Weekly hours of informal caregiving and paid work, and the risk of cardiovascular disease

Weekly hours of informal caregiving and paid work, and the risk of cardiovascular disease Abstract Background Little is known on the association between weekly hours of informal caregiving and risk of cardiovascular disease (CVD). The objective was to investigate the individual and joint effects of weekly hours of informal caregiving and paid work on the risk of CVD. Methods Pooled analysis with 1396 informal caregivers in gainful employment, from the Swedish Longitudinal Occupational Survey of Health and the Whitehall II study. Informal caregiving was defined as care for an aged or disabled relative. The outcome was CVD during 10 years follow-up. Analyzes were adjusted for age, sex, children, marital status and occupational grade. Results There were 59 cases of CVD. Providing care >20 h weekly were associated with a higher risk of CVD compared to those providing care 1–8 h weekly (hazard ratio = 2.63, 95%CI: 1.20; 5.76), irrespectively of weekly work hours. In sensitivity analyzes, we found this risk to be markedly higher among long-term caregivers (6.17, 95%CI: 1.73; 22.1) compared to short-term caregivers (0.89, 95%CI: 0.10; 8.08). Caregivers working ≥55 h weekly were at higher risk of CVD (2.23, 95%CI: 1.14; 4.35) compared to those working 35–40 h weekly. Those providing care >8 h and working ≤40 h weekly had a higher risk of CVD compared to those providing care 1–8 h and working ≤40 h (3.23, 95%CI: 1.25; 8.37). Conclusion A high number of weekly hours of informal caregiving as opposed to few weekly hours is associated with a higher risk of CVD, irrespectively of weekly work hours. The excess risk seemed to be driven by those providing care over long periods of time. Introduction Informal caregiving, defined as unpaid care for elderly or disabled family members, is common in working age.1–3 Helping those in need is meaningful to most people, and informal caregiving saves the health-care system considerable expenses in terms of relieving formal caregiving services.4 However, informal caregiving may lead to an emotional and physical burden for the caregiver.3,5 Furthermore, chronic stress from caregiving6 may lead to a dysregulation of the cardiovascular system via a hyper activation of the hypothalamic-pituitary-adrenal axis.7 It has, therefore, been hypothesized that informal caregiving may increase the risk of cardiovascular disease (CVD).8–10 In order to initiate preventive strategies to reduce health consequences of informal caregiving, it becomes important to unveil certain conditions under which caregiving becomes a stress burden with consequences for health. Caregiving hours may be used as an indicator of intensity, but only few studies have addressed the relationship between caregiving hours and risk of CVD. A study based on the US Nurses’ Health Study showed that women providing spousal care for ≥9 hours weekly had a higher risk of CHD compared to non-caregivers.11 Additionally, results indicate that weekly hours of caregiving had a greater impact on risk of CHD compared to perceived stress from caregiving.11 However, Nurses’ Health Study only included female nurses, which may limit generalizability to the general working population. In another study among British civil servants, an association between informal caregiving and CHD was found only for caregivers in poor health.12 In an additional analysis, they found no increased risk of CHD in those providing ≤5 h or >5 h of weekly care compared to non-caregivers. However, dividing at 5 h weekly may not capture those with high intensity caregiving, i.e. many weekly hours. The previous literature is limited by having non-caregivers as the reference, since being a non-caregiver does not include the mental burden of caregiving responsibilities, which caregivers have to face regardless of whether they provide caregiving for one or 30 h weekly. By only including caregivers, we will be able to thoroughly investigate the contribution of weekly hours of caregiving. This is highly relevant in terms of designing support systems aimed at relieving some of the caregiving burden. A further limitation in previous research on caregiving and CVD risk relates to the double burden from informal caregiving and paid work. Work stress and long working hours are independent risk factors for CHD13–15 and it is important to recognize that work and family life are strongly intertwined and should not be treated separately in relation to health consequences.16–18 In support of this, several studies have shown that spillover between work and family life may be associated with low self-reported health.19–22 Presently, we sought to investigate the individual and joint effects of weekly hours of informal caregiving and paid work on the risk of CVD, in informal caregivers in gainful employment. Methods This is a multi-cohort study pooling individual level data from the Swedish Longitudinal Occupational Survey of Health (SLOSH)23 and the Whitehall II study from the UK.24 SLOSH is an open cohort and comprises 40 877 individuals, representative of the Swedish workforce. Whitehall II includes 10 308 British civil servants from 20 London-based departments. Year 2008 was the baseline in SLOSH (response rate 61%), and years 1991–93 were the baseline in Whitehall II (response rate 87%). The total study population encompassed 1396 informal caregivers in gainful employment (Supplementary Appendix A). The cohorts has been approved by the appropriate ethics committees and participants have given consent to be included. Weekly hours of informal caregiving and paid work Informal caregiving was in both cohorts assessed by asking whether participants provided care for an aged or disabled relative. We included only those who answered ‘yes.’ We categorized weekly hours of caregiving in three groups; 1–8 h (reference), 9–20 h and >20 h.11 In Whitehall II, we had additional information on long-term caregiving, distinguishing those who provided informal care 2–3 years following baseline (long-term caregivers) from those provided care at baseline only (short-term caregivers).9,12 For weekly work hours, individuals in SLOSH were asked how many hours they worked each week, and in Whitehall II individuals were asked how many hours they worked on a normal workday. To harmonize this information, we multiplied daily work hours in Whitehall II by five, based on the assumption that individuals in the civil service work 5 days a week. Weekly work hours were categorized as <35 h, 35–40 h (reference), 41–54 h and ≥55 h per week.13,25 We investigated the joint effect of weekly hours of informal caregiving and paid work more thoroughly by dividing participants into four mutually exclusive groups: (1) Those providing care 1–8 h and working ≤40 h weekly, (2) providing care 1–8 h and working >40 h weekly, (3) providing care >8 h and working ≤40 h weekly and (4) providing care >8 h and working >40 h weekly. Ascertainment of CVD We used register-based information on diagnosis of CVD defined according to the 10th edition of the International Classification of Disease (ICD).26 CVD encompassed diagnoses CHD (ICD-10 codes I20–I25) and stroke (ICD-10 codes I60–I63). Individuals with a CVD diagnosis at baseline were excluded from the study. Covariates Confounding and mediating variables related to CVD were identified based on prior knowledge and using the method of directed acyclic graphs (Supplementary Appendix B). Confounding variables included age, sex, children (yes, no), marital status (yes, no) and occupational grade (low wage and manual laborers, lower non-manual and midlevel managers, and upper non-manual workers and managing staff). Smoking and BMI measured at baseline, were most likely mediating factors on the causal pathway from workload of caregiving and work and risk of CVD,27 and were only adjusted for in a sensitivity analysis. Statistical analyzes We applied a Cox regression model with time from baseline as the underlying time axis. The outcome was time until CVD, death from other causes or the end of follow-up, whichever came first; with 6 years follow-up in SLOSH and 10 years follow-up in Whitehall II. The Cox model was implemented in a random-effect meta-analysis and cohort specific estimates were combined and weighted in a joint model.28 The degree of heterogeneity between cohorts was assessed using I2, which describes the percentage of total variation across studies due to heterogeneity.29 We considered I2 values of 25%, 50% and 75% to represent low, moderate and high heterogeneity, respectively.29 We applied a random effects model, because there may be between-group variation in SLOSH and Whitehall II on the investigated associations, due to different follow-up time and welfare-state differences. Analyzes showed no effect modification by sex (Supplementary Appendix C4). We applied two analytical models: model 1 was adjusted for age and sex, and model 2 was also adjusted for children, marital status and occupational