In New Zealand, as in many developed countries, the older adult population is the fastest growing age group, Statistics New Zealand, 2019, with many of these individuals making the shift to live in community dwelling retirement villages. While living in residential care (in rest home or respite care) is associated with higher sedentariness (Ministry of Health, 2013), there is a lack of research on the association between physical activity and health outcomes in individuals living independently in community-style retirement villages. Increasing our understanding of the health correlates of this population will go towards improving wellness and extending disability free life years in this population.

Given the damaging effects that cardiovascular decline can have on quality of life and independent living in older adults, a considerable amount of research has been conducted to determine the best means of delaying or reducing arterial stiffness. Chronological aging has been identified as one of the main clinical determinants of central artery stiffness and it is especially pronounced among adults aged 50+ (Tsai et al., 2017; Ceciliato et al., 2020) for both men and women (van der Heijden-Spek et al., 2000). Moreover, the PWV increase in people with hypertension has been found to be higher (0.93 m/s per decade, p < .001) than in normotensive people (0.44 m/s per decade) (Diaz et al., 2017).

Although it is impossible to stop the natural process of aging, there are factors that may help to slow this age-associated increase in arterial stiffness. A reduction in sedentary behaviour has been found by Ahmadi-Abhari et al. (2017) to be highly associated with a slowing of the natural progression of age-related stiffening of the arteries. In older populations prone to increased arterial stiffness (Wen et al., 2015), regular physical activity is an essential component of healthy aging (Chodzko-Zajko et al., 2009). To best support healthy aging, national health organizations, such as New Zealand’s Ministry of Health, recommend the inclusion of aerobic, resistance, and balance components to exercise routines targeted at older populations (Chodzko-Zajko et al., 2009; Ministry of Health, 2013). These recommendations are typically supported by current research, for example, habitual physical activity from mid-life to late-life is associated with lower central arterial stiffness and pulse pressure in older adults (Tanaka and Safar, 2005). Similarly, healthy active people with high maximal oxygen uptake, and few cardiovascular disease risk factors have considerably lower central pulse wave velocity than individuals who are sedentary with cardiovascular disease risk factors (7.0 ± 1.1 and 8.2 ± 1.4 m/s mean ± SD respectively) (Dobrin, 1983; Deiseroth et al., 2019).

The heterogeneity in aging and physical ability (from regularly and independently physically active to frail and inactive) makes it difficult to prioritize the exact types of physical activity, or combinations of physical activity components needed for optimal health benefits (Bangsbo et al., 2019). Regardless, some studies have attempted to tease out more guidance regarding different exercise modalities effects on arterial stiffness which would be important for the older population.

Regular aerobic exercise improves arterial stiffness significantly, with larger affects among those starting with stiffer arteries (PWV ≥8 m/s). However, resistance exercise had no effect on PWV nor did the combined aerobic plus resistance exercise (Ashor et al., 2014). While research has typically found resistance training to be ineffective in reducing arterial stiffness (Ceciliato et al., 2020), some have found that while high-intensity resistance training is unlikely to improve arterial stiffness, low-intensity resistance training or whole-body vibration training may be helpful in reducing carotid–femoral PWV (Jurik et al., 2021).

For older adults, balance is important and when balance is compromised their ability to live independently reduces. While studies investigating the effect of balance, or balance exercises are uncommon, there is some evidence to suggest that improving both core strength and balance is more effective in reducing arterial stiffness than resistance training (Kujawski et al., 2018), perhaps due to the strong association between physical activity levels, balance and core strength. Furthermore, older adults (over 65 years old) with higher PWV (>10 m/s) appear to have more difficulty in maintaining their balance when compared to those with lower PWV, even after controlling for age (Peultier-Celli et al., 2021). Again, it could be that better balance comes as a result of, at least in part, its association with higher physical activity levels.

