FormalPara Key Points for Decision Makers

Informal caregiving involves millions of individuals who spend billions of hours caring for patients who would otherwise require care from the healthcare system. This resource should be seen as scarce and as valuable as any other health service within the health system. Therefore, it is important to consider the value of informal caring as part of the health technology assessment.

The provision of informal caregiving is adversely associated with the workplace productivity of caregivers in terms of increased absenteeism, a greater chance of presenteeism and higher working hour tension.

Policymakers should prioritise increasing access to publicly funded or subsidised formal care. If there were more formal care options available, informal caregivers might have less caregiving burden and be more productive at work.

1 Introduction

Informal caregiving is mostly an unpaid job provided by family members and friends to the elderly and people with disability or long-term health conditions [1]. About 65% of informal caregivers across Organization for Economic Co-operation and Development (OECD) countries were women in 2019 [2]. In 2018, one in ten Australians provided informal care, representing 2.65 million people [3]. Informal caregiving has known adverse effects on the health of carers, leads to high rates of depression and anxiety, and causes emotional and physical strain as well as stress [4, 5]. Policies to support informal caregivers in many OECD countries have included the provision of paid or unpaid care leave, cash benefits, cash-for-care allowances for recipients, periods of paid leave for informal carers and counselling/training services [2, 6]. Despite this, the work performance of informal caregivers continues to be negatively impacted as there is a competing demand between informal caregiving and paid work [7].

Unpaid carers contribute substantial value to the health system and society. Informal caregiving involves millions of individuals who spend billions of hours caring for patients who would otherwise require care from the healthcare system. This resource should be seen as scarce and as valuable as any other health service within the health system. Therefore, it is important to consider the value of informal caring as part of the health technology assessment. There are direct and indirect costs associated with the investment of time of the caregivers. There may be lost income because they are unable to work or unable to work as much as they performed before. There are indirect productivity losses to society. If a carer cannot come in to do their job in the workplace, society will miss out on the formal value of that carer’s time. They may be otherwise volunteering in a capacity that would add value to society. Policymakers are often interested in knowing the value of the reduction in the burden on carers if there is a new treatment or a new health technology available that reduces the burden placed on informal carers. The goal of health technology assessment is to ensure the utilization of scarce and limited resources in the most efficient way. Therefore, it is important to take these values and burdens into consideration when health technology assessors consider the value of new medicines of new healthcare technology. Economic evaluations that disregard informal care would undervalue a health intervention that reduces the caregiving burden [8].

Informal caregiving for family members with long-term illnesses or disabilities burdens the caregiver [9]. Some empirical studies have examined the association between informal caregiving and productivity at work. However, results from these studies on the association between informal caregiving and productivity at work are mixed and inconclusive. For instance, a study by Bryan (2012), using matched employer–employee data, examined the association between employees’ access to flexible working arrangements and the amount of informal care provision to sick or elderly friends and relatives. The authors found that about 13% more hours of informal care were associated with flexitime and the ability to reduce working hours [10]. Chen et al. (2017) utilized data from the 1991–2009 China Health and Nutrition Survey and examined the association between informal care and labour market outcomes among married women of working age. They found that the intensity of caregiving significantly reduced weekly hours of work among the study population [11]. Fakeye et al. (2022) estimated caregiving-related absenteeism and presenteeism among US adults. They found that formal caregiving reduced productivity at work by one-third, with productivity loss higher among older adult caregivers [12]. Johnson and Lo Sasso (2001) examined the association between labour supply, time help to parents and financial assistance to parents among US working-age individuals. They found that the provision of informal care to parents by women reduced work hours, while the provision of financial assistance increased the number of hours spent working [13].

Mazanec et al. (2011) examined the association between caregiving burden and productivity loss in the USA. They found that a greater number of caregiving hours was associated with greater work productivity loss, among others [4]. Using a prospective cohort study of working adults in France, Finland, and the UK, Mortensen et al. (2017) examined the effects of job strain and informal caregiving on long-term sickness absence with special attention to gender differences. They found that high job strain and informal caregiving were significantly associated with a higher risk of sickness absence among women [14]. Van Houtven et al. (2013) also examined the association between informal caregiving and work in the USA. The authors found that the provision of informal care by women who remain working reduced work by 3–10 hours per week, while there was little effect of informal caregiving on working hours among men [15]. Wolf and Soldo (1994) estimated the effect of the provision of care to older parents on employment and hours of work among married women and found no evidence of the effect of the provision of care to older parents on employment and hours of work [16]. Yamada and Satoshi (2015) examined the association between informal care provision and labour market outcomes in Japan, and found that the provision of family care for parents adversely affected the probability of working hours worked and employment status, with the effect being greater among female caregivers compared with male caregivers [17]. A study in Australia by Gray and Hughes (2005) on the use of flexible working arrangements to provide adult care compared with the use by those caring for children found that one in seven working individuals with caregiving responsibilities desire but cannot utilise flexible working hours, shift work and work-from-home possibilities. However, this study did not examine the between-person differences in the association between informal caregiving and productivity burdens such as absenteeism, presenteeism and working hour tension [18].

