Are baby-boomers healthy enough to keep working? Health as a mediator of extended labour force participation.
Buckley, Jennifer ; O'Dwyer, Lisel ; Tucker, Graeme 等
Introduction
Structural ageing of Australia's population means that
baby-boomers (born 1946-65) are an important group for policy. They form
an unusually large cohort; their socio-demographic characteristics
differ substantially from previous generations; and the socio-economic
context in which they will age has altered significantly over recent
decades (Buckley 2008).
One important impact of structural ageing is a reduction in
forecast economic growth (Treasury 2010: ix). This change will stretch
fiscal capacity, as expenditure related to aged care, pension provision
and health will rise as the population ages (Productivity Commission
2005: xii). Interventions to manage these pressures stem from the three
Ps: population, productivity and participation (Productivity Commission
2005; Treasury 2010). Increasing total labour participation rates is
seen as critical as the aged dependency ratio will fall from 5:1 today
to a projected 2.7:1 in 2050 (Treasury 2010: viii), due to fewer new
entrants to the labour market and more workers aged 55 and over who are
more likely to exit the work force or work part-time (Treasury 2010a:
11; Access Economics 2001).
The participation component of the three Ps framework involves
re-engaging workers with low or no labour force participation and
implementing policies to encourage older workers to delay their exit
from the workforce and facilitate part-time work in retirement
(Productivity Commission 2005; Treasury 2010). In terms of overall
participation rates, a large 10 percentage point increase in the
participation rate of male workers aged 55 years and over would yield at
most a 2 percentage point increase in overall participation (Treasury
2002: 28). However, this increase would still have significant fiscal
benefits in relation to government expenditure on age pensions and
private retirement incomes (Treasury 2002; Casey et al. 2003: 11;
Productivity Commission 2005). In addition, as work has been shown to
offer both health and social benefits, extended workforce participation
has the potential to improve health outcomes through facilitating active
engagement, thus reducing government health expenditure and contributing
to better quality of life for older people (Aquino et al. 1996; Dave et
al. 2008).
The baby-boomer cohort currently forms the core of the mature age
workforce; yet most work-related research focuses on older workers per
se and does not differentiate between baby-boomers and preceding
cohorts. The terms 'baby-boomers', 'mid-life',
'mature' and 'older' are sometimes used
synonymously. This means that many studies are interpreted as
representing baby-boomers, even when the study samples do not fit within
conventionally defined dates for the Australian baby boom (1946-1965;
1946-1961). (1)
Accurate definition of cohorts and age groups is important, as
factors influencing health, lifestyle choices, workability and work
decisions of individuals from different cohorts vary according to the
social context prevailing when they were born and as they matured (Ryder
1997). Cohort effects are particularly important given the dramatic
changes in work participation by age since 1970. Table 1 shows
significant falls in male participation and substantial increases for
females in all older age groups. This reflects both changing
socioeconomic conditions and the rise of the dual income household as
well as improved opportunities for women in employment and education
since the 1970s. A similar trend is evident in the United States (Poulos
& Nightingale 1997: 16).
Research on the health of older Australians has focused on
disability and disease prevalence and the 'burden' on the
health system (AIHW 2006; Adams et al. 2008; Atlantis et al. 2009) while
research on extending labour participation focuses on re-skilling,
re-training, and workplace management (Dawe 2009; Lawton Smith 2009; De
Luna et al. 2010). However, little Australian research examines the
relationship between health and workforce participation in older
workers. In addition, most of the data sets available to Australian
researchers use self-report measures of chronic conditions and few
contain clinically measured data (Gruszin & Szuster 2010). The
effect of health on labour force participation has generally been
assessed using summary health indicators such as self-rated health
(Maurer et al. 2007), or by examining the impact of a single chronic
condition such as diabetes (Brown et al. 2005) or mental health
conditions (Butterworth et al. 2006). Relatively few Australian studies
have compared the effects of different chronic conditions on labour
force participation or examined the effect of multiple conditions.