grade; analyzes of caregiving hours were further adjusted for work hours and those of work hours were adjusted for caregiving hours. Interaction was assessed on a multiplicative and additive scale according to recommendations by the STROBE guidelines.30 We assessed multiplicative interaction by including a product term between caregiving and work hour variables in the Cox model. Additive interaction was assessed with the synergy index,31 which represents the additional risk from exposure to informal caregiving and work hours when interaction is present, relative to the risk from exposure when there is no interaction. In supplemental analyzes, we investigated risk of CVD for long-term caregivers compared to short-term caregivers. Here, we excluded the first 3 years of follow-up to minimize risk of reverse causation bias. Results Among the 1396 informal caregivers, 59 participants (4%) developed CVD within 10 410 person-years at risk; 50 cases of CHD and 9 cases of stroke. In SLOSH, there were 14 (2%) cases during 6 years follow-up and in Whitehall II, there were 45 (6%) cases during 10 years follow-up. Baseline characteristics of participants are shown in table 1. Table 1 Baseline characteristics of informal caregivers in SLOSH and Whitehall II   SLOSH  Whitehall II  Total  1–8 h care ≤40 h work  1–8 h care >40 h work  >8 h care ≤40 h work  >8 h care >40 h work  Informal caregivers  N = 700  N = 696  N = 1396  N = 504  N = 496  N = 262  N = 134  Women  68%  39%  54%  56%  41%  71%  58%  Age (Mean ± SD)  52 ± 9  50 ± 5  51 ± 7  52 ± 7  51 ± 7  50 ± 9  51 ± 7  Children  48%  63%  56%  55%  59%  52%  51%  Low occupational grade  12%  35%  23%  19%  34%  11%  22%  Smoking  19%  15%  17%  13%  14%  23%  25%  Obese  21%  11%  16%  14%  15%  18%  25%  Weekly caregiving hours                1–8 h  69%  74%  72%  100%  100%  ––  ––  9–20 h  17%  16%  16%  ––  ––  55%  61%  >20 h  14%  10%  12%  ––  ––  45%  39%  Weekly work hours                <35 h  21%  4%  13%  20%  ––  29%  ––  35–40 h  33%  52%  42%  80%  ––  71%  ––  41–54 h  34%  35%  34%  ––  77%  ––  75%  ≥55 h  12%  9%  11%  ––  23%  ––  25%    SLOSH  Whitehall II  Total  1–8 h care ≤40 h work  1–8 h care >40 h work  >8 h care ≤40 h work  >8 h care >40 h work  Informal caregivers  N = 700  N = 696  N = 1396  N = 504  N = 496  N = 262  N = 134  Women  68%  39%  54%  56%  41%  71%  58%  Age (Mean ± SD)  52 ± 9  50 ± 5  51 ± 7  52 ± 7  51 ± 7  50 ± 9  51 ± 7  Children  48%  63%  56%  55%  59%  52%  51%  Low occupational grade  12%  35%  23%  19%  34%  11%  22%  Smoking  19%  15%  17%  13%  14%  23%  25%  Obese  21%  11%  16%  14%  15%  18%  25%  Weekly caregiving hours                1–8 h  69%  74%  72%  100%  100%  ––  ––  9–20 h  17%  16%  16%  ––  ––  55%  61%  >20 h  14%  10%  12%  ––  ––  45%  39%  Weekly work hours                <35 h  21%  4%  13%  20%  ––  29%  ––  35–40 h  33%  52%  42%  80%  ––  71%  ––  41–54 h  34%  35%  34%  ––  77%  ––  75%  ≥55 h  12%  9%  11%  ––  23%  ––  25%  Abbreviations: SD, Standard Deviation. Table 1 Baseline characteristics of informal caregivers in SLOSH and Whitehall II   SLOSH  Whitehall II  Total  1–8 h care ≤40 h work  1–8 h care >40 h work  >8 h care ≤40 h work  >8 h care >40 h work  Informal caregivers  N = 700  N = 696  N = 1396  N = 504  N = 496  N = 262  N = 134  Women  68%  39%  54%  56%  41%  71%  58%  Age (Mean ± SD)  52 ± 9  50 ± 5  51 ± 7  52 ± 7  51 ± 7  50 ± 9  51 ± 7  Children  48%  63%  56%  55%  59%  52%  51%  Low occupational grade  12%  35%  23%  19%  34%  11%  22%  Smoking  19%  15%  17%  13%  14%  23%  25%  Obese  21%  11%  16%  14%  15%  18%  25%  Weekly caregiving hours                1–8 h  69%  74%  72%  100%  100%  ––  ––  9–20 h  17%  16%  16%  ––  ––  55%  61%  >20 h  14%  10%  12%  ––  ––  45%  39%  Weekly work hours                <35 h  21%  4%  13%  20%  ––  29%  ––  35–40 h  33%  52%  42%  80%  ––  71%  ––  41–54 h  34%  35%  34%  ––  77%  ––  75%  ≥55 h  12%  9%  11%  ––  23%  ––  25%    SLOSH  Whitehall II  Total  1–8 h care ≤40 h work  1–8 h care >40 h work  >8 h care ≤40 h work  >8 h care >40 h work  Informal caregivers  N = 700  N = 696  N = 1396  N = 504  N = 496  N = 262  N = 134  Women  68%  39%  54%  56%  41%  71%  58%  Age (Mean ± SD)  52 ± 9  50 ± 5  51 ± 7  52 ± 7  51 ± 7  50 ± 9  51 ± 7  Children  48%  63%  56%  55%  59%  52%  51%  Low occupational grade  12%  35%  23%  19%  34%  11%  22%  Smoking  19%  15%  17%  13%  14%  23%  25%  Obese  21%  11%  16%  14%  15%  18%  25%  Weekly caregiving hours                1–8 h  69%  74%  72%  100%  100%  ––  ––  9–20 h  17%  16%  16%  ––  ––  55%  61%  >20 h  14%  10%  12%  ––  ––  45%  39%  Weekly work hours                <35 h  21%  4%  13%  20%  ––  29%  ––  35–40 h  33%  52%  42%  80%  ––  71%  ––  41–54 h  34%  35%  34%  ––  77%  ––  75%  ≥55 h  12%  9%  11%  ––  23%  ––  25%  Abbreviations: SD, Standard Deviation. As seen in figure 1, compared to those who provided caregiving 1–8 h weekly, those who provided 9–20 weekly hours of caregiving (HR = 3.31, 95%CI: 0.53; 20.5) or > 20 h weekly (HR = 2.63; 95%CI: 1.20; 5.76) were at higher risk of CVD (Supplementary Appendix C1). There was high heterogeneity with an I2 value of 85% for the estimate on 9–20 h. Cohort specific estimates suggest that the risk of CVD was higher in SLOSH (HR = 8.71, 95%CI: 2.85; 26.6) compared to Whitehall II (HR = 1.35, 95%CI: 0.59; 3.13) (Supplementary Appendix C2, C3). There was no noteworthy heterogeneity for >20 weekly hours of caregiving. As also seen in figure 1, those working ≥55 h weekly had a higher risk of CVD compared to those working 35–40 h weekly (HR = 2.52, 95%CI: 1.14; 4.35). For those working <35 h and those working 41–54 h weekly there were no association with CVD. Additional analyzes showed that estimates attenuated slightly when adjusting for smoking and BMI (Supplementary Appendix C5). Figure 1 View largeDownload slide Associations between weekly caregiving, work hours and CVD during 10 years follow-up in 1396 informal caregivers from SLOSH and the Whitehall II study. Adjusted for age, sex, children, marital status, occupational grade, along with caregiving hours adjusted for work hours and vice versa Figure 1 View largeDownload slide Associations between weekly caregiving, work hours and CVD during 10 years follow-up in 1396 informal caregivers from SLOSH and the Whitehall II study. Adjusted for age, sex, children, marital status, occupational grade, along with caregiving hours adjusted for work hours and vice versa In figure 2, we present the joint model of weekly hours of caregiving and work. We did not find a higher risk of CVD in the group who provided few hours (1–8 h weekly) of caregiving and working >40 h (HR = 1.34; 95%CI: 0.67; 2.68). However, individuals who provided >8 h of caregiving and worked ≤40 hours per week had a higher risk of CVD compared with the reference group (HR = 3.23, 95%CI: 1.25; 8.37) (Supplementary Appendix C6). However, there was moderate heterogeneity in the latter analysis: Those exposed to many caregiving hours and long work hours in SLOSH had a markedly higher risk of CVD (HR = 8.17, 95%CI: 1.41; 47.4) compared to those with few caregiving hours and short work hours, while this was less supported in Whitehall II (HR = 1.22, 95%CI: 0.33; 4.51) (Supplementary Appendix C7, C8). Figure 2 View largeDownload slide Joint associations between weekly caregiving and work hours and CVD during 10 years follow-up in 1396 informal caregivers from SLOSH and the Whitehall II study. Adjusted for age, sex, children, marital status and occupational grade. Multiplicative interaction, P = 0.26, additive interaction, synergy index = 0.73 Figure 2 View largeDownload slide Joint associations between weekly caregiving and work hours and CVD during 10 years follow-up in 1396 informal caregivers from SLOSH and the Whitehall II study. Adjusted for age, sex, children, marital status and occupational grade. Multiplicative interaction, P = 0.26, additive interaction, synergy index = 0.73 In further analyzes using Whitehall II data (figure 3), we found that long-term caregivers providing 9–20 h (HR = 4.41, 95%CI: 0.88; 22.0) and >20 h of weekly caregiving (HR = 6.17, 95%CI: 1.73; 22.1) were at markedly higher risk of CVD compared to long-term caregivers providing caregiving for 1–8 h. For short-term caregivers, there was no association with CVD in either group (Supplementary Appendix C9). We found a multiplicative interaction between long-term caregiving and weekly hours of caregiving using a binary variable with >8 h and ≤8 h of caregiving (multiplicative interaction: P = 0.04, additive interaction: synergy index = –1.49). In general, there was no attenuation of hazard ratios in the multiple adjusted analyzes, compared to the analyzes adjusted for age and sex only. Figure 3 View largeDownload slide The association between long-term caregiving and CVD during 7 years follow-up in 616 informal caregivers from the Whitehall II study. Adjusted for age, sex, children, marital status, occupational grade, along with