In summary, while existing studies on the productivity burden of informal caregiving for disabled or elderly relatives have been conducted in other developed countries [4, 7, 10, 13,14,15,16,17, 19], no study has examined the between-person difference in the association between informal caregiving and productivity burden in Australia. Furthermore, results from previous studies cannot be generalised to the Australian population. We aim to fill this gap in the literature by examining the between-person differences in the association between informal caregiving and three measures of productivity following a retrospective, longitudinal research design.

2 Methods

2.1 Data Source

We used data from the Household, Income and Labour Dynamics in Australia (HILDA) survey for the empirical analyses. HILDA is a nationally representative longitudinal study that collects data annually from over 13,000 individuals from over 7000 households using a multi-stage sampling procedure since 2001. It is designed and managed by the Melbourne Institute of Applied Economic and Social Research with the financial support of the Australian Government Department of Social Services. The survey collects data from household members aged 15 years or above using a combination of self-completion questionnaires alongside in-person and telephone interviews conducted by professional interviewers. The survey collects detailed information on a wide range of topics, such as wealth, retirement, fertility, health, education and demographic characteristics. To perform the empirical analyses, we have selected the HILDA dataset since it contains information on informal caregiving, absenteeism, presenteeism, preferred hours to work and actual hours to work alongside socio-demographic and lifestyle characteristics at different periods. The details of the survey can be found elsewhere [20].

2.2 Analytic Sample and Missing Data

This study analysed data from 11 waves (waves 11–21) of the HILDA survey, spanning the period from 2011 to 2021. The analysis is restricted to individuals currently in paid employment for the regression analyses of presenteeism and working hour tension. However, we considered both employed and unemployed respondents for the regression analysis of absenteeism. Our sample is confined to respondents having complete data on the main variable of interest (informal caregiving) and outcome variables (absenteeism, presenteeism and working hour tension). The final analytic sample comprises 97,521 yearly observations (18,612 unique participants) for absenteeism, and 96,818 yearly observations (18,139 unique participants) for presenteeism and working hour tension. Figure 1 depicts the flow of participants into the analytic sample.

Fig. 1
figure 1

Participants flow into the analytic sample and missing data

Full size image

2.3 Measures

2.3.1 Outcome Variables

We measure productivity through absenteeism (count variable), presenteeism (binary variable) and working hour tension (count variable). The outcome variable, absenteeism, is a self-reported measure of the number of leave taken in the previous 12 months. Total leave of a respondent was ascertained through asking four questions: how many days of paid annual leave, sick leave, unpaid leave and other leave (maternity, paternity, bereavement, etc.) have you taken in last 12 months. To determine absenteeism, we additionally counted the number of days a responder was unemployed over the previous fiscal year. We summed four forms of leave and days unemployed to compute a respondent’s overall absenteeism in the previous 12 months.

Another outcome, presenteeism, was derived from the short form (SF-36) Health Survey, administered yearly in the HILDA survey as part of the self-completion questionnaire (SCQ). More specifically, individuals are asked whether, as a result of any emotional problems, they have experienced any of the following in the past four weeks: ‘cutting down the amount of time you spent on work or other activities’, ‘accomplished less than you would like’ and ‘did not do work or other activities as carefully as usual’. Participants were also asked whether they have experienced any of the following three events in the past four weeks due to any physical problems: ‘cut down the amount of time spent on work or other activities’, ‘accomplished less than would like’ and ‘were limited in the kind of work’. The responses were recorded in binary form: yes or no. Using these six questions, we construct a binary indicator taking the value of 1 if a respondent answered ‘Yes’ to any of the questions in that year and 0 otherwise.