Which health variables are most important for arterial health of older adults continues to be debated. Therefore, the aim of this study was to investigate the relationship between strength, balance, cardiovascular fitness, and physical activity levels to determine which health variables can predict arterial stiffness in an older population residing in community dwelling retirement villages.

There was a wide degree of variability between the participants in each of the villages, particularly regarding their age and PWV (Table 2). Pearson’s correlations were run for each of the health variables individually with PWV (Table 3).

Model A of the multivariate regression explained 16% of the variance and was a significant predictor of PWV, F (4, 75) = 3.52, p < 0.01, with an R ² of 0.16. The predicted effect on PWV was: 17.09 + 0.03 (Grip strength) - 0.79 (Balance) −0.09 (Sit-to-Stand) - 0.01 (6-min walk).

Model B explained 50% of the variance and was a significant predictor of PWV, F (11, 68) = 6.08, p < 0.001), with an R ² of 0.5. The predicted model in this regression analysis was: 1.05–0.73 (balance) - 0.09 (Sit-to-stand) - 0.02 (diastolic blood pressure, DBP) + 0.56 (Number of risk factors) +0.09 (age) + 0.06 (systolic blood pressure, SBP) + 0.02 (grip strength) and betas of +0.001 were found with the remaining variables, 6-min walk, total energy expenditure Kcal/week, total PA min/week, and BMI, suggesting small predictive contributions.

Model C explained 45% of the variance and was a significant predictor of PWV, F (3, 76) = 20.53, p < 0.001, with R ² = 0.45. The participant’s predicted PWV using this model is as follows: 3.71 + 0.59 (number of risk factors) + 0.11 (age) + 0.05 (SBP).

The aim of the present study was to explore the relative importance of strength, balance and physical activity on arterial stiffness in an older population. Our key findings indicate risk factors such as older age, higher SBP and a larger number of health risk factors were the strongest predictors of higher PWV. Moreover, higher physical activity levels (measured through self-report physical activity and 6-MWT) are better predictors of PWV than balance and hand-grip strength and sit-to-stand. Our findings are supported in previous research on males older than 50, but slightly younger than our participants. For example, among 54- to 75-year-old men, but with similar blood pressures, endurance-trained men showed significantly attenuated arterial stiffness indexes (25% lower), and Augmentation Index scores (36% lower) than their physically inactive counterparts (Vaitkevicius et al., 1993; Vlachopoulos et al., 2010). Our study suggests that engaging in regular aerobic exercise more strongly correlated with lower arterial stiffness and SBP than resistance training (as measured by hand-grip strength and sit-to-stand ability which are indications of muscular fitness). Nevertheless, a combination of the two types of exercises has been found in other studies to be advantageous to PWV and SBP. A recent study by Tardelli, et al. (2022) on spontaneously hypertensive rats (SHR) showed that after 8-weeks of combined aerobic and resistance training, arterial stiffness was reduced, resulting in a better autonomic balance to the heart and potentially explaining the lowering of blood pressure. SHR are inbred genetic rats with experimental hypertension created to mimic human essential hypertension for research trials (Gardner, et al., 2011)). Moreover, Shimojo et al. (2018) studied menopausal effects in ovariectomised SHR suggesting that combined aerobic and resistance training reduced mean arterial blood pressure by 12% compared to their physically inactive counterparts, heart rate by 8%, vascular sympathetic modulation by 40%, and improved overall baroreflex sensitivity. Perhaps the resistance training types (ladder adapted to rats, versus lifting weights or using exercise bands by older adults) played a role in these differences in findings.