caregiving hours adjusted for work hours and vice versa. Long-term caregiving: Informal caregiving at baseline and 3 years later, short-term caregiving: Informal caregiving at baseline. Interaction between long-term caregiving and caregiving hours (≤8/>8 h). Multiplicative interaction, P = 0.04, additive interaction, synergy index = –1.49 Figure 3 View largeDownload slide The association between long-term caregiving and CVD during 7 years follow-up in 616 informal caregivers from the Whitehall II study. Adjusted for age, sex, children, marital status, occupational grade, along with caregiving hours adjusted for work hours and vice versa. Long-term caregiving: Informal caregiving at baseline and 3 years later, short-term caregiving: Informal caregiving at baseline. Interaction between long-term caregiving and caregiving hours (≤8/>8 h). Multiplicative interaction, P = 0.04, additive interaction, synergy index = –1.49 Discussion In this longitudinal study of informal caregivers in gainful employment, we found that those who provided many weekly hours of informal caregiving were at markedly higher risk of CVD opposed to few weekly hours, irrespectively of weekly work hours. In addition, we found that many caregiving hours were associated with a six times higher risk of CVD among long-term caregivers, whereas many caregiving hours were not associated with CVD risk among short-term caregivers. We also found that long working hours was associated with a higher risk of CVD, but there was only weak evidence to support a more detrimental effect among those with double burden from long caregiving- and work hours. Based on results from the 2001 UK Census, it is estimated that 10% of the UK population provide informal caregiving, with a peak at age 45–59, in which almost 20% provide informal caregiving.32 Furthermore, it is estimated that around 75% of caregivers in the UK are gainfully employed (Scotland excluded).1 In Sweden, it is estimated that 25% of the population provides help to a relative or other dependent.2 Given the growing population of elderly reported by the World Health Organization, these numbers are likely to be increasing.33 Thus, caregiving seems to be a public health concern as gainfully employed caregivers who provide >20 h of weekly caregiving may have a markedly higher risk of CVD compared to those providing care for 1–8 h weekly. In the Nurses’ Health Study, it was previously shown that female nurses providing weekly care for a disabled or ill spouse ≥9 h had a 1.8 times higher risk of CHD compared to non-caregivers.11 Our data complement these findings by showing that subjects providing >20 hours of caregiving for aged or disabled relatives had a 2.6 times higher risk of CVD compared to those providing care for 1–8 h weekly, irrespectively of number of weekly work hours. In another study, there was no increased risk of CHD in caregivers providing >5 weekly hours of caregiving.12 However, non-caregivers were used as the reference group, and there is potentially a large difference in providing >5 weekly hours and >20 weekly hours of care. A study on married couples also investigated the association between spousal caregiving and the risk of CVD,9 but based on reports from the care recipient instead of the caregiver. Here, high intensity caregiving was defined as ≥14 h of weekly caregiving, and they found a 35% higher risk of CVD compared to a joint reference group of those providing <14 h of weekly caregiving and non-caregivers. In line with our study, they found that long-term caregivers had a higher risk of CVD compared to short-term caregivers, also using two consecutive waves separated by 2 years as the definition of long-term caregiving.9 We had no information of how many years prior to baseline individuals provided caregiving, and results on long-term caregiving may, therefore, be underestimated. Including only caregivers provides a direct test of the effect of the number of hours of caregiving among people all exposed to some degree of caregiving strain. Furthermore, by including weekly work hours, we were able to investigate the joint workload of weekly caregiving and work hours, which has not been investigated previously. Our results from the joint model showed that many weekly hours of caregiving was associated with higher risk of CVD irrespectively of work hours, unless the caregiver worked >55 h weekly. In a previous large-scale meta-analysis, working >55 h weekly was found to be associated with a modest increased CVD risk,13 whereas we found a more than double risk of CVD. The stronger effect observed presently is likely caused by only including informal caregivers, who may be more vulnerable to the health consequences of long working hours. In Sweden, individuals are eligible for care provided by the state34; and informal caregiving is, therefore, on a more voluntary basis than in the UK where the responsibility of long-term care is placed with individuals and their families.35 We found some indication that many weekly hours of care and risk of CVD was stronger in SLOSH compared with Whitehall II. This finding could be explained with caregivers in Sweden choosing to provide care for recipients in their own household. Thus, it has been suggested that caregiving inside the household is worse for health compared with caregiving outside the household, due to the emotional stress of having the care recipient around all the time.36 Regarding generalizability of findings, the Nurses’ Health Study included female nurses11 and the study on married couples had a study population with a mean age of 64 years,9 and thus, it is likely that participants were out of the labor market. We included caregivers also in the study by Buyck et al.,12,24 but supplemented with a Swedish population of caregivers.23 Thus, with a broader range of individuals in the labor force from two European countries with different welfare-state systems, our results may be generalized to a broader population of informal caregivers in western countries. However, it should be acknowledged that all participants included were in steady work situations. Thus, given that job insecurity is a stressor that may also affect health,37 we may have underestimated the true association between weekly caregiving hours and risk of CVD in a working population, which additionally includes people with precarious jobs and temporary unemployment. Furthermore, by including only informal caregivers in the study sample, we recognize that our focus was only on a selection of the total population. Strengths of this study include the highly comparable longitudinal data on exposures and outcome from two occupational cohorts. In addition, including register-based ICD-10 diagnostic codes strengthens the validity of information. However, there were few cases of CVD, and results should, therefore, be replicated before drawing strong conclusions. People with missing information on exposures and covariates were more like to be male compared with complete cases. This gender difference is however unlike to have biased our estimates since this missing is unlikely to be associated with both informal caregiving hours and future events of CVD. A common issue in informal caregiving research, which we were not able to account for, is the ability to distinguish between care recipients. For example, it has been shown that caregiving for a spouse may be more detrimental for health compared with caregiving for e.g. a disabled parent.11 Thus, some of the effect of caregiving hours may be due to high intensity spousal caregiving, as it has been shown that spousal caregiving is particularly time-intensive.3 Most likely because you are more likely to live in the same household as your spouse, while this is not always the case with a disabled parent. However, by adjusting for age we may have partially adjusted for some of the effect from care recipients. Thus, caregiving for a parent would be more common in younger age whereas caregiving for a spouse would be more common in older age.3 Conclusions Previous studies have shown associations between informal caregiving and risk of CVD. In our study population of informal caregivers, we found that many as opposed to few weekly hours of informal caregiving was associated with a higher risk of CVD, irrespectively of weekly work hours. This, association seemed to be driven by those had provided long-term care. We also found that very long working hours were associated with a higher risk of CVD in informal caregivers, indicating that these workers may be a particularly vulnerable group. Our results emphasize the importance of preventive strategies and support systems for those who provide intensive and long-term caregiving. Supplementary data Supplementary data are available at EURPUB online. Acknowledgement The authors thank all participants, researchers and support staffs who have contributed to the cohort studies. Funding This work was supported by the Danish Work Environment Foundation (grant no. 12-2013-03). MK was supported by NordForsk, the MRC (K013351), and the Academy of Finland (311492). The SLOSH study was supported by the Swedish Council for Working Life and Social Research (FAS, grants no. 2005-0734) and the Swedish Research Council (VR, grants no. 2009-6192 and 2013-1645). The Whitehall II Study is supported by grants from The UK Medical Research Council (MR/K013351/1; G0902037), British Heart Foundation (RG/13/2/30098, PG/11/63/29011) and the US National Institutes of Health (R01HL36310, R01AG013196) have supported collection of data in the Whitehall II Study. Conflicts of interest: Dr. Kivimäki reports grants from NordForsk, the MRC (K013351), and the Academy of Finland (311492), during the conduct of the study. Dr. Lange reports personal fees from Novo Nordisk outside the submitted work. Key points In a population of informal caregivers, we show that more than 20 weekly hours of informal caregiving is associated with markedly higher risk of CVD compared to providing few weekly hours of caregiving, irrespectively of number of weekly work hours. The higher risk of CVD among caregivers seemed to be driven by those had provided long-term care. Informal caregivers working more than 55 h weekly had more than twice the risk of developing CVD compared with those working 35–40 h weekly. References 1 Beesley L. Informal Care in England. King’s Fund, 2006. 