This paper explores associations between health and employment
status in a sample of South Australian baby-boomers. The aim is to
provide insights into how baby-boomers' health status might mediate
the success of work extension policies. This will provide a health
perspective for work-related policies designed to retain or re-engage
mature age workers. However, as health and work are both embedded in a
social context we also examine how social change has influenced the
labour force participation of baby-boomers and how this varies within
the cohort. We address the following research questions:
* How does labour participation vary within the baby-boomer cohort
and what are the factors associated with different levels of
participation?
* What are the associations between health and employment status
and do these vary by socio-demographic characteristics?
Literature review
Social change, policy change and labour market participation
As different age cohorts have experienced social, economic and
political changes at different stages in their life course, labour
market participation is also likely to vary by cohort. Although there is
a substantial literature on the workforce participation of mature age
workers, research describing patterns of labour force participation
within the baby-boomer cohort is sparse. One exception is NATSEM's
study which used HILDA data to show that men were more likely to be in
full-time work and women to work part-time, although in the oldest
subgroup of baby-boomers (aged 60-64) gender differences diminished,
with part-time work becoming more common for men nearing retirement
(Gong & McNamara 2011). Just over a third of all baby-boomers worked
a standard week (35-40 hours) but men were more likely to work long paid
hours (50 or more), while women were more likely to work fewer paid
hours.
Poor health and labour market participation of ageing baby-boomers
Improvements in population health status are often cited to justify
increasing the age for receiving the Age Pension and superannuation.
However, assumptions that baby-boomers will have better health than
their predecessors are generally based on mortality data and longer life
expectancies (McDonald 2011: 34) rather than on empirical evidence of
actual health status.
There is no consistent Australian evidence of either the
compression or expansion of morbidity. Less severe disability has
expanded slightly (AIHW 2006; Mathers 2007) while Schofield and
colleagues (2007) used Australian National Health Survey (NHS) data to
show that the proportion of persons aged 40-64 years reporting five or
more long-term conditions has increased from 25 to 32 per cent between
1995 and 2005. Atlantis and colleagues (2009) reported significant
increases in the prevalence of diabetes, high cholesterol and high blood
pressure for those aged 60 and over between 1989-90 and 2004-05. A third
study comparing persons aged 53-62 in the 1989-90 NHS with baby-boomers
aged 53-62 in the 2008-09 NHS found that the proportion of baby-boomers
with obesity, diabetes, hearing loss, multiple conditions, and asthma
was at least double that of their predecessors, while there were also
substantial increases in arthritis, migraine, back problems, high
cholesterol and alcohol risk (Buckley 2012).
Analyses of the 2001 and 2004-05 NHS (Zhang et al. 2009), and Wave
1 HILDA Survey data (Cai & Kalb 2009) indicated that the impact of
poor health on labour participation was stronger for workers aged 50 or
more than for younger workers. Not surprisingly, the impact of health on
labour participation was often much stronger for those with more than
one condition (Schofield et al. 2007; Schofield et al. 2008; Park 2010).
Zhang and colleagues (2009) found that, overall, the negative effects Of
chronic conditions on labour force participation were more significant
for males than females and that this was particularly the case for
mental health conditions. Conversely, several studies have found the
effect Of poor self-rated health (SRH) on labour force participation to
be greater for women than men (Schofield et al. 2007; Cai & Cong
2009; Gilfillan & Andrews 2010). Due to the use of different health
indicators and methods, it is difficult to compare findings regarding
the impact of chronic conditions on labour force participation
Retirement decisions and transitions
The decision to retire or to continue in the workforce is
influenced by a wide range of factors such as caring responsibilities,
financial status, health, job satisfaction, gender, the labour force
environment, lifestyle aspirations, government policy and the overall
economic context (Quine et al. 2006; Everingham et al. 2007). Factors
influencing the retirement decision vary by socio-economic status, which
is correlated with the capacity to accumulate financial resources for
retirement. Occupational type and work conditions contribute to factors
which have a more direct impact on the retirement decision, such as job
satisfaction, financial need and work/life stress (Knox 2003; Quine et
al. 2006). Those with more neutral feelings about their work may choose
to retire simply because they want more leisure and personal time (Onyx
& Baker 2006).