The weak (sit-to-stand and balance tests) to moderate (total self-report physical activity and 6-MWT) relationship between these tests and arterial health is possibly due to the nature of the impact of the physical activity on the arterial walls. Tanaka and Safar (2005) have hypothesised that short-term physical activity (measured self-report physical activity volume) likely improves arterial stiffness via improvements in vasoconstrictor tone in smooth muscle tissue as well as enhanced nitric oxide bioavailability which regulates brachial artery elasticity. Conversely, long-term regular physical activity appears to prevent structural changes in the arterial walls that relate to the covalent crosslinking between long-lived proteins (such as collagen) and advanced glycation end-products (Takeuchi, 2020). While aerobic training introduced following the onset of cardiovascular disease can work to reduce arterial stiffness in older adults, the longevity of the improvements is unclear (Madden et al., 2013), which may explain the beneficial, but weak to moderate relationship between physical activity and PWV in our study. Moreover, our results indicated a weak correlation between PWV and the anaerobically-powered sit-to-stand test, whereas there was a moderate correlation between PWV and the aerobically-powered 6-MWT (See Table 3). However, a recent systematic review and meta-analysis demonstrated no significant correlation between PWV and high intensity interval training as well as moderate-intensity interval training (Way et al., 2019).

The strength of the relationship between a higher number of risk factors and increased PWV may help to explain the moderate relationship between overall physical activity and PWV (see Table 3). In this regard, findings by Madden et al. (2013) suggest that there is at least some irreversible arterial damage that occurs in individuals with multiple risk factors for arterial stiffness, such as advanced age, diabetes, hypertension or hypercholesterolemia. Therefore, in those participants with one or more risk factors for cardiovascular disease, the beneficial effect of physical activity on PWV is not strong enough to overcome the damaging effect of the risk factors on the arterial walls (and therefore off-set the beneficial relationship between physical activity and PWV).

Regarding the relative importance of each of the fitness criteria, our findings suggest that overall physical activity volume (self-report calculated to Kcal and MET min/week) and aerobic fitness (6-MWT) (see Table 3) are moderately related to PWV, but grip strength had no effect on PWV and sit-to-stand was only weakly compatible with PWV. These findings showed higher significance in previous meta-analyses by Zhang et al. (2018) as well as Ashor et al. (2014), indicating improvements in PWV are associated with aerobic exercise, and trended towards significance with combined aerobic and resistance training. These changes may be due to improvements in peak oxygen uptake (Deiseroth et al., 2019) and endothelial function (Zhang et al., 2018). Unlike the intervention studies reported in the meta-analyses above, our cross-sectional study provides evidence supporting the role of habitual daily physical activity in supporting arterial health in older people.

Finally, our study indicates that BMI is a poor predictor for arterial stiffness (see Table 3). While there seems to be a paradox regarding the arterio-protective nature of metabolically-healthy obesity on arterial stiffness (Yuan et al., 2020), particularly in the young-middle aged populations (Corden et al., 2013; Yuan et al., 2020), the absence of a relationship between BMI and arterial stiffness in our older cohort is likely due to the BMI measurement ( W e i g h t   k g h e i g h t m 2 ) itself. As height is static, changes in BMI would be due to the participant’s weight, which is influenced by changes in either the fat mass, or lean mass of the individual. An increase in fat mass is likely to be associated with an increase in arterial stiffness in middle-age to older populations (Benetos et al., 2009; Corden et al., 2013), while an increase in lean leg mass is associated with reduced arterial stiffness (Snijder et al., 2004). Therefore, an increase in BMI could either be beneficial or not depending on the reason for its elevation. Considering that BMI rises with increased fat mass or reduced muscle mass, it would be prudent to investigate age-related changes, such as sarcopenia. Low thigh cross sectional area (Huisman et al., 2015), and sarcopenia (Ochi et al., 2010) in older men have also been associated with increased arterial stiffness. Therefore, as BMI is unable to differentiate between fat and lean mass, it is perhaps not surprising that it was not associated with arterial stiffness in our study. Future research should measure body composition rather than BMI to elucidate the relationship between fat mass, muscle mass and PWV.