Working hour tension is a count variable derived from two variables following previous research [21]. The authors derived working hour tension by subtracting hours per week usually worked in all jobs from hours preferred to work per week. If the weekly worked hours are higher or equal to the preferred working hour, then the derived variable ‘working hour tension’ takes the value zero. Working hour tension will be non-negative integers if a respondent’s preferred hours to work is greater than weekly worked hours.

2.3.2 Exposure Variable

Informal caregiving is the act of providing assistance to family members, friends or neighbours who are in need. The HILDA survey identifies informal carers by asking respondents if they provide ongoing care or assistance to a family member with a long-term health condition, a disability or who is old. The self-administered follow-up questionnaire asks about respondents’ time commitment in a typical week to caregiving. These questions are used to identify informal carers and quantify their caregiving commitments (in hours per week). We classified informal caregiving responsibilities as non-caregiver, lighter caregiving (< 5 hours/week), moderate caregiving (5–19 hours/week) and intensive caregiving (20 or more hours/week) following a prior study [22].

2.3.3 Control Variables

We included potential confounders that are important correlates of workplace productivity based on previous research. The socio-demographic covariates included in the study are age, gender, relationship status, education, annual household disposable income, Indigenous status and region of residence. We have included childcare use as an exposure variable in the regression analyses. Childcare use is constructed from three variables (childcare used for any not yet at school child, childcare used for any school-age child during term and childcare used by any school-age child during school holidays). The responses to these questions were recorded in binary form (not used and used). If a respondent answered ‘used’ to any of the three questions, we considered them to have utilised childcare. The current study included two key measures of health-related characteristics (weight and disability status) and three health-related behavioural characteristics (smoking, alcohol consumption and physical activity). Moreover, the authors also considered a wide range of job-related characteristics (farm size, employment contract, occupation, supervisory responsibilities, union membership, paid holiday leave, sick leave and overall job satisfaction) as the potential confounders. We provide a detailed description of the control variables in Table 5 (Appendix).

2.4 Estimation Strategy

2.4.1 Conceptual Framework

We begin our conceptual framework by noting that workers who provide informal care owing to the presence of a disabled family member must decide whether to go to work or stay at home. This decision may also be influenced by a variety of other factors. For example, workers might become unwell and opt not to report to work. This process will be determined by the health and health-related behaviour of the individual. In addition, workplace policies (e.g. if the employee is eligible for sick leave or sick pay, employment contract, etc.) and work culture (kind of vocation, whether the employee has a supervisory relationship, farm size, etc.) will also influence this choice. Hence, the decision to report to work or not depends on the costs and benefits of the alternatives available to the workers.

In making a work participation choice, we assume workers build expectations about their on-the-job productivity and their firm’s productivity loss. Workers might stay at home if they expect reduced workplace productivity owing to informal caregiving duties. On the other hand, if the worker’s absence causes considerable productivity losses for the firm and the worker’s employment is at risk, they are more likely to report to work despite informal caregiving commitments. Of course, the worker might fail to recognise the expectations, in which case neither absenteeism, presenteeism nor work hour tension would emerge.

In model estimation using our conceptual framework, three essential observations should be emphasised. First, the productivity loss associated with informal caregiving is contingent on several factors, not all of which may be observable in our data. Consequently, estimation error may be present in the analysis. Second, absenteeism, presenteeism and work hour tension may be conceptually related, but we cannot estimate them simultaneously. This is because we do not observe every incidence of informal caregiving scenarios with a matching work attendance decision. Lastly, it is unclear whether absenteeism, presenteeism and work hour tension are inherently negative outcomes from the standpoint of either the employer or the employee. Continuing to work when excessively burdened with informal caregiving obligations, for instance, may not result in a loss of productivity provided the work environment is accommodating enough to meet the worker’s circumstances. Similarly, the firm’s productivity loss might well be minimised if employees with caregiving responsibilities choose to stay at home rather than report to work.

