The relation of work-related factors with ambulatory blood pressure and nocturnal blood pressure dipping among aging workers

High blood pressure (BP) is one of the leading risk factors for premature death worldwide (GBD 2015 Risk Factors Collaborators 2016). In addition to traditional risk factors for hypertension (e.g., obesity and high salt intake) (GBD 2015 Risk Factors Collaborators 2016; He et al. 2013), work-related factors, such as job strain and shift work, have been found to associate with hypertension (Esquirol et al. 2011; Gilbert-Ouimet et al. 2014).

After the introduction of the job demands–control model of occupational stress by Karasek in the 1970s, there has been growing interest to study the relation of job strain (i.e., imbalance between job demands and job control) and susceptibility to illness. Job strain is known to associate with elevated office BP, but this association may be even stronger when BP is measured using 24-h ambulatory monitoring (Gilbert-Ouimet et al. 2014). Compared to office BP, ambulatory BP offers possibility to measure nocturnal BP, a particularly strong predictor of cardiovascular morbidity and mortality (Hermida et al. 2011; Kikuya et al. 2005). Furthermore, the nocturnal decline in BP (i.e., nocturnal BP dipping) has been shown to have incremental prognostic value on cardiovascular events independent of 24-h systolic BP, particularly among hypertensive patients (Salles et al. 2016). In previous studies, individuals with reduced nocturnal BP dipping (non-dippers) have been noted to have an increased risk for cardiovascular diseases compared to persons with normal nocturnal BP dipping (dippers) (Salles et al. 2016).

Prior studies on the association between job strain and ambulatory BP have discovered that higher job strain is positively associated with systolic BP and to a lesser extent with diastolic BP (Cesana et al. 1996; Clays et al. 2007; Fauvel et al. 2001, 2003; Landsbergis et al. 1994, 2003a, b; Maina et al. 2011; Schnall et al. 1992, 1998; Tobe et al. 2005, 2007). However, it is unclear how job demands and job control separately associate with ambulatory BP and especially with nocturnal BP dipping in aging workers (Landsbergis et al. 2013). Two previous studies have examined the relation of job strain and nocturnal BP dipping in middle-aged participants (Clays et al. 2011; Fan et al. 2013), and only one study (Mc Carthy et al. 2014) with job control as the only exposure variable has focused on nocturnal BP dipping in the older workers (aged 50–69 years).

Another important work characteristic in terms of 24-h BP is work-time mode. Shift work is associated with circadian misalignment (Douma and Gumz 2018) which can affect BP profiles (Morris et al. 2017). Indeed, it has been shown that shift work raises BP and shift workers are classified more often as non-dippers than day workers (Fialho et al. 2006; Kitamura et al. 2002; Lo et al. 2008; Munakata et al. 2001; Ohira et al. 2000; Yamasaki et al. 1998). Again, most of these studies included only individuals aged under 40 years or had a study sample of under 60 participants. It is, therefore, unclear whether the association between work-related factors and diurnal BP profile persists among aging workers who are at a considerably greater absolute risk of cardiovascular disease than younger individuals.

The aim of this study was to examine the association of work-related factors, including occupational status, work-time mode, job demands, and job control with ambulatory BP patterns among aging workers (age ≥ 55 years). We particularly focused on the association of work-related factors with nocturnal BP dipping (defined as BP decrease during sleeping time) which has been scarcely studied previously.

Characteristics of the study participants are shown in Table 1. The majority of participants were women (75%) and mean age was 62.2 ± 2.8 years. The mean awake BP was 128.7/77.8 mmHg, while the mean asleep BP was 110.8/63.9 mmHg. 50% of the participants had awake hypertension and 44% asleep hypertension. 21% were categorized as systolic non-dippers and 11% as diastolic non-dippers.

In our study, shift workers had significantly higher awake systolic BP than regular day workers, 136.5 mmHg (95% CI 131.9–141.1) vs. 132.5 mmHg (95% CI 128.3–136.6). The diastolic awake or asleep BPs did not differ significantly between these groups. We found no difference in awake or asleep BP values between manual and non-manual workers, workers with high and low job demands, or workers with high and low job control (Table 2).

Table 3 shows the nocturnal BP dipping in categories by work-related factors. Diastolic nocturnal BP dipping was significantly greater in shift workers than in regular day workers (19% vs. 17%, p = 0.03). Systolic nocturnal BP dipping was significantly lower in participants with high job demands than in those with low job demands (14% vs. 16%, p = 0.04). In contrast, we did not find any significant associations between occupational status, job control, and nocturnal BP dipping values.

Table 4 shows the nocturnal BP dipping results from a sensitivity analysis after excluding 52 participants on antihypertensive medication. In these analyses, we observed no significant differences in BP dipping values between groups categorized by work-related factors.

In this study, we examined the association of occupational status, work-time mode, job demands and job control with ambulatory BP among aging workers. We focused particularly on the relation of work-related factors and nocturnal BP dipping. We observed that individuals who performed shift work had significantly greater systolic awake BP and diastolic BP dipping than regular day-time workers. In addition, participants with high job demands had significantly lower systolic BP dipping than participants with low job demands. Job control and occupational status did not differentiate ambulatory BP profiles in aging workers.