There are a number of limitations that need to be addressed in our study. As participants opted into the study, there may be bias towards more physically fit individuals and there were more women in the study. Each of the three villages recruited into the study had different participant profiles (See Table 2). However, as this was not an intervention study, and all data was pooled prior to analysis, the diversity of the participants in each of the retirement villages likely improved the ecological validity of our findings to the general population. This study took place over a year, therefore, participants were tested in different seasons. This seasonal change may have impacted participant’s levels of physical activity and that could have impacted arterial stiffness. The small sample size of this study limits the extent to which the findings can be applied to the older population in general and there were predominantly women. Some participants had CVDs and others did not. We tried to accounted for these individual health variations by creating the “number of (CVD) risk factors” variable (smoker, heart incidents, previous stroke, systolic blood pressure ≥140 mmHg, and/or taking anti-hypertension medication). We also collected a list of all the medications participants were taking. We controlled for participants taking anti-hypertension-specific medications (namely, ACE Inhibitors, ARBs and Beta-Blockers) as they directly impact PWV and they were among the most commonly taken in our sample. Stratifying the study population based on medication use, particularly corticosteroids and other relevant drug groups, allows for a more granular examination of their impact on arterial stiffness. The scope of this study, however, did not allow us to rationalise controlling for other medications as they were not frequent enough among participants, the could have complex interactions our study would not be able to detect, and the combinations varied too much from person to person. Body composition (BIA or DEXA) was not collected, nor was leg strength, which could have been informative variables to assess. Some participants had a light meal, i.e., were not fasted, prior to measurement due to medication needing to be taken with food. Although the recommendation is to have at least two measurements taken of cfPWV with a difference <0.5 m/s (Townsend et al., 2015), several PWV measures in our study took up to 50 min to get a quality-controlled tick, so in some instances, we were only able to collect one PWV measure. With participants scheduled back to back, we did the best we could to fit in two PWV measurements but in several cases, we had to settle for one due to the next participant arriving and limitations on room bookings.

Nevertheless, some of the limitations provide insight for future studies. For example, when measuring PWV using tonometry in older adults who may have larger necks and less elastic skin, it is important to allow for an additional 30 min to increase the likelihood of obtaining a second PWV measure. Also, in the absence of muscle mass bioimpedance, future studies might include measuring calf circumference in addition to grip strength for a muscle mass estimate for older adults (Quiñonez-Olivas, et al., 2016). It could have also been interesting to investigate the number of years since hypertension diagnosis.

Given the prevalence of comorbidities and polypharmacy (multiple medications taken simultaneously) in older adults, investigating potential synergistic or antagonistic effects between different drug classes is important to understand their broader interactions (Sinnott and Bradley, 2015). The inclusion criteria of designing such a study would need to control for the specific mix of drug classes taken, which was another limitation of this study. For example, the presence of diabetes for a long time is correlated with accelerated arterial stiffening indirectly through multiple mechanisms including, inflammation, endothelial dysfunction, elevated blood pressure, and changes in the structural components of the arterial walls (Petrie et al., 2018). By mitigating inflammation, corticosteroids might theoretically contribute to improved vascular health, making arterial stiffness a co-beneficiary of anti-diabetes medication (Laurent et al., 2006; Lunder et al., 2021). Polypharmacy can complicate the interpretation of associations, so careful consideration of medication interactions is warranted, particularly in the development of novel treatments (Salwe et al., 2016). Although it was beyond the scope of this study, future research should aim to delineate the specific mechanisms through which medications impact PWV and explore optimal therapeutic strategies that balance the benefits and potential risks associated with different drug classes (Janić et al., 2014).

In conclusion, this study reinforces the importance of habitual, lifelong physical activity in maintaining arterial health in an older population. However, the adverse effect of high systolic blood pressure, increasing age and overall numbers of cardiovascular risk factors had a stronger ability to predict arterial stiffness in an older population than fitness-based tests. Therefore, while engaging in physical activity later in life will have a beneficial effect on vascular health, regular physical activity throughout the lifespan, amongst other cardiovascular risk-reduction strategies, will have larger impact on pulse wave velocity in aging adults.