2.4.2 Model Estimation: Absenteeism and Working Hour Tension

According to this study, absenteeism \(({A}_{it})\) is measured by the number of absent days in the last 12 month period, which is a count variable. Similarly, person i experience working hour tension (number of hours like to work – number of hours actually worked) at time t is denoted as \({W}_{it}\), which is also count variable that. As a result, we estimate random-effects Poisson regression models of the following form [23]:

$$\mathrm{Pr} \left({A}_{it}={a}_{it}|{x}_{it}\right)= \frac{{e}^{-{\lambda }_{it}}{{\lambda }_{it}}^{{a}_{it}}}{{a}_{it}! }$$
(1)

and

$$\mathrm{Pr} \left({W}_{it}={w}_{it}|{x}_{it}\right)= \frac{{e}^{-{\lambda }_{it}}{{\lambda }_{it}}^{{w}_{it}}}{{w}_{it}! }$$
(2)

Where, \({\lambda }_{it}= {exp}^{{x}_{it}\beta }\); \({x}_{it}\) is a vector of control and exposure variables (informal caregiving \({IC}_{it}\), childcare-related characteristics \({C}_{it}\), health-related characteristics and behaviour \({H}_{it}\), job characteristics \({JC}_{it}\) and demographic \({X}_{it}\)).

Thus, the random-effects Poisson model is specified in our study in the following form:

$$\mathrm{log}\left({\lambda }_{it}\right)={\gamma }_{0}+ {{\gamma }_{1}IC}_{it}+ {{\gamma }_{2}C}_{it}+ {\gamma }_{3}{H}_{it}+{\gamma }_{4}{JC}_{it}+ {\gamma }_{5}{X}_{it}+{\overline{X} }_{i}\eta + {\alpha }_{i}+{\varepsilon }_{ij}$$
(3)

The primary parameter of interest to be estimated is \({\gamma }_{1}\) indicating the impact of informal caregiving on productivity loss.

2.4.3 Model Estimation: Presenteeism

Since our measure of presenteeism is a binary indicator variable, we use a random-effects logistic model to examine the link between productivity loss and informal care. Consequently, we postulate that the following logistic functional forms may be employed to characterise presenteeism associated productivity losses resulting from informal caregiving in the family:

$${P}_{it}^{*}=\Lambda ({IC}_{it}, {C}_{it}{,H}_{it},{JC}_{it},{X}_{it}, {\alpha }_{i}, {\varepsilon }_{it})$$
(4)

Where \({P}_{it}^{*}\) is the person \(i\mathrm{^{\prime}}\)s continuous latent response, reflecting the propensity to experience presenteeism at time \(t\). The independent variable \({IC}_{it}\) is an ordinal indicator on the level of informal caregiving, \({C}_{it}\) indicates childcare use, \({H}_{it}\) is a vector of health-related characteristics and behaviour, \({JC}_{it}\) is a vector of job characteristics, \({X}_{it}\) is a vector of demographic controls, \({\alpha }_{i}\) is unobserved time-invariant heterogeneity and \({\varepsilon }_{it}\) is the error term. The subscript \(it\) refers to person \(i\) and time \(t\) respectively.

Since \({P}_{it}^{*}\) is unobserved, the following is the observed response:

$${P}_{it}=\left\{\begin{array}{c}1 , if \; {P}_{it}^{*}>0\\ 0, otherwise\end{array}\right.$$
(5)

The random-effects logistic model (see Wooldridge, J.M., 2010 for details) for individuals experiencing presenteeism is as follows [24]:

$$log\left(\frac{\mathrm{Pr}\left({P}_{it}=1\right)}{1-\mathrm{Pr}\left({P}_{it}=1\right)}\right)= {{\beta }_{1}IC}_{it}+ {\beta }_{2}{H}_{it}+{\beta }_{3}{JC}_{it}+ {\beta }_{4}{X}_{it}+{\alpha }_{i}$$
(6)

With \({\alpha }_{i \sim } N(0,{\sigma }_{\alpha }^{2})\) ; \({\beta }_{i}\) \(\left[i=1\dots 4\right]\) is a vector of parameters to be estimated.

As with our analysis of absenteeism, presenteeism and work hour tension, the main parameter of interest \({\beta }_{1}\) will capture the impact of productivity loss due to informal caregiving variable \({IC}_{it}\). We expect higher odds of productivity loss for a higher level of informal caregiving.

3 Results

3.1 Descriptive Analyses

Table 1 presents the socio-demographic, health and employment characteristics of the analytic sample at baseline, the final wave and the pooled in all waves. The majority of the sample were middle-aged adults (45.82%), partnered (63.89%), non-Indigenous (97.85%) and lived in major cities (69.89%). Table 2 also shows that 23.87% were obese, 17.41% had a disability, 16.41% currently smoked, 86.38% consumed alcohol and 64.87% did not engage in the recommended amount of physical exercise. A larger proportion of individuals worked in a small firm (42.69%), had a permanent employment contract (67.85%), were professionals (26.01%), had no supervisory duties (56.14%), were not union members (77.61%), had paid holiday leave (73.58%) and had paid sick leave (73.89%). The average level of job satisfaction among workers is 7.73, with a standard deviation of 1.56 (pooled data).