Although socioeconomic status (SES) is one way of framing the
retirement decision making process, Knox (2003) found that income alone
had little impact on exit decisions. Discounting the impact of external
factors, such as caring responsibilities, health problems, and
preferences of spouses, a synthesis of the literature suggests that the
retirement decisions of high income individuals who can afford to retire
are likely to be based on whether anticipated satisfaction from
increased leisure in retirement will exceed current job satisfaction. By
contrast, low SES individuals' work exit decisions are more likely
to be influenced by financial necessity than by choice (Quine et al.
2006; Everingham et al. 2007).
In the 1980s and 1990s, retirement was used as a mechanism to
alleviate pressure on the labour market and facilitate the transfer of
jobs across generations (Cornman & Kingson 1996). The comparatively
lower participation levels of older male baby-boomers, when compared to
those in the same age groups in 1970, can partly be traced to this
policy (Jackson et al. 2006: 316). Many of the incentives for early
retirement have been removed or are being reviewed as part of the
Australian Government' strategies to increase participation in
older workers (Productivity Commission 2005; Commonwealth of Australia
2008; 2009). The superannuation preservation age is being progressively
increased to 60 years; payments primarily directed at mature age
recipients, such as Widows Allowance and Mature Age Allowance have been
phased out; and the eligibility criteria for the Disability Pension has
been tightened (Carson & Kerr 2003; ABS 2010). In addition,
eligibility for the Age Pension is being progressively increased to 67
years for both men and women (Commonwealth of Australia 2009).
Although women's labour force participation has risen
dramatically in recent decades, interruptions caused by childbirth and
other family commitments make it difficult for women to secure
attractive part-time work opportunities in later life (Gilfillan &
Andrews 2010). There is substantial evidence indicating that education
is a strong mediator of workforce participation in older women (Quine et
al. 2006; Everingham et al. 2607, Gilfillan & Andrews 2010), with
better educated women likely to stay in the workforce longer even if
they have caring responsibilities (Warner-Smith et al. 2006). Financial
need (including the need to save for retirement) also drives many older
women to continue working, especially for divorced women or single
parents.
Poor health and retirement
The overall effect of poor health is to reduce labour force
participation, but it also precipitates early retirement. The most
commonly cited reason for retirement in the 45-54 age-group according to
the 2011 ABS Retirement and Retirement Intentions Survey was
ill-health/disability (men, 56 per cent; women, 28 per cent) (ABS 2011).
Although the primary reason given for retirement in the 55-64 age group
was access to retirement funds (48 per cent men; 33 per cent women), the
proportion citing ill health/disability (23 per cent men; 19 per cent
women) is still substantial. In addition, the 'access to retirement
funds' option simply tells us that those ticking this option
retired because they could, but it does not provide information about
the factors which influenced this decision.
Using cross-sectional data from the ABS 2003 Survey of Disability,
Ageing and Carers, Schofield and colleagues (2008. 448) found that 45.6
per cent of respondents aged 45-64 years who were not in the workforce,
had left due to a chronic health condition while Cai and Kalb (2004: 13)
used HILDA data to show that 39 per cent of men aged 50-64 and 21
percent of women aged 50-60 who were not employed, had ceased their last
job due to their own illness or disability.
Only Zhang and colleagues (2009) included mental health as a
chronic condition variable despite reports of an increase in prevalence
of high psychological distress in the Australian population (AIHW 2008).
This is an interesting omission, particularly as there is evidence to
indicate that mental health conditions tend to co-occur with other
chronic conditions (AIHW 2008, 2010; Holden et al. 2010; Zhang et al.
2009: 101). This suggests that the increase in multi-morbidity
consequent on an ageing population is likely to be accompanied by an
increase in psychological distress, which also has implications for an
ageing workforce.