2 Szebehely M. Anhörigomsorg, förvärvsarbete och försörjning. In: Lönsamt arbete––familjeansvarets fördelning och konsekvenser . Stockholm: Elanders Sverige AB, 2014. 3 Hunt GG, Reinhard S, Greene R. Caregiving in the U.S. Research report by the AARP – Public Policy Institute. June 2015. 4 AARP – Public Policy Institute. Valuing the Invaluable: A New Look at the Economic Value of Family Caregiving . Research report by the AARP – Public Policy Institute, 2007: 1– 12. 5 Wolff JL, Spillman BC, Freedman VA, Kasper JD. A national profile of family and unpaid caregivers who assist older adults with health care activities. JAMA Intern Med  2016; 176: 372– 9. http://dx.doi.org/10.1001/jamainternmed.2015.7664 Google Scholar CrossRef Search ADS PubMed  6 Dich N, Lange T, Head J, Rod NH. Work stress, caregiving, and allostatic load. Psychosom Med  2015; 77: 539– 47. http://dx.doi.org/10.1097/PSY.0000000000000191 Google Scholar CrossRef Search ADS PubMed  7 Dragoş D, Tănăsescu MD. The effect of stress on the defense systems. J Med Life  2010; 3: 10– 8. Google Scholar PubMed  8 von Känel R, Mausbach BT, Patterson TL, et al.   Increased Framingham Coronary Heart Disease Risk Score in dementia caregivers relative to non-caregiving controls. Gerontology  2008; 54: 131– 7. Google Scholar CrossRef Search ADS PubMed  9 Capistrant BD, Moon JR, Berkman LF, Glymour MM. Current and long-term spousal caregiving and onset of cardiovascular disease. J Epidemiol Community Heal  2013; 18: 1199– 216. 10 Haley WE, Roth DL, Howard G, Safford MM. Caregiving strain and estimated risk for stroke and coronary heart disease among spouse caregivers: differential effects by race and sex. Stroke  2010; 41: 1– 13. Google Scholar CrossRef Search ADS PubMed  11 Lee S, Colditz G. a, Berkman LF, Kawachi I. Caregiving and risk of coronary heart disease in U.S. women. Am J Prev Med  2003; 24: 113– 9. Google Scholar CrossRef Search ADS PubMed  12 Buyck J-F, Ankri J, Dugravot A, et al.   Informal caregiving and the risk for coronary heart disease: the Whitehall II study. J Gerontol A Biol Sci Med Sci  2013; 68: 1316– 23. http://dx.doi.org/10.1093/gerona/glt025 Google Scholar CrossRef Search ADS PubMed  13 Kivimäki M, Jokela M, Nyberg ST, et al.   Long working hours and risk of coronary heart disease and stroke : a systematic review and meta-analysis of published and unpublished data for 603 838 individuals. Lancet  2015; 386: 1739– 46. Google Scholar CrossRef Search ADS PubMed  14 Kivimäki M, Nyberg ST, Batty GD, et al.   Job strain as a risk factor for coronary heart disease: a collaborative meta-analysis of individual participant data. Lancet  2012; 380: 1491– 7. Google Scholar CrossRef Search ADS PubMed  15 Rosengren A, Hawken S, Ounpuu S, et al.   Association of psychosocial risk factors with risk of acute myocardial infarction in 11119 cases and 13648 controls from 52 countries (the INTERHEART study): case-control study. Lancet  2004; 364: 953– 62. http://dx.doi.org/10.1016/S0140-6736(04)17019-0 Google Scholar CrossRef Search ADS PubMed  16 Krantz G, Östergren P-O. Double exposure. The combined impact of domestic responsibilities and job strain on common symptoms in employed Swedish women. Eur J Public Health  2001; 11: 413– 9. Google Scholar CrossRef Search ADS PubMed  17 Evans JM. Firms’ contribution to the reconciliation between work and family life. OECD Labour Mark Soc Policy Occas Pap  2001; 48: 1– 60. 18 Hobson CJ, Delunas L, Kesic D. Compelling evidence of the need for corporate work/life balance initiatives. J Employ Couns  2001; 38: 38– 44. http://dx.doi.org/10.1002/j.2161-1920.2001.tb00491.x Google Scholar CrossRef Search ADS   19 Hammer TH, Saksvik PØ, Nytrø K, et al.   Expanding the psychosocial work environment: workplace norms and work-family conflict as correlates of stress and health. 2004; 9: 83– 97. 20 Amstad FT, Meier LL, Fasel U, et al.   A meta-analysis of work-family conflict and various outcomes with a special emphasis on cross-domain versus matching-domain relations. J Occup Health Psychol  2011; 16: 151– 69. http://dx.doi.org/10.1037/a0022170 Google Scholar CrossRef Search ADS PubMed  21 Floderus B, Hagman M, Aronsson G, et al.   Work status, work hours and health in women with and without children. Occup Environ Med  2009; 66: 704– 10. http://dx.doi.org/10.1136/oem.2008.044883 Google Scholar CrossRef Search ADS PubMed  22 Krantz G, Berntsson L, Lundberg U. Total workload, work stress and perceived symptoms in Swedish male and female white-collar employees. Eur J Public Health  2005; 15: 209– 14. http://dx.doi.org/10.1093/eurpub/cki079 Google Scholar CrossRef Search ADS PubMed  23 Leineweber C, Baltzer M, Magnusson Hanson LL, Westerlund H. Work-family conflict and health in Swedish working women and men: a 2-year prospective analysis (the SLOSH study). Eur J Public Health  2013; 23: 710– 6. http://dx.doi.org/10.1093/eurpub/cks064 Google Scholar CrossRef Search ADS PubMed  24 Marmot M, Brunner E. Cohort Profile: the Whitehall II study. Int J Epidemiol  2005; 34: 251– 6. http://dx.doi.org/10.1093/ije/dyh372 Google Scholar CrossRef Search ADS PubMed  25 Heikkila K, Nyberg ST, Madsen IEH, et al.   Long working hours and cancer risk: a multi-cohort study. Br J Cancer  2016; 114: 813– 8. http://dx.doi.org/10.1038/bjc.2016.9 Google Scholar CrossRef Search ADS PubMed  26 WHO. ICD-10 version: 2016 [Internet]. Available at: http://apps.who.int/classifications/icd10/browse/2016/en#/IX (24 May 2017, date last accessed). 27 Hammer LB, Sauter S. Total worker health and work-life stress. J Occup Environ Med  2013; 55: S25– 9. Google Scholar CrossRef Search ADS PubMed  28 Dersimonian R, Laird N. Meta-analysis in clinical trials *. 1986; 7: 177– 88. 29 Higgins JPT, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ Br Med J  2003; 327: 557– 60. http://dx.doi.org/10.1136/bmj.327.7414.557 Google Scholar CrossRef Search ADS   30 Vandenbroucke JP, Von Elm E, Altman DG, et al.   Strengthening the reporting of observational studies in epidemiology (STROBE): explanation and elaboration. PLoS Med  2007; 4: 1628– 54. Google Scholar CrossRef Search ADS   31 Mutsert R, De Jager KH, Zoccali C, et al.   The effect of joint exposures: examining the presence of interaction. Kidney Int  2009; 75: 677– 81. http://dx.doi.org/10.1038/ki.2008.645 Google Scholar CrossRef Search ADS PubMed  32 Dahlberg L, Demack S, Bambra C. Age and gender of informal carers: a population-based study in the UK. Health Soc Care Community  2007; 15: 439– 45. http://dx.doi.org/10.1111/j.1365-2524.2007.00702.x Google Scholar CrossRef Search ADS PubMed  33 WHO. Good Health Adds Life to Years Global Brief for World Health Day, 2012, 2012: 28. 34 Fukushima N, Adami J, Palme M. The long-term care system for the elderly in Sweden. Eur Netw Econ Policy Res Institutes  2010; 89: 1– 17. 35 Comas-Herrera A, Pickard L, Wittenberg R, et al.   The long-term care system for the elderly in England. Eur Netw Econ Policy Res Institutes  2010; 74: 1– 27. 36 Kaschowitz J, Brandt M. Health effects of informal caregiving across Europe: a longitudinal approach. Soc Sci Med  2017; 173: 72– 80. http://dx.doi.org/10.1016/j.socscimed.2016.11.036 Google Scholar CrossRef Search ADS PubMed  37 Niedhammer I, Chastang JF, Sultan-Taieb H, et al.   Psychosocial work factors and sickness absence in 31 countries in Europe. Eur J Public Health  2013; 23: 622– 9. http://dx.doi.org/10.1093/eurpub/cks124 Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2017. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The European Journal of Public Health Oxford University Press

Weekly hours of informal caregiving and paid work, and the risk of cardiovascular disease

Loading next page...