Table 3 displays the distribution of the outcome variables (absenteeism, presenteeism and working hour tension) and main variable of interest (informal caregiving) at the baseline (2011), final (2021) and pooled across all waves. The mean level of absenteeism is 39.31 days with a substantial standard deviation of 73.53 days. Among the study participants, nearly one-third reported presenteeism (31.73%). The mean working hour tension is 1.78 with a standard deviation of 5.15 h. Table 2 also indicates that 4.35%, 3.10% and 0.66% of participants provided lighter, moderate and intensive informal caregiving, respectively (pooled data).

Table 1 Distribution of the analytic sample (socio-demographic, health and job-related characteristics): baseline, final and pooled across all waves
Full size table
Table 2 Distribution of the outcome and main variables of interest: baseline, final and pooled across all waves
Full size table
Table 3 Results from the longitudinal random-effects Poisson regression (models 1 and 3) and random-effects logistic regression models (model 2)
Full size table

Figure 2 displays the mean absence/leave days at work by the informal caregiving responsibilities of the study sample. We found that the average number of absence days is highest among those with intensive caregiving obligations in all years. For example, average absence/leave days of the respondents with intensive caregiving duties is 54.54 days, followed by moderate (41.27 days) and no caregiving responsibilities (40.33 days) in 2021.

Fig. 2
figure 2

Mean absence/leave days over the informal caregiving role

Full size image

Figure 3 shows a breakdown of the presenteeism rate among the study population according to their level of informal caregiving responsibilities. The graph indicates that participants with no caregiving role had the lowest rate of presenteeism in all years. The rate of presenteeism in the most recent year (2021) was highest among the respondents who provide intensive care (64.41%), followed by those with a moderate (46.55%), a lighter (46.54%) and a non-caregiving role (35.87%).

Fig. 3
figure 3

Rate of presenteeism over the informal caregiving role

Full size image

Figure 4 outlines the mean working hour tension by the study sample’s informal caregiving responsibilities. As expected, workers with intensive caregiving responsibilities had the highest working hour tension rate in most studied years. In 2021, we found that mean working hour tension is highest among person who had moderate caregiving duties (2.53 h), followed by intensive (1.92 h) and lighter caregiving duties (1.46 h).

Fig. 4
figure 4

Mean working hour tension over the informal caregiving role

Full size image

3.2 Regression Modelling

Table 3 presents the adjusted regression analyses, which were based on 97,521 (model 1) and 96,818 (models 2 and 3) person–year observations. After adjusting for socio-demographic, health and job-related covariates, we found that the rates of absence/leave at work are 1.03, 1.02 and 1.19 times higher in case of lighter [IRR (95% CI): 1.03 (1.02, 1.03)], moderate [IRR (95% CI): 1.02 (1.02, 1.03)] and intensive caregiving duties [IRR (95% CI): 1.19 (1.18, 1.21)], respectively, compared with no caregiving duties. The regression results also indicated the predicted change in the odds of presenteeism associated with different levels of informal caregiving (model 2). The results indicate that all things being equal, the odds of reporting presenteeism are higher amongst workers with intensive caregiving responsibilities [OR (95% CI): 1.78 (1.41, 2.25)], followed by lighter [OR (95% CI): 1.24 (1.13, 1.37)] and moderate [OR (95% CI): 1.23 (1.10, 1.38)] caregiving responsibilities compared with non-caregiving counterparts (model 2). Our results also show that the rates of having working hour tension are significantly greater in workers with intensive caregiving responsibilities [IRR (95% CI): 1.19 (1.11, 1.27)], followed by moderate [IRR (95% CI): 1.15 (1.11, 1.19)] and lighter [IRR (95% CI): 1.08 (1.04, 1.12)] caregiving responsibilities than their peers without caregiving responsibilities given other covariates reference categories are held constant (model 3).

3.3 Informal Caregiving—Attributable Costs of Absenteeism

Table 4 presents the costs of lost productivity in terms of absenteeism due to informal caregiving. The result shows that, on average, an individual with lighter, moderate and intensive caregiving role had incurred AUD 276.13, AUD 246.81 and AUD 1927.16, respectively, in absenteeism costs annually compared with their counterparts without caregiving duties (Appendix Tables 6 and 7 contains further details).

