study of Barrea et al. (2017) has positively linked a healthy diet to better PhA scores, independent of gender, age, and BMI.
One interesting finding from our study is that the intervention and control groups showed similar improvements in PhA scores, with no differences between groups. This result may be related to the slightly higher stepping time increase observed in the control group, raising the question of whether walking or cycling desks may be more effective at improving PhA scores than SSD in the office setting, as they could increase energy expenditure and metabolic rates in addition to reducing sitting time (Oye-Somefun et al., 2021). Based on the SUFHA findings for other health-related outcomes, the reduction in sitting time was enough to impact psychological consequences such as overall fatigue, the need to recover after work, or musculoskeletal discomfort (Júdice et al., 2024). Still, the analyses in this study suggest that for significant changes to occur in PhA, the standing and sitting time changes may need to be higher in magnitude or potentially need more time.
Another noteworthy finding from our study is that despite being middle-aged, our participants had PhA values corresponding to those of healthy elderly subjects (Mattiello et al., 2020). This finding raises concerns about the potential harmful effects of high SB and low activity levels, as long-term exposure to excessive sitting time may be responsible for these lower PhA values. More studies are needed to investigate the relationship between activity/SB patterns and PhA levels. More randomized controlled studies in different contexts and populations, considering PhA as the primary outcome, are necessary to deepen the understanding of the effects of SB reduction on this variable. Also, it might be interesting to compare the effects of interventions focused on reducing SB and increasing physical activity levels (i.e., replacing SB for active time) in comparison with interventions focusing only on improving one behavior to analyze their impact on PhA and the magnitude of differences they have on this variable.
4.1. Limitations
Despite the novelty factor of this study, some limitations need to be addressed. The sample size was mainly composed of women, which makes it challenging to compare the impact of SB in PhA between genders. This is particularly relevant since evidence has previously shown that the amount and type of SB in adult populations differs between genders (Prince et al., 2020). Therefore, studies involving a more even number of participants of both genders are necessary to further understand potential differences in the impact of SSD interventions and SB reduction in PhA between men and women. Also, it is essential to conduct studies that include different ethnic groups to analyze if there is a difference in the impact between different ethnic backgrounds since it has been previously found that differences in SB patterns exist between different cultural and ethnic backgrounds (Biddle et al., 2019). One limitation of this study is that it was not possible to measure the muscle mass of the participants, which is strongly associated with PhA (Martins et al., 2021). Further research is needed to understand how varying muscle mass levels can mediate SB’s effects on PhA. Finally, the fact that the intervention group received two gathered strategies (i.e., SSD and motivational prompts) which the control group did not receive, this fact prevents us from concluding about the sole impact of SSD on SB reduction and PhA. Future interventions aiming to determine the impact of SSD on PhA may be designed in such way that can allow this differentiation to be performed.
4.2. Practical implications
The findings of this study contribute to the investigation of the impact of SB on cellular health, specifically in PhA, an indicator of cell integrity and function. The lack of significance in the results suggests that only reducing SB may not improve PhA, and it may be essential to increase activity levels. However, reducing SB may slow PhA decreases. Therefore, practitioners may consider replacing SB with active work (e.g., treadmill desks and portable pedaling devices) to achieve maximum benefits and results.
5. Conclusion
This six-month SSD intervention coupled with motivational prompts failed to increase the PhA of office workers. However, the marginal improvements suggest that SB interventions may need to integrate other strategies to increase office workers’ activity levels to improve PhA levels. When analyzing the mechanisms that influence PhA scores, one can state that, while it is essential to increase activity levels, reducing SB may prevent PhA decreases. Interestingly, our middle-aged sedentary office workers presented PhA levels comparable to the reference values of elderly individuals, which is concerning and potentially suggests that long-term exposure to SB patterns may lead to low PhA scores. Further studies are needed to understand the link between SB and PhA and explore combined interventions that aim to replace SB with active alternatives. For a more comprehensive analysis, future studies should present inter-individual variability data and integrate a more gender-balanced sample to enable a deeper understanding of the heterogeneous response levels to interventions, as well as explore longer interventions, that may be necessary for changes in PhA to occur.
Data Accessibility Statement
Anonymized trial data will be available for non-commercial research purposes only upon request to the PI.
Declaration of Generative AI and AI-assisted technologies in the writing process
While preparing this work, the authors utilized Grammarly’s AI-powered writing assistance to enhance the manuscript’s language and readability. Subsequently, the authors reviewed and edited the resulting content as necessary, taking full responsibility for the publication’s final version.
Ethics and Consent
The study has received approval from the ethics committee of the Universidade Lusófona, with approval number D0522. It was conducted following the Declaration of Helsinki for Human Studies.
Before entering the trial, each participant was given a detailed oral and written study description. Participants had the opportunity to ask questions before providing written informed consent. All participants signed an informed consent form.
Acknowledgements
We want to thank the participants for their time and effort. This work was supported by the ILIND (FAZER+/ILIND/CIDEFES/1/2022).
Competing Interests
The authors have no competing interests to declare.
Author Contributions
PBJ conceptualized the study. PBJ, HS, SCT, and GBR developed the methodology. PGFR wrote the original draft, and PGFR and PBJ reviewed and edited the manuscript. All authors have read and agreed to the published version of the manuscript.