The impact of the work environment on health
Although work generally has a beneficial effect on health (Price et
al. 2002; Winefield et al. 2002) it can also contribute to stress,
anxiety or depression, particularly where the workplace environment is
characterised by stressors such as high demand, low control, and poor
supervisor or co-worker support (Michie & Williams 2003; Skinner
& Pocock 2008), with work environment stressors potentially leading
to a premature departure from the workforce (Butterworth et al. 2006;
Park 2010). Equally, it is well recognised that long work hours (48 or
more) are associated with more work-life conflict, poorer health and
negative impacts on social relationships (Caruso 2006; Pocock 2006). In
addition, the impact of the work environment on health has been shown to
differ by gender. Park (2010) found that for women, job strain (high
demand/low control) was a significant predictor of early retirement,
while for men, supervisor support in the workplace was more important.
Data and methodology
We draw on data from the North West Adelaide Health Study (NWAHS),
a population based biomedical cohort study of a representative cross
section of urban-based adults (18-90 years old) (Grant et al. 2006;
2009). A key strength of this dataset is that it is based on medically
and clinically diagnosed conditions rather than self-reported data. The
conceptual framework of the NWAHS and the methods of selection have been
described previously (Grant et al. 2006).
The analysis of variations in labour force participation within the
cohort (Research question 1), and of the associations between health and
employment status (Research question 2), is based on a sample of
baby-boomers (1946-1965) drawn from Stage 2 (2004-06) of the NWAHS.
Stage 2 included clinical follow-up visits and additional surveys, and
was conducted between 2004 and 2006 (n=3,206; 79 per cent), with
baby-boomers constituting 36.8 per cent of the sample (n=1,179). As
previous research (Buckley 2011) shows that there are substantial health
and socio-demographic differences between older (1946-1955) and younger
(1956-1965) baby-boomers, our initial bivariate analyses were conducted
separately for each of these two groups. In 2007 when the employment
data was collected, older boomers were aged 52-61 and younger boomers
were aged 42-51. We also examined the data by gender overall and, in
some instances, within both older and younger baby-boomer sub-cohorts.
Variations in the level of labour force participation were assessed
through employment status (full-time; part-time; casual; unemployed;
retired; home duties; unable to work) and hours per week in paid work.
'Casual' workers may work full-time or part-time but do not
have paid leave entitlements. The 'unable to work' category
does not specify why the respondent was unable to work; hence
identification with this category does not necessarily imply illness but
could, for example, also relate to caring responsibilities. Variables
used to assess the socio-economic characteristics associated with
different levels of labour participation are listed in Table 4. The
independent variables used to assess associations between health and
employment status included: self-rated health (fair-poor;
good-excellent); clinically measured and/or doctor-diagnosed chronic
conditions; and multiple morbidity. Parameters for the chronic
conditions used in this analysis are included in Appendix 1. Several of
the conditions have been designated national health priority areas due
to their significant contribution to the national burden of disease
(AIHW 2010). Dependent variables included employment status and whether
respondents were in the labour force (ILF) or not in the labour force
(NILF).
Data were analysed using the Statistical Package for the Social
Sciences (SPSS) versions 15 and 17, and Pearson's chi-square test
was used to determine statistically significant differences. The
significance level for hypothesis tests was set at five per cent. A
multivariate logistic regression model was used to test for associations
between labour force participation and chronic conditions, and
socio-demographic characteristics. Models were fitted using backward
elimination of non-significant variables. Based on the bivariate
analyses, we considered the following variables for inclusion in the
model: income, pension status, age group or cohort, sex, living
arrangements, educational attainment, marital status, country of birth
and housing tenure. Chronic conditions considered for inclusion were:
depression, arthritis, diabetes, heart disease, chronic kidney disease,
asthma and chronic obstructive pulmonary disorder (COPD).
Income, pension status, chronic kidney disease and asthma were not
included in the model. Most of the variations in income and pension
status are explained by labour force participation; they are primarily
consequences and not indicators of labour force participation and so
were excluded from the model. Chronic kidney disease and asthma were
excluded as they had a significance probability greater than .25 and
hence did not meet the requirements set out by Hosmer and Lemeshow
(2000).