Table 4 Costs of absenteeism attributed to informal caregiving according to gender
Full size table

4 Discussion

4.1 Key Findings

Informal care is crucial to the well-being of the elderly and individuals with long-term health conditions or disabilities. The services provided by informal caregivers are vital to society. However, the intensity of caregiving may negatively impact the workplace productivity of caregivers. The primary goal of this study is to investigate the indirect costs of informal caregiving in terms of absenteeism, presenteeism and working hour tension. We utilised data from 11 successive waves of the HILDA survey conducted between 2011 and 2021. Longitudinal random-effects logistic regression and negative binomial regression models were employed to assess the influence of informal caregiving on workplace productivity among working-age Australians. We observed that being an informal carer significantly reduced productivity at work. More specifically, intensive caregiving is linked to a higher number of days missed from work. We also found that all types of informal caregiving responsibilities are associated with an increased likelihood of presenteeism at work. Further, our findings demonstrated that moderate informal caregiving involvement could increase working hour tension.

Prior research on the burden of informal caregiving has concentrated mostly on the relationship between caregiving and labour force participation. Relatively little is known regarding the extent to which workplace productivity of the caregivers’ changed by their caregiving responsibility. This research contributes to the current body of knowledge by pointing out the indirect costs incurred by employees who have informal caregiving duties. This study considered three measures of productivity (absenteeism, presenteeism and working hour tension) while examining the impact of informal caregiving burden on the workplace productivity of caregivers. The present study is one of the few that employed longitudinal data to investigate how informal caregiving affects the job productivity of caregivers. Using simple cross-sectional data while exploring the relationships between informal caregiving and lost productivity may provide a biased estimate due to omitted variable bias. Using longitudinal data, this study could keep track of the same people over time and see how their productivity changed as their caregiving responsibilities changed. This makes the present study unique among previous research on the topic.

We found evidence that employees providing informal care were more likely to be absent from work, which matches the results of two recent longitudinal studies in European countries [14, 19]. Men and women are equally likely to be absent from work due to caregiving responsibilities in the Netherlands [19]. However, informal caregiving increases sickness absence exclusively among women in France, Finland, and the UK [14]. A significant finding of the current study is informal caregiving is linked to a higher rate of presenteeism. This confirms the results of a previous study showing that informal caregiving is linked to reduced productivity while at work [4]. A recent study in the USA also found that nearly one in four employed caregivers has reduced productivity while at work [12].

Another key finding of our study is that informal caregivers experience greater working hour tension (desired working hours are greater than the actual working hours) than their counterparts, which is in line with several previous studies. A recent longitudinal study in the Netherlands confirmed that taking up informal caregiving responsibilities reduce women’s working hours [19]. The provision of informal caregiving had a negative impact on the number of hours worked in Japan [17] and a Chinese study discovered that intensive caregiving considerably reduces weekly working hour [11]. Further, several US studies concluded that females with informal caregiving provision worked fewer hours than their counterparts [13, 15]. However, a study conducted in the USA indicated that caregiving for an older parent does not prevent married women from participating in paid employment or reducing the number of hours worked [16].

Previous research has uncovered the mechanisms by which informal caregiving impacts the workplace productivity of caregivers. For example, an Australian study revealed that one in seven working individuals with caregiving responsibilities desire but cannot utilise flexible working hour, shift work and work-from-home possibilities [18]. Due to the 24 hour day, informal caregiving responsibilities may discourage individuals from entering the workforce. Moreover, the increased absenteeism rates among informal caregivers may be attributable to their added stress and workload [19]. Informal care provision leads to negative health outcomes arising from physical and mental strain [6]. Studies in the Netherlands, Germany, European countries, and the UK have shown that informal care provision has negative effects on the physical and mental health of informal caregivers over the long term, especially for high-intensity caregivers [6, 25,26,27,28], and this could have a negative effect on productivity at work (presenteeism). The emotional strain of caregiving makes it difficult for carers to effectively carry out their job responsibilities while at work. Therefore, caregivers may cut back their working hours or be absent due to unpredictable caregiving responsibilities [29].