 
/lp/ou_press/weekly-hours-of-informal-caregiving-and-paid-work-and-the-risk-of-uIkQdsESL4
Publisher
Oxford University Press
Copyright
© The Author(s) 2017. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved.
ISSN
1101-1262
eISSN
1464-360X
D.O.I.
10.1093/eurpub/ckx227
Publisher site
See Article on Publisher Site

Abstract

Abstract Background Little is known on the association between weekly hours of informal caregiving and risk of cardiovascular disease (CVD). The objective was to investigate the individual and joint effects of weekly hours of informal caregiving and paid work on the risk of CVD. Methods Pooled analysis with 1396 informal caregivers in gainful employment, from the Swedish Longitudinal Occupational Survey of Health and the Whitehall II study. Informal caregiving was defined as care for an aged or disabled relative. The outcome was CVD during 10 years follow-up. Analyzes were adjusted for age, sex, children, marital status and occupational grade. Results There were 59 cases of CVD. Providing care >20 h weekly were associated with a higher risk of CVD compared to those providing care 1–8 h weekly (hazard ratio = 2.63, 95%CI: 1.20; 5.76), irrespectively of weekly work hours. In sensitivity analyzes, we found this risk to be markedly higher among long-term caregivers (6.17, 95%CI: 1.73; 22.1) compared to short-term caregivers (0.89, 95%CI: 0.10; 8.08). Caregivers working ≥55 h weekly were at higher risk of CVD (2.23, 95%CI: 1.14; 4.35) compared to those working 35–40 h weekly. Those providing care >8 h and working ≤40 h weekly had a higher risk of CVD compared to those providing care 1–8 h and working ≤40 h (3.23, 95%CI: 1.25; 8.37). Conclusion A high number of weekly hours of informal caregiving as opposed to few weekly hours is associated with a higher risk of CVD, irrespectively of weekly work hours. The excess risk seemed to be driven by those providing care over long periods of time. Introduction Informal caregiving, defined as unpaid care for elderly or disabled family members, is common in working age.1–3 Helping those in need is meaningful to most people, and informal caregiving saves the health-care system considerable expenses in terms of relieving formal caregiving services.4 However, informal caregiving may lead to an emotional and physical burden for the caregiver.3,5 Furthermore, chronic stress from caregiving6 may lead to a dysregulation of the cardiovascular system via a hyper activation of the hypothalamic-pituitary-adrenal axis.7 It has, therefore, been hypothesized that informal caregiving may increase the risk of cardiovascular disease (CVD).8–10 In order to initiate preventive strategies to reduce health consequences of informal caregiving, it becomes important to unveil certain conditions under which caregiving becomes a stress burden with consequences for health. Caregiving hours may be used as an indicator of intensity, but only few studies have addressed the relationship between caregiving hours and risk of CVD. A study based on the US Nurses’ Health Study showed that women providing spousal care for ≥9 hours weekly had a higher risk of CHD compared to non-caregivers.11 Additionally, results indicate that weekly hours of caregiving had a greater impact on risk of CHD compared to perceived stress from caregiving.11 However, Nurses’ Health Study only included female nurses, which may limit generalizability to the general working population. In another study among British civil servants, an association between informal caregiving and CHD was found only for caregivers in poor health.12 In an additional analysis, they found no increased risk of CHD in those providing ≤5 h or >5 h of weekly care compared to non-caregivers. However, dividing at 5 h weekly may not capture those with high intensity caregiving, i.e. many weekly hours. The previous literature is limited by having non-caregivers as the reference, since being a non-caregiver does not include the mental burden of caregiving responsibilities, which caregivers have to face regardless of whether they provide caregiving for one or 30 h weekly. By only including caregivers, we will be able to thoroughly investigate the contribution of weekly hours of caregiving. This is highly relevant in terms of designing support systems aimed at relieving some of the caregiving burden. A further limitation in previous research on caregiving and CVD risk relates to the double burden from informal caregiving and paid work. Work stress and long working hours are independent risk factors for CHD13–15 and it is important to recognize that work and family life are strongly intertwined and should not be treated separately in relation to health consequences.16–18 In support of this, several studies have shown that spillover between work and family life may be associated with low self-reported health.19–22 Presently, we sought to investigate the individual and joint effects of weekly hours of informal caregiving and paid work on the risk of CVD, in informal caregivers in gainful employment. Methods This is a multi-cohort study pooling individual level data from the Swedish Longitudinal Occupational Survey of Health (SLOSH)23 and the Whitehall II study from the UK.24 SLOSH is an open cohort and comprises 40 877 individuals, representative of the Swedish workforce. Whitehall II includes 10 308 British civil servants from 20 London-based departments. Year 2008 was the baseline in SLOSH (response rate 61%), and years 1991–93 were the baseline in Whitehall II (response rate 87%). The total study population encompassed 1396 informal caregivers in gainful employment (Supplementary Appendix A). The cohorts has been approved by the appropriate ethics committees and participants have given consent to be included. Weekly hours of informal caregiving and paid work Informal caregiving was in both cohorts assessed by asking whether participants provided care for an aged or disabled relative. We included only those who answered ‘yes.’ We categorized weekly hours of caregiving in three groups; 1–8 h (reference), 9–20 h and >20 h.11 In Whitehall II, we had additional information on long-term caregiving, distinguishing those who provided informal care 2–3 years following baseline (long-term caregivers) from those provided care at baseline only (short-term caregivers).9,12 For weekly work hours, individuals in SLOSH were asked how many hours they worked each week, and in Whitehall II individuals were asked how many hours they worked on a normal workday. To harmonize this information, we multiplied daily work hours in Whitehall II by five, based on the assumption that individuals in the civil service work 5 days a week. Weekly work hours were categorized as <35 h, 35–40 h (reference), 41–54 h and ≥55 h per week.13,25 We investigated the joint effect of weekly hours of informal caregiving and paid work more thoroughly by dividing participants into four mutually exclusive groups: (1) Those providing care 1–8 h and working ≤40 h weekly, (2) providing care 1–8 h and working >40 h weekly, (3) providing care >8 h and working ≤40 h weekly and (4) providing care >8 h and working >40 h weekly. Ascertainment of CVD We used register-based information on diagnosis of CVD defined according to the 10th edition of the International Classification of Disease (ICD).26 CVD encompassed diagnoses CHD (ICD-10 codes I20–I25) and stroke (ICD-10 codes I60–I63). Individuals with a CVD diagnosis at baseline were excluded from the study. Covariates Confounding and mediating variables related to CVD were identified based on prior knowledge and using the method of directed acyclic graphs (Supplementary Appendix B). Confounding variables included age, sex, children (yes, no), marital status (yes, no) and occupational grade (low wage and manual laborers, lower non-manual and midlevel managers, and upper non-manual workers and managing staff). Smoking and BMI measured at baseline, were most likely mediating factors on the causal pathway from workload of caregiving and work and risk of CVD,27 and were only adjusted for in a sensitivity analysis. Statistical analyzes We applied a Cox regression model with time from baseline as the underlying time axis. The outcome was time until CVD, death from other causes or the end of follow-up, whichever came first; with 6 years follow-up in SLOSH and 10 years follow-up in Whitehall II. The Cox model was implemented in a random-effect meta-analysis and cohort specific estimates were combined and weighted in a joint model.28 The degree of heterogeneity between cohorts was assessed using I2, which describes the percentage of total variation across studies due to heterogeneity.29 We considered I2 values of 25%, 50% and 75% to represent low, moderate and high heterogeneity, respectively.29 We applied a random effects model, because there may be between-group variation in SLOSH and Whitehall II on the investigated associations, due to different follow-up time and welfare-state differences. Analyzes showed no effect modification by sex (Supplementary Appendix C4). We applied two analytical