We checked the effect of using cohort (1946-1945 and 1956-1965),
five year age groups, and five year birth cohorts, as alternative
predictors in the model. Using Akaike's Information Criteria (AIC)
(Burnham & Anderson 2004), the best model was identified as the
model containing five year age groups. As age was not linear in the
logit, it could not be included in the model as a continuous predictor
and was therefore included in the model as a categorical variable.
All relevant two way interactions were included in the model, and
again, predictors were culled using backward elimination of
non-significant terms.
Results
Research Question 1--labour force participation
Table 2 sets out the employment status of males and females in each
baby boomer age cohort. It can be seen that older boomers (1946-1955)
were less likely to be working full-time and were more likely to be
retired, irrespective of gender. However, only older males were
significantly more likely to be in the unable to work category. Older
boomer males also approached significance for being more likely to be in
part-time employment compared to younger boomer males (p =.055).
Table 3 shows that just over one-third of all baby-boomers worked a
standard week (35-40 hours). Not unexpectedly, females worked far fewer
paid hours than males, with only 14.7 per cent working more than 40
hours per week (Table 3). However, there was little difference in the
hours worked by each cohort and although about five per cent more of the
younger cohort worked between 41-49 hours, the proportion working extra
long hours (50 or more per week) was substantial for both age cohorts
(17 per cent). Extra long hours were primarily worked by males (27 per
cent) rather than females (6.5 per cent).
Table 4 sets out the socio-economic characteristics associated with
different employment categories in older workers and shows a distinct
trend for social risk factors to increase in tandem with reduced
attachment to the workforce. Not surprisingly, both full-time and
part-time work are associated with more supportive living arrangements
(marriage, family with children) and more resources in terms of
education and income. Social disadvantage was most strongly associated
with being casual, unemployed or unable to work, with the indicators of
disadvantage increasing in that order. The majority of casual workers
were female and had only a secondary education. Casual workers were the
least disadvantaged of these three subgroups but were still
significantly more likely than full-time or part-time workers to be
renting and on a low income.
Those who were unemployed or unable to work were more likely than
full-time or part-time workers to: live alone; be single; have an income
of less than $20,000; be on a government pension; and to rent their
house.
The results suggest respondents in casual employment and those who
were unemployed or unable to work experienced both long term and
multiple disadvantage, with more of these respondents in rental housing,
with low levels of social support, and low incomes. The comparatively
small numbers in these categories mean that results from more fine
grained analysis cannot be regarded as definitive; however, bivariate
analysis of unpublished NWAHS data does suggest that the degree of
disadvantage associated with these employment categories is likely to be
mediated by gender.
Table 4 shows that the majority of respondents in the home duties
category were in the 1956-1965 age cohort and, as would be expected, the
majority were female. Respondents in this category were more likely to
have a secondary education at most and the proportion with a
certificate/diploma or tertiary education was substantially smaller than
for full-time or part-time workers. Figure 1 shows the distinct impact
of education on employment patterns for women, with the proportion of
women in the home duties category decreasing as education levels rise. A
substantially higher proportion of tertiary educated women were employed
full-time, with this particularly evident for older (1946-55)
baby-boomer women who may well have reached the empty nest stage.
[FIGURE 1 OMITTED]
Figure 2 shows that these trends are largely confirmed by Census
data. While there are some differences between Census and NWAHS data in
relation to levels of education and full-time and part-time employment,
the relationship between education and home duties is comparable.
[FIGURE 2 OMITTED]
Around 15 per cent of older boomers (1946-1955) had retired but
there were virtually no retirees in the younger (1956-1965) age group.
ABS Retirement Survey data collected in 2006-2007 (ABS 2008) shows a
similar trend.