4.2 Implications for Policy and Practice

The substantial indirect costs of informal caregiving cannot be overlooked. The present study’s findings have implications for formulating public policies regarding informal caregiving. Due to an ageing population, working-age individuals in developed countries, particularly women, face the challenge of balancing caregiving with paid work. Since the working-age population offers a large portion of informal care, it is crucial to understand how it influences workplace productivity to create regulations that assist caregivers. Substitution between formal and informal care may be an effective strategy for enhancing workplace productivity. The results suggest that policymakers should prioritise increasing access to publicly funded or subsidised formal care. This suggestion is in line with Australia’s recent reforms that include the introduction of a National Disability Insurance Scheme [30] and Nursing Home Care Reforms [31] aiming to extend formal care. If more formal care options were available, informal caregivers might have less caregiving burden and be more productive at work. Time spent providing care is associated with the quality of life of both the caregiver and the person being cared for. It has been demonstrated that an improvement in the care recipient’s quality of life reduces the time spent providing care and improves the caregivers’ quality of life [32]. Evidence also suggests that public long-term care insurance programme helps to lessen the burden on primary caregivers [17]. In addition, this study emphasises the necessity for care-friendly workplace regulations that will protect caregivers from productivity loss and allow them to continue working healthily. There is evidence that flexible time and reduced working hour opportunity result in more hours of informal caregiving [10].

4.3 Role in Health Technology Assessment (HTA)

Economic evaluation is the comparative analysis of alternative courses of action in terms of costs and effectiveness. Our results provide evidence that informal caregiving is associated with productivity loss. Insight into the costs of informal caregiving is helpful for efficient resource allocation. Economic evaluations of interventions to improve the outcomes of children, the disabled or the elderly should include the cost of caregiving when measuring the cost effectiveness of an intervention from the societal perspective. These costs include productivity loss (not able to work, reduced work hours, less ability to take higher income jobs), as well as travel and support costs. Our findings will serve as an input for future economic evaluations regarding productivity loss associated with informal caregiving. Alongside the development of techniques to integrate informal care into economic evaluation, policymakers are beginning to recognise the significance and necessity of incorporating informal care into policy decisions [33]. Several earlier studies evaluated interventions that jointly or primarily targeted carers [34,35,36,37]. For example, a study conducted in Australia demonstrates the costs and benefits of implementing an evidence-based reablement program that aims to enhance the function and capability of dementia patients [34]. According to the findings of the study, dementia patients and their carers endure the costs, while the Australian health and social care system benefits the most from the implementation of the programme. Their findings suggest that decision-makers in Australia should offer financial incentives to assist dementia patients and their carers to engage in reablement programs in Australia [34].

4.4 Limitations and Avenues for Further Research

This study has two major limitations that need to be acknowledged. First, this study’s unbalanced longitudinal research design prevents drawing any conclusions about the causal effects of informal caregiving on workplace productivity. Second, the study’s findings could be susceptible to bias due to the use of self-reported data on time spent for informal caregiving. There is evidence that if participants felt social pressure, they would overstate their readiness to work extra hours [29]. Apart from these, there are limitations with how each of the key dependent variables are measured. For example, we measured presenteeism through the SF-36 questionnaire which is not exclusively designed to assess presenteeism. Widely utilised instruments, such as the Work Productivity and Activity Impairment Questionnaire (WPAI), the Health and Work Performance Questionnaire (HPQ) and the Work Limitations Questionnaire (WLQ), may be able to analyse absenteeism and presenteeism behaviour of respondents more accurately. Given these caveats, the authors call for a well-designed cohort study to determine the causal effect of informal caregiving on workplace productivity. Besides, future studies may point out the increasing rate of switching to part-time employment from full time, and job losses due to informal caregiving.

5 Conclusion

Informal caregiving is demanding and challenging. The demand for informal care is anticipated to increase worldwide due to the ageing population. Therefore, knowing the impact of informal caregiving on workplace productivity is essential. This study provides new evidence on the relationship between informal caregiving and workplace productivity, focusing particularly on absenteeism, presenteeism and working hour tension. We found that the provision of informal caregiving is adversely associated with the workplace productivity of caregivers in terms of increased absenteeism, a greater chance of presenteeism and higher working hour tension. The negative effect of informal caregiving on health, and adverse labour market outcomes continues to be a major concern in most developed countries as they rely on informal care to reduce the burden on the health system. Our findings provide evidence to inform the design and effective implementation of policies that improve the productivity of informal caregivers at work. Policymakers may respond to the excessive demand for informal care by increasing the availability of paid care. However, the impact of informal caregiving on workplace productivity must be well understood before implementing the policy.