models: model 1 was adjusted for age and sex, and model 2 was also adjusted for children, marital status and occupational grade; analyzes of caregiving hours were further adjusted for work hours and those of work hours were adjusted for caregiving hours. Interaction was assessed on a multiplicative and additive scale according to recommendations by the STROBE guidelines.30 We assessed multiplicative interaction by including a product term between caregiving and work hour variables in the Cox model. Additive interaction was assessed with the synergy index,31 which represents the additional risk from exposure to informal caregiving and work hours when interaction is present, relative to the risk from exposure when there is no interaction. In supplemental analyzes, we investigated risk of CVD for long-term caregivers compared to short-term caregivers. Here, we excluded the first 3 years of follow-up to minimize risk of reverse causation bias. Results Among the 1396 informal caregivers, 59 participants (4%) developed CVD within 10 410 person-years at risk; 50 cases of CHD and 9 cases of stroke. In SLOSH, there were 14 (2%) cases during 6 years follow-up and in Whitehall II, there were 45 (6%) cases during 10 years follow-up. Baseline characteristics of participants are shown in table 1. Table 1 Baseline characteristics of informal caregivers in SLOSH and Whitehall II   SLOSH  Whitehall II  Total  1–8 h care ≤40 h work  1–8 h care >40 h work  >8 h care ≤40 h work  >8 h care >40 h work  Informal caregivers  N = 700  N = 696  N = 1396  N = 504  N = 496  N = 262  N = 134  Women  68%  39%  54%  56%  41%  71%  58%  Age (Mean ± SD)  52 ± 9  50 ± 5  51 ± 7  52 ± 7  51 ± 7  50 ± 9  51 ± 7  Children  48%  63%  56%  55%  59%  52%  51%  Low occupational grade  12%  35%  23%  19%  34%  11%  22%  Smoking  19%  15%  17%  13%  14%  23%  25%  Obese  21%  11%  16%  14%  15%  18%  25%  Weekly caregiving hours                1–8 h  69%  74%  72%  100%  100%  ––  ––  9–20 h  17%  16%  16%  ––  ––  55%  61%  >20 h  14%  10%  12%  ––  ––  45%  39%  Weekly work hours                <35 h  21%  4%  13%  20%  ––  29%  ––  35–40 h  33%  52%  42%  80%  ––  71%  ––  41–54 h  34%  35%  34%  ––  77%  ––  75%  ≥55 h  12%  9%  11%  ––  23%  ––  25%    SLOSH  Whitehall II  Total  1–8 h care ≤40 h work  1–8 h care >40 h work  >8 h care ≤40 h work  >8 h care >40 h work  Informal caregivers  N = 700  N = 696  N = 1396  N = 504  N = 496  N = 262  N = 134  Women  68%  39%  54%  56%  41%  71%  58%  Age (Mean ± SD)  52 ± 9  50 ± 5  51 ± 7  52 ± 7  51 ± 7  50 ± 9  51 ± 7  Children  48%  63%  56%  55%  59%  52%  51%  Low occupational grade  12%  35%  23%  19%  34%  11%  22%  Smoking  19%  15%  17%  13%  14%  23%  25%  Obese  21%  11%  16%  14%  15%  18%  25%  Weekly caregiving hours                1–8 h  69%  74%  72%  100%  100%  ––  ––  9–20 h  17%  16%  16%  ––  ––  55%  61%  >20 h  14%  10%  12%  ––  ––  45%  39%  Weekly work hours                <35 h  21%  4%  13%  20%  ––  29%  ––  35–40 h  33%  52%  42%  80%  ––  71%  ––  41–54 h  34%  35%  34%  ––  77%  ––  75%  ≥55 h  12%  9%  11%  ––  23%  ––  25%  Abbreviations: SD, Standard Deviation. Table 1 Baseline characteristics of informal caregivers in SLOSH and Whitehall II   SLOSH  Whitehall II  Total  1–8 h care ≤40 h work  1–8 h care >40 h work  >8 h care ≤40 h work  >8 h care >40 h work  Informal caregivers  N = 700  N = 696  N = 1396  N = 504  N = 496  N = 262  N = 134  Women  68%  39%  54%  56%  41%  71%  58%  Age (Mean ± SD)  52 ± 9  50 ± 5  51 ± 7  52 ± 7  51 ± 7  50 ± 9  51 ± 7  Children  48%  63%  56%  55%  59%  52%  51%  Low occupational grade  12%  35%  23%  19%  34%  11%  22%  Smoking  19%  15%  17%  13%  14%  23%  25%  Obese  21%  11%  16%  14%  15%  18%  25%  Weekly caregiving hours                1–8 h  69%  74%  72%  100%  100%  ––  ––  9–20 h  17%  16%  16%  ––  ––  55%  61%  >20 h  14%  10%  12%  ––  ––  45%  39%  Weekly work hours                <35 h  21%  4%  13%  20%  ––  29%  ––  35–40 h  33%  52%  42%  80%  ––  71%  ––  41–54 h  34%  35%  34%  ––  77%  ––  75%  ≥55 h  12%  9%  11%  ––  23%  ––  25%    SLOSH  Whitehall II  Total  1–8 h care ≤40 h work  1–8 h care >40 h work  >8 h care ≤40 h work  >8 h care >40 h work  Informal caregivers  N = 700  N = 696  N = 1396  N = 504  N = 496  N = 262  N = 134  Women  68%  39%  54%  56%  41%  71%  58%  Age (Mean ± SD)  52 ± 9  50 ± 5  51 ± 7  52 ± 7  51 ± 7  50 ± 9  51 ± 7  Children  48%  63%  56%  55%  59%  52%  51%  Low occupational grade  12%  35%  23%  19%  34%  11%  22%  Smoking  19%  15%  17%  13%  14%  23%  25%  Obese  21%  11%  16%  14%  15%  18%  25%  Weekly caregiving hours                1–8 h  69%  74%  72%  100%  100%  ––  ––  9–20 h  17%  16%  16%  ––  ––  55%  61%  >20 h  14%  10%  12%  ––  ––  45%  39%  Weekly work hours                <35 h  21%  4%  13%  20%  ––  29%  ––  35–40 h  33%  52%  42%  80%  ––  71%  ––  41–54 h  34%  35%  34%  ––  77%  ––  75%  ≥55 h  12%  9%  11%  ––  23%  ––  25%  Abbreviations: SD, Standard Deviation. As seen in figure 1, compared to those who provided caregiving 1–8 h weekly, those who provided 9–20 weekly hours of caregiving (HR = 3.31, 95%CI: 0.53; 20.5) or > 20 h weekly (HR = 2.63; 95%CI: 1.20; 5.76) were at higher risk of CVD (Supplementary Appendix C1). There was high heterogeneity with an I2 value of 85% for the estimate on 9–20 h. Cohort specific estimates suggest that the risk of CVD was higher in SLOSH (HR = 8.71, 95%CI: 2.85; 26.6) compared to Whitehall II (HR = 1.35, 95%CI: 0.59; 3.13) (Supplementary Appendix C2, C3). There was no noteworthy heterogeneity for >20 weekly hours of caregiving. As also seen in figure 1, those working ≥55 h weekly had a higher risk of CVD compared to those working 35–40 h weekly (HR = 2.52, 95%CI: 1.14; 4.35). For those working <35 h and those working 41–54 h weekly there were no association with CVD. Additional analyzes showed that estimates attenuated slightly when adjusting for smoking and BMI (Supplementary Appendix C5). Figure 1 View largeDownload slide Associations between weekly caregiving, work hours and CVD during 10 years follow-up in 1396 informal caregivers from SLOSH and the Whitehall II study. Adjusted for age, sex, children, marital status, occupational grade, along with caregiving hours adjusted for work hours and vice versa Figure 1 View largeDownload slide Associations between weekly caregiving, work hours and CVD during 10 years follow-up in 1396 informal caregivers from SLOSH and the Whitehall II study. Adjusted for age, sex, children, marital status, occupational grade, along with caregiving hours adjusted for work hours and vice versa In figure 2, we present the joint model of weekly hours of caregiving and work. We did not find a higher risk of CVD in the group who provided few hours (1–8 h weekly) of caregiving and working >40 h (HR = 1.34; 95%CI: 0.67; 2.68). However, individuals who provided >8 h of caregiving and worked ≤40 hours per week had a higher risk of CVD compared with the reference group (HR = 3.23, 95%CI: 1.25; 8.37) (Supplementary Appendix C6). However, there was moderate heterogeneity in the latter analysis: Those exposed to many caregiving hours and long work hours in SLOSH had a markedly higher risk of CVD (HR = 8.17, 95%CI: 1.41; 47.4) compared to those with few caregiving hours and short work hours, while this was less supported in Whitehall II (HR = 1.22, 95%CI: 0.33; 4.51) (Supplementary Appendix C7, C8). Figure 2 View largeDownload slide Joint associations between weekly caregiving and work hours and CVD during 10 years follow-up in 1396 informal caregivers from SLOSH and the Whitehall II study. Adjusted for age, sex, children, marital status and occupational grade. Multiplicative interaction, P = 0.26, additive interaction, synergy index = 0.73 Figure 2 View largeDownload slide Joint associations between weekly caregiving and work hours and CVD during 10 years follow-up in 1396 informal caregivers from SLOSH and the Whitehall II study. Adjusted for age, sex, children, marital status and occupational grade. Multiplicative interaction, P = 0.26, additive interaction, synergy index = 0.73 In further analyzes using Whitehall II data (figure 3), we found that long-term caregivers providing 9–20 h (HR = 4.41, 95%CI: 0.88; 22.0) and >20 h of weekly caregiving (HR = 6.17, 95%CI: 1.73; 22.1) were at markedly higher risk of CVD compared to long-term caregivers providing caregiving for 1–8 h. For short-term caregivers, there was no association with CVD in either group (Supplementary Appendix C9). We found a multiplicative interaction between long-term caregiving and weekly hours of caregiving using a binary variable with >8 h and ≤8 h of caregiving (multiplicative interaction: P = 0.04, additive interaction: synergy index = –1.49). In general, there was no attenuation of hazard ratios in the multiple adjusted analyzes, compared to the analyzes adjusted for age and sex only. Figure 3 View largeDownload slide The association between long-term caregiving and CVD during 7 years follow-up in 616 informal caregivers from the Whitehall II study. Adjusted