The majority of retirees from the NWAHS were married and were more
likely to be living with a partner only and less likely to be living in
a family with children, reflecting the fact that nearly all of the
retired were earlier born baby-boomers (1946-1955) and hence more likely
to have reached the empty nest stage. The retired were more likely to
have an income of less than $20,000 per annum and 39.5 per cent were in
receipt of some sort of government pension. However, they were
substantially more likely to own their home outright, with this being
the case for 82.9 per cent of this group (Table 4).
[FIGURE 3 OMITTED]
Research Question 2
Figure 3 shows the percentage of baby-boomers in each employment
category (y axis) and the proportion in each category with either
good-excellent or fair-poor self-rated health (SRH) (x axis). Apart from
the home duties category, SRH tends to fall as work force attachment
decreases, with the full-time employed most likely to have
good-excellent SRH and those with an employment status of
'retired' or 'unable to work' most likely to have
fair-poor SRH.
Around 20 per cent of respondents had fair-poor SRH and the
majority of these (68.7 per cent) were in the labour force. However, the
proportion of younger (1956-1965) baby-boomers with fair-poor SRH in the
labour force (82 per cent) was substantially higher than the proportion
of older (1946-1955) boomers (55.1 per cent). Figure 4 shows that in
both age cohorts those with fair-poor SRH were less likely than those
with good-excellent SRH to be in the labour force. For the 1946-1955 age
cohort this did not vary by gender, however, in the younger cohort,
males with fair-poor SRH were significantly more likely than females to
be in the labour force (p=.008) (NWAHS 2011).
[FIGURE 4 OMITTED]
Analysis of the whole cohort shows that there were no significant
differences in employment status for respondents with asthma or kidney
disease (Table 5). However, diabetes, arthritis, COPD, poor self rated
health (SRH), and multiple conditions were all significantly associated
with being retired or unable to work. Arthritis was also significantly
associated with casual employment while those who were depressed were
significantly more likely to be unemployed, casual, or unable to work.
When adjusted for sex, the odds of not being in the labour force
(NILF) for older baby-boomers (1946-1955) were significantly higher for
those who had COPD, depression, diabetes and arthritis but not for those
with asthma, cardiovascular disease or kidney disease (Table 6).
However, for the younger cohort (1956-1965), the odds of being out of
the labour force were only significantly greater for those with
diabetes, cardiovascular disease and COPD.
When adjusted for age, the odds of not being in the labour force
were significantly higher for both males and females who had COPD, while
the odds for those with depression or arthritis were only significantly
higher for males, and the odds for those with diabetes were only
significantly higher for females.
The results from the logistic regression (Table 8) indicate that
workforce participation in the baby-boomer cohort was positively
associated with being younger, male, having a tertiary education, and
paying off a mortgage. Chronic conditions negatively associated with
workforce participation were diabetes and COPD. We also tested for
two-way interactions between sex and age and the chronic conditions, but
the only interaction of significance was between sex and arthritis. The
relationship between arthritis and non-participation in the workforce is
stronger for males than females.
Discussion
Variations in baby-boomers' labour force participation Our
analysis of workforce participation shows that just over one-quarter of
baby-boomers are marginally attached or not attached to the workforce.
Consistent with findings from NATSEM (Gong & McNamara 2011), we
found that only around one-third of employed baby-boomers worked a
standard week (35-40 hours) and that extra long hours (50 or more) were
worked by nearly 30 per cent of males, irrespective of age sub-cohort.
Consistent with the literature on retirement intentions and decisions
(Knox 2003; Jackson et al. 2006; Onyx & Baker 2006; Quine et al.
2006; Everingham et al. 2007), the results presented in Table 6 suggest
that early retirees can be broadly divided into two main subgroups,
those who have reasonable means and choose to retire early and those who
may not be particularly well off and retire involuntarily. Approximately
a quarter of retirees had a household income of $60,000 or more and, in
line with Onyx and Baker (2006), the desire for increased leisure may
well have been a key factor in their retirement decision. However,
around half of the retirees had a household income of $40,000 or less
and about 40 per cent were in receipt of a government pension,
suggesting that the labour force exits of individuals in this group were
involuntary. This view is supported by the fact that respondents were 61
or younger at the time of the survey, and hence were not eligible for an
Age Pension. Consequently, those on a government pension must have been
in receipt of some other form of assistance such as a Disability
Pension, Mature Age Allowance or Carer's Payment.