To our knowledge, this is one of the first studies to study BP dipping among aging workers in this scale. We observed that shift workers demonstrated a significantly greater awake systolic BP and diastolic BP dipping compared to regular day-time workers. Unlike in our study, shift work has been related to an unfavorable dipping profile in three of four previous studies, (Fialho et al. 2006; Kitamura et al. 2002; Lo et al. 2008; Yamasaki et al. 1998). However, these studies included mainly younger individuals and only one study was performed using a single 24-h BP monitoring (Yamasaki et al. 1998), whereas the others examined the dipping profile changes in participants working in rotating shifts. In the study that was performed on middle-aged nurses untreated for hypertension, participants working in the evening or night shifts were more often systolic BP non-dippers than day-shift workers (Yamasaki et al. 1998). This difference was explained mainly by the higher systolic sleep BP. In our study, the differences in BP dipping values were explained mainly by the shift workers’ higher awake BP. These differences between studies may be partly explained by occupational differences because the participants of the previously mentioned study were nurses whereas our participants had a wide array of occupations. One major difference is also that, in contrast to the study with a sample consisting of nurses, only few of shift workers performed night shifts in our study. It has been earlier reported that circadian misalignment is the mechanism underlying the higher BP in shift workers (Morris et al. 2017). In addition, shift workers sleep less than day workers, express higher anger-in scores and, therefore, have probably more sympathetic nerve activity (Ohira et al. 2000). Therefore, it is possible that shift workers in our study had all these features, but most importantly, most of them were sleeping in the night in contrast to night-shift workers. Possibly, our shift workers had increased day BP because of stress but the relaxing effect of sleep was adequate for them as they had no circadian misalignment. In addition, our participants were also older and some were using antihypertensive medication. The exclusion of the participants with antihypertensive medication improved the nocturnal BP dipping profiles in general and abolished the significant difference between the regular day-time workers and shift workers. This can be partly explained by reduced statistical power. In addition, this could also reflect the fact that antihypertensive medication usually affects day-time BP values more than night-time values due to the more common dosing of the medications in the morning (Mancia and Parati 2004).

Studies on the relation of job demands and BP profiles have been extremely limited. In our study, we observed that workers with high job demands demonstrated a significantly smaller nocturnal systolic BP dipping than workers with low job demands. No differences in awake or asleep BP values were observed between these two groups. Although decreased nocturnal dipping has been related to adverse cardiovascular risk in previous studies (Salles et al. 2016), the clinical significance of this finding in individuals with high job demands remains unclear and requires further study in a prospective setting.

One earlier study has reported that low job control is related to higher asleep systolic BP and decreased nocturnal systolic BP dipping in younger hypertensive men (Fan et al. 2013). In older workers, low control, measured by possibility for influence and development at work, has also been associated with higher asleep systolic BP, but not with systolic nocturnal BP dipping (Mc Carthy et al. 2014). In our study, the differences in awake and asleep BP values and systolic nocturnal BP dipping between the low and high control groups remained small, even after performing a sensitivity analysis without participants on hypertensive medication. Two prior studies have also examined the association between job strain and ambulatory BP (Brown et al. 2006; Landsbergis et al. 2003b) regarding the occupational status. These studies reported that particularly workers performing manual labor may be more exposed to high BP induced by job strain. Our findings on the relation between job strain and ambulatory BP in aging workers, however, were limited.

The strengths of our study include having a relatively large number of participants with numerous occupational backgrounds, compared to previous smaller studies. Most of the previous studies considering work-time mode were also often performed at single workplaces in which psychosocial exposures may have a limited variance. Moreover, we included people with and without hypertension. The results of our study are, therefore, more generalizable to the aging general population as older workers commonly use antihypertensive medication. The current study has also some limitations which warrant discussion. First, although aging workers have been studied less in this context, they are also a distinct group of employees who may be somewhat healthier than the general population. This is reflected by the relatively low mean awake BP of the study sample (129/78 mmHg). It is also important to note that the proportion of non-dippers in our study was low (systolic/diastolic: 21/11%), which could partly explain the lack of significant differences between the groups. Furthermore, the awake BP values used in our study were the means of the whole awake period which could result in smaller between-group differences compared with mean BPs that would have been calculated separately for work- and free-time. We used shortened version of the Job Content Questionnaire although it has been shown to correlate with the complete scales (Fransson et al. 2012). Moreover, only a few of shift workers were performing night shifts. Besides, further analysis on sleep quality during the ambulatory monitoring would have helped to elucidate the impact of sleep quality on ambulatory BP but these data are unavailable in both study cohorts. In addition, as a cross-sectional study, this study can not offer information on the causal association between work-related factors and ambulatory BP.

We conclude that these data provide evidence that shift work and high job demands are associated with ambulatory BP profile. Although shift work was related to a higher awake BP in our study, this increased cardiovascular risk may be offset by the fact that these individuals demonstrated greater BP dipping. Furthermore, hypertensive individuals with high job demands might benefit from 24-h ambulatory BP measurement as our findings suggest that they may have an increased risk of being non-dippers. Future prospective studies are needed to understand causality between these factors.