for age, sex, children, marital status, occupational grade, along with caregiving hours adjusted for work hours and vice versa. Long-term caregiving: Informal caregiving at baseline and 3 years later, short-term caregiving: Informal caregiving at baseline. Interaction between long-term caregiving and caregiving hours (≤8/>8 h). Multiplicative interaction, P = 0.04, additive interaction, synergy index = –1.49 Figure 3 View largeDownload slide The association between long-term caregiving and CVD during 7 years follow-up in 616 informal caregivers from the Whitehall II study. Adjusted for age, sex, children, marital status, occupational grade, along with caregiving hours adjusted for work hours and vice versa. Long-term caregiving: Informal caregiving at baseline and 3 years later, short-term caregiving: Informal caregiving at baseline. Interaction between long-term caregiving and caregiving hours (≤8/>8 h). Multiplicative interaction, P = 0.04, additive interaction, synergy index = –1.49 Discussion In this longitudinal study of informal caregivers in gainful employment, we found that those who provided many weekly hours of informal caregiving were at markedly higher risk of CVD opposed to few weekly hours, irrespectively of weekly work hours. In addition, we found that many caregiving hours were associated with a six times higher risk of CVD among long-term caregivers, whereas many caregiving hours were not associated with CVD risk among short-term caregivers. We also found that long working hours was associated with a higher risk of CVD, but there was only weak evidence to support a more detrimental effect among those with double burden from long caregiving- and work hours. Based on results from the 2001 UK Census, it is estimated that 10% of the UK population provide informal caregiving, with a peak at age 45–59, in which almost 20% provide informal caregiving.32 Furthermore, it is estimated that around 75% of caregivers in the UK are gainfully employed (Scotland excluded).1 In Sweden, it is estimated that 25% of the population provides help to a relative or other dependent.2 Given the growing population of elderly reported by the World Health Organization, these numbers are likely to be increasing.33 Thus, caregiving seems to be a public health concern as gainfully employed caregivers who provide >20 h of weekly caregiving may have a markedly higher risk of CVD compared to those providing care for 1–8 h weekly. In the Nurses’ Health Study, it was previously shown that female nurses providing weekly care for a disabled or ill spouse ≥9 h had a 1.8 times higher risk of CHD compared to non-caregivers.11 Our data complement these findings by showing that subjects providing >20 hours of caregiving for aged or disabled relatives had a 2.6 times higher risk of CVD compared to those providing care for 1–8 h weekly, irrespectively of number of weekly work hours. In another study, there was no increased risk of CHD in caregivers providing >5 weekly hours of caregiving.12 However, non-caregivers were used as the reference group, and there is potentially a large difference in providing >5 weekly hours and >20 weekly hours of care. A study on married couples also investigated the association between spousal caregiving and the risk of CVD,9 but based on reports from the care recipient instead of the caregiver. Here, high intensity caregiving was defined as ≥14 h of weekly caregiving, and they found a 35% higher risk of CVD compared to a joint reference group of those providing <14 h of weekly caregiving and non-caregivers. In line with our study, they found that long-term caregivers had a higher risk of CVD compared to short-term caregivers, also using two consecutive waves separated by 2 years as the definition of long-term caregiving.9 We had no information of how many years prior to baseline individuals provided caregiving, and results on long-term caregiving may, therefore, be underestimated. Including only caregivers provides a direct test of the effect of the number of hours of caregiving among people all exposed to some degree of caregiving strain. Furthermore, by including weekly work hours, we were able to investigate the joint workload of weekly caregiving and work hours, which has not been investigated previously. Our results from the joint model showed that many weekly hours of caregiving was associated with higher risk of CVD irrespectively of work hours, unless the caregiver worked >55 h weekly. In a previous large-scale meta-analysis, working >55 h weekly was found to be associated with a modest increased CVD risk,13 whereas we found a more than double risk of CVD. The stronger effect observed presently is likely caused by only including informal caregivers, who may be more vulnerable to the health consequences of long working hours. In Sweden, individuals are eligible for care provided by the state34; and informal caregiving is, therefore, on a more voluntary basis than in the UK where the responsibility of long-term care is placed with individuals and their families.35 We found some indication that many weekly hours of care and risk of CVD was stronger in SLOSH compared with Whitehall II. This finding could be explained with caregivers in Sweden choosing to provide care for recipients in their own household. Thus, it has been suggested that caregiving inside the household is worse for health compared with caregiving outside the household, due to the emotional stress of having the care recipient around all the time.36 Regarding generalizability of findings, the Nurses’ Health Study included female nurses11 and the study on married couples had a study population with a mean age of 64 years,9 and thus, it is likely that participants were out of the labor market. We included caregivers also in the study by Buyck et al.,12,24 but supplemented with a Swedish population of caregivers.23 Thus, with a broader range of individuals in the labor force from two European countries with different welfare-state systems, our results may be generalized to a broader population of informal caregivers in western countries. However, it should be acknowledged that all participants included were in steady work situations. Thus, given that job insecurity is a stressor that may also affect health,37 we may have underestimated the true association between weekly caregiving hours and risk of CVD in a working population, which additionally includes people with precarious jobs and temporary unemployment. Furthermore, by including only informal caregivers in the study sample, we recognize that our focus was only on a selection of the total population. Strengths of this study include the highly comparable longitudinal data on exposures and outcome from two occupational cohorts. In addition, including register-based ICD-10 diagnostic codes strengthens the validity of information. However, there were few cases of CVD, and results should, therefore, be replicated before drawing strong conclusions. People with missing information on exposures and covariates were more like to be male compared with complete cases. This gender difference is however unlike to have biased our estimates since this missing is unlikely to be associated with both informal caregiving hours and future events of CVD. A common issue in informal caregiving research, which we were not able to account for, is the ability to distinguish between care recipients. For example, it has been shown that caregiving for a spouse may be more detrimental for health compared with caregiving for e.g. a disabled parent.11 Thus, some of the effect of caregiving hours may be due to high intensity spousal caregiving, as it has been shown that spousal caregiving is particularly time-intensive.3 Most likely because you are more likely to live in the same household as your spouse, while this is not always the case with a disabled parent. However, by adjusting for age we may have partially adjusted for some of the effect from care recipients. Thus, caregiving for a parent would be more common in younger age whereas caregiving for a spouse would be more common in older age.3 Conclusions Previous studies have shown associations between informal caregiving and risk of CVD. In our study population of informal caregivers, we found that many as opposed to few weekly hours of informal caregiving was associated with a higher risk of CVD, irrespectively of weekly work hours. This, association seemed to be driven by those had provided long-term care. We also found that very long working hours were associated with a higher risk of CVD in informal caregivers, indicating that these workers may be a particularly vulnerable group. Our results emphasize the importance of preventive strategies and support systems for those who provide intensive and long-term caregiving. Supplementary data Supplementary data are available at EURPUB online. Acknowledgement The authors thank all participants, researchers and support staffs who have contributed to the cohort studies. Funding This work was supported by the Danish Work Environment Foundation (grant no. 12-2013-03). MK was supported by NordForsk, the MRC (K013351), and the Academy of Finland (311492). The SLOSH study was supported by the Swedish Council for Working Life and Social Research (FAS, grants no. 2005-0734) and the Swedish Research Council (VR, grants no. 2009-6192 and 2013-1645). The Whitehall II Study is supported by grants from The UK Medical Research Council (MR/K013351/1; G0902037), British Heart Foundation (RG/13/2/30098, PG/11/63/29011) and the US National Institutes of Health (R01HL36310, R01AG013196) have supported collection of data in the Whitehall II Study. Conflicts of interest: Dr. Kivimäki reports grants from NordForsk, the MRC (K013351), and the Academy of Finland (311492), during the conduct of the study. Dr. Lange reports personal fees from Novo Nordisk outside the submitted work. Key points In a population of informal caregivers, we show that more than 20 weekly hours of informal caregiving is associated with markedly higher risk of CVD compared to providing few weekly hours of caregiving, irrespectively of number of weekly work hours. The higher risk of CVD among caregivers seemed to be driven by those had provided long-term care. Informal caregivers working more than 55 h weekly had more than twice the risk of developing CVD compared with those working 35–40 h weekly. References 1 Beesley L. Informal Care in England. King’s Fund, 2006. 2 Szebehely M. Anhörigomsorg, förvärvsarbete och försörjning. In: Lönsamt arbete––familjeansvarets fördelning och konsekvenser . Stockholm: Elanders Sverige AB, 2014. 3 Hunt GG, Reinhard S, Greene R. Caregiving in the U.S. Research report by the AARP – Public Policy Institute. June 2015. 4 AARP – Public Policy Institute. Valuing the Invaluable: A New Look at the Economic Value of Family Caregiving . Research report by the AARP – Public Policy Institute, 2007: 1– 12. 5 Wolff JL, Spillman BC, Freedman VA, Kasper JD. A national profile of family and unpaid caregivers who assist older adults with health care activities. JAMA Intern Med  2016; 176: 372– 9. http://dx.doi.org/10.1001/jamainternmed.2015.7664 Google Scholar CrossRef Search ADS PubMed  6 Dich N, Lange T, Head J, Rod NH. Work stress, caregiving, and allostatic load. Psychosom Med  2015; 77: 539– 47. http://dx.doi.org/10.1097/PSY.0000000000000191 Google Scholar CrossRef Search ADS PubMed  7 Dragoş D, Tănăsescu MD. The effect of stress on the defense systems. J Med Life  2010; 3: 10– 8. Google Scholar PubMed  8 von Känel R, Mausbach BT, Patterson TL, et al.   Increased Framingham Coronary Heart Disease Risk Score in dementia caregivers relative to non-caregiving controls. Gerontology  2008; 54: 131– 7. Google Scholar CrossRef Search ADS PubMed  9 Capistrant BD, Moon JR, Berkman LF, Glymour MM. Current and long-term spousal caregiving and onset of cardiovascular disease. J Epidemiol Community Heal  2013; 18: 1199– 216. 10 Haley WE, Roth DL, Howard G, Safford MM. Caregiving strain and estimated risk for stroke and coronary heart disease among spouse caregivers: differential effects by race and sex. Stroke  2010; 41: 1– 13. Google Scholar CrossRef Search ADS PubMed  11 Lee S, Colditz G. a, Berkman LF, Kawachi I. Caregiving and risk of coronary heart disease in U.S. women. Am J Prev Med  2003; 24: 113– 9. Google Scholar CrossRef Search ADS PubMed  12 Buyck J-F, Ankri J, Dugravot A, et al.   Informal caregiving and the risk for coronary heart disease: the Whitehall II study. J Gerontol A Biol Sci Med Sci  2013; 68: 1316– 23. http://dx.doi.org/10.1093/gerona/glt025 Google Scholar CrossRef Search ADS PubMed  13 Kivimäki M, Jokela M, Nyberg ST, et al.   Long working hours and risk of coronary heart disease and stroke : a systematic review and meta-analysis of published and unpublished data for 603 838 individuals. Lancet  2015; 386: 1739– 46. Google Scholar CrossRef Search ADS PubMed  14 Kivimäki M, Nyberg ST, Batty GD, et al.   Job strain as a risk factor for coronary heart disease: a collaborative meta-analysis of individual participant data. Lancet  2012; 380: 1491– 7. Google Scholar CrossRef Search ADS PubMed  15 Rosengren A, Hawken S, Ounpuu S, et al.   Association of psychosocial risk factors with risk of acute myocardial infarction in 11119 cases and 13648 controls from 52 countries (the INTERHEART study): case-control study. Lancet  2004; 364: 953– 62. http://dx.doi.org/10.1016/S0140-6736(04)17019-0 Google Scholar CrossRef Search ADS PubMed  16 Krantz G, Östergren P-O. Double exposure. The combined impact of domestic responsibilities and job strain on common symptoms in employed Swedish women. Eur J Public Health  2001; 11: 413– 9. Google Scholar CrossRef Search ADS PubMed  17 Evans JM. Firms’ contribution to the reconciliation between work and family life. OECD Labour Mark Soc Policy Occas Pap  2001; 48: 1– 60. 18 Hobson CJ, Delunas L, Kesic D. Compelling evidence of the need for corporate work/life balance initiatives. J Employ Couns  2001; 38: 38– 44. http://dx.doi.org/10.1002/j.2161-1920.2001.tb00491.x Google Scholar CrossRef Search ADS   19 Hammer TH, Saksvik PØ, Nytrø K, et al.   Expanding the psychosocial work environment: workplace norms and work-family conflict as correlates of stress and health. 2004; 9: 83– 97. 20 Amstad FT, Meier LL, Fasel U, et al.   A meta-analysis of work-family conflict and various outcomes with a special emphasis on cross-domain versus matching-domain relations. J Occup Health Psychol  2011; 16: 151– 69. http://dx.doi.org/10.1037/a0022170 Google Scholar CrossRef Search ADS PubMed  21 Floderus B, Hagman M, Aronsson G, et al.   Work status, work hours and health in women with and without children. Occup Environ Med  2009; 66: 704– 10. http://dx.doi.org/10.1136/oem.2008.044883 Google Scholar CrossRef Search ADS PubMed  22 Krantz G, Berntsson L, Lundberg U. Total workload, work stress and perceived symptoms in Swedish male and female white-collar employees. Eur J Public Health  2005; 15: 209– 14. http://dx.doi.org/10.1093/eurpub/cki079 Google Scholar CrossRef Search ADS PubMed  23 Leineweber C, Baltzer M, Magnusson Hanson LL, Westerlund H. Work-family conflict and health in Swedish working women and men: a 2-year prospective analysis (the SLOSH study). Eur J Public Health  2013; 23: 710– 6. http://dx.doi.org/10.1093/eurpub/cks064 Google Scholar CrossRef Search ADS PubMed  24 Marmot M, Brunner E. Cohort Profile: the Whitehall II study. Int J Epidemiol  2005; 34: 251– 6. http://dx.doi.org/10.1093/ije/dyh372 Google Scholar CrossRef Search ADS PubMed  25 Heikkila K, Nyberg ST, Madsen IEH, et al.   Long working hours and cancer risk: a multi-cohort study. Br J Cancer  2016; 114: 813– 8. http://dx.doi.org/10.1038/bjc.2016.9 Google Scholar CrossRef Search ADS PubMed  26 WHO. ICD-10 version: 2016 [Internet]. Available at: http://apps.who.int/classifications/icd10/browse/2016/en#/IX (24 May 2017, date last accessed). 27 Hammer LB, Sauter S. Total worker health and work-life stress. J Occup Environ Med  2013; 55: S25– 9. Google Scholar CrossRef Search ADS PubMed  28 Dersimonian R, Laird N. Meta-analysis in clinical trials *. 1986; 7: 177– 88. 29 Higgins JPT, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ Br Med J  2003; 327: 557– 60. http://dx.doi.org/10.1136/bmj.327.7414.557 Google Scholar CrossRef Search ADS   30 Vandenbroucke JP, Von Elm E, Altman DG, et al.   Strengthening the reporting of observational studies in epidemiology (STROBE): explanation and elaboration. PLoS Med  2007; 4: 1628– 54. Google Scholar CrossRef Search ADS   31 Mutsert R, De Jager KH, Zoccali C, et al.   The effect of joint exposures: examining the presence of interaction. Kidney Int  2009; 75: 677– 81. http://dx.doi.org/10.1038/ki.2008.645 Google Scholar CrossRef Search ADS PubMed  32 Dahlberg L, Demack S, Bambra C. Age and gender of informal carers: a population-based study in the UK. Health Soc Care Community  2007; 15: 439– 45. http://dx.doi.org/10.1111/j.1365-2524.2007.00702.x Google Scholar CrossRef Search ADS PubMed  33 WHO. Good Health Adds Life to Years Global Brief for World Health Day, 2012, 2012: 28. 34 Fukushima N, Adami J, Palme M. The long-term care system for the elderly in Sweden. Eur Netw Econ Policy Res Institutes  2010; 89: 1– 17. 35 Comas-Herrera A, Pickard L, Wittenberg R, et al.   The long-term care system for the elderly in England. Eur Netw Econ Policy Res Institutes  2010; 74: 1– 27. 36 Kaschowitz J, Brandt M. Health effects of informal caregiving across Europe: a longitudinal approach. Soc Sci Med  2017; 173: 72– 80. http://dx.doi.org/10.1016/j.socscimed.2016.11.036 Google Scholar CrossRef Search ADS PubMed  37 Niedhammer I, Chastang JF, Sultan-Taieb H, et al.   Psychosocial work factors and sickness absence in 31 countries in Europe. Eur J Public Health  2013; 23: 622– 9. http://dx.doi.org/10.1093/eurpub/cks124 Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2017. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved.

Journal

The European Journal of Public HealthOxford University Press

Published: Dec 22, 2017

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off