Variations in associations between health status and labour force
participation
Consistent with previous research we found a strong association
between poor health and reduced labour force participation. In line with
Cai and Kalb (2009) and Zhang (2009) we found this association to be
stronger for older baby-boomers (52-61 years).
The role of poor health in triggering an individual's exit
from the labour force is likely to be mediated by a wide range of
inter-related structural and individual level factors (Brown &
Vickerstaff 2011: 531-532). Not surprisingly, age is a key factor, with
our findings consistent with previous studies, such as AIHW (2009) and
NATSEM (Gong & McNamara 2011), which show a stronger association
between poor health and workforce exits for older (1946-1955)
baby-boomers than for younger (1956-1965) baby-boomers. The much higher
labour force participation of younger baby-boomers with fair-poor SRH is
likely to be influenced by two factors. First, most of the chronic
conditions examined in this study are age related and occur with
increasing frequency and severity as individuals grow older (AIHW 2010:
287). Therefore, younger baby-boomers are likely to be in the early
stages of chronic disease and may find work easier to manage than those
with more advanced conditions. Second, younger baby-boomers may have
greater motivation to stay in the workforce despite ill health for
several reasons: they have bad less time to accumulate the assets
necessary for a comfortable retirement (Buckley 2011: 229); they are
less likely to be able to access an alternative income such as
superannuation (Kelly 2009); and they are more likely to have children
living at home (NWAHS 2011).
Limitations and future research
This study has a number of limitations. First, the relatively small
cell sizes in the casual, unemployed, and unable to work categories
makes a detailed and robust analysis of gender differences within these
groups difficult. Second, the cross-sectional nature of the study means
that causality cannot be determined. However, with the exception of
depression, being out of the workforce is unlikely to be the cause of
the chronic conditions examined in this study. Finally, we do not know
from the available data why people are unable to work. The reasons may
include illness, disability or caring responsibilities.
Given that between 10 and 20 per cent of baby-boomers suffer from
either depression and/or some other form of psychological distress or
mental illness (Buckley 2011: 414), it will be important for future
research to explore the causal linkages between mental health conditions
and work, and how factors such as gender, psychological make-up and work
and home environments might mediate these linkages. Identifying gender
differences in the underlying causes of depression could usefully inform
initiatives aimed at reducing depression in older workers, given the
finding that males not in the labour force were more likely to be
depressed than females not in the labour force and females in the labour
force were more likely to be depressed than males in the labour force.
This is particularly the case if the cause of depression is linked to
having or not having work, or is linked to the way in which work is
experienced, or to issues associated with the work-home interface.
Although a cross-sectional design means that we cannot establish the
causal direction between depression and work, Winefield (1995: 196)
points out that causality is likely to work both ways; some individuals
are likely to become unemployed because they are anxious or depressed
while others may become psychologically distressed as a consequence of
losing their job. Previous research has found the financial stress and
social exclusion associated with precarious employment and unemployment
to be key contributing factors to depression (Winefield 1995: 187; Price
et al. 2002). This resonates with our data which shows that respondents
in casual work, or who were unemployed, experienced greater disadvantage
than those in secure employment, and they were also more likely to be
depressed.
Implications
As the health of the current cohort of older workers appears to be
worse than that of previous cohorts, the practicality of enforcing
participation through policies which make it difficult to retire is
questionable. Calls to further increase the access age for the Age
Pension and superannuation to age 70 (Daley et al. 2012: 53) may well
cause hardship and possibly contribute to poorer health in the long
term.
Baby-boomers' exits from the labour force due to ill health,
prior to reaching Aged Pension eligibility age of 65, have already begun
to occur. A large proportion (44.9 per cent) of those who rated their
own health as fair or poor in the cohort born between 1946 and 1955 are
not in the labour force, equating to 9.9 per cent of this group.
However, as the odds of not being in the labour force for those with
poor health increases with age, and the health of younger boomers
appears to be on a similar trajectory to that of their older
counterparts (Buckley 2011: 192), it is reasonable to expect that a
similar proportion of the younger age cohort will also leave the labour
force over the next decade.
Poor health in older workers and their consequent reduced
participation levels means that the expected fiscal benefits in relation
to government expenditure on age pensions, and in relation to private
retirement incomes, may be less than anticipated (Treasury 2002;
Productivity Commission 2005). The direct costs to the community are
also considerable, with Schofield and colleagues estimating that reduced
labour participation due to poor health reduces GDP by around $12
billion per annum (Schofield et al. 2008). Individuals who retire
involuntarily due to poor health are also likely to have poorer outcomes
in retirement (Quine et al. 2007) and have had less opportunity to
accumulate the financial assets that cushion this stage of life.
An obvious policy implication arising from this study is the need
to use both primary and secondary prevention strategies to improve
health in older workers. Early diagnosis of chronic conditions should be
implemented and/ or maintained because timely care reduces morbidity and
improves function. For secondary prevention it will be important to
prioritise the development and uptake of effective self-management
programmes to minimise the impact of chronic conditions. Given the
substantial proportion of baby-boomers with fair-poor health who
continue to work, it: will be important to develop both primary and
secondary prevention programmes specifically for the workplace, as this
is a prime site for intervention. Workplace health initiatives would be
further facilitated by engaging outside expertise such as employer
representative organisations and allied health organisations. Overall,
there is a need to increase health literacy in this cohort as ABS data
(2006) show this to be lower in baby-boomers than in younger age groups.
Individual level strategies, such as diet and physical activity
interventions for weight loss; and cognitive behavioural therapy and/or
drug treatments for depression, will continue to be important (Schofield
et al. 2008: 449-450). We must also refine our knowledge of workplace
factors which influence health, and develop strategies which minimise
their negative effects and enhance their positive effects. For example,
certain workplace conditions, such as shift work, weekend work and
variable work schedules, have been shown to have a detrimental impact on
health while others, such as flexible work conditions, have a positive
effect (Costa et al. 2006; Magee et al. 2010). Flexible work conditions,
such as working from home and flexible hours, may be particularly
important in enabling chronically ill older workers to maintain their
participation and avoid further deterioration in their health. Future
research and future health initiatives must account for variations in
work strain, work-family conflict and work-life imbalance, and in health
status and early exits, by age and gender (Pocock et al. 2007; Park
2010; Buckley 2011).
Given the significant impact that chronic disease has on labour
participation and productivity (Schofield et al. 2008; AIHW 2009), and
that chronic disease is associated with age, it is in employers'
interest to create and maintain a healthy work environment. New health
promotion and prevention initiatives must be tailored to a variety of
work environments and to the needs of workers in diverse occupations.
There are still substantial gaps in our knowledge about the most
effective and efficient ways to intervene (Dugdill et al. 2008; Anderson
et al. 2009). However, the evidence suggests that healthy workplace
initiatives should focus on both the physical and the psychosocial work
environment (Siegrist et al. 2009). It will also be important to
continue to fund initiatives such as the 'Work life balance
strategy' developed by Safework SA (n.d.), which increases employer
awareness of how work policies, practices and demands impact on
employees' health, and which assist them to develop and negotiate
work arrangements that are health enhancing rather than detrimental.
Conclusion
Although the majority of baby-boomers report good-excellent health,
a significant minority do not. This is not surprising, given the high
proportion of boomers who are overweight or obese (74.8 per cent) and
who do not exercise at an adequate level (68.4 per cent) (Buckley 2011:
418). This paper has shown that, if current trajectories continue, poor
health will limit the proportion of baby-boomers who are capable of
extending their working lives or who are able to stave off retirement
until they can access the Age Pension, particularly as the minimum age
at which people become eligible for the Pension has recently been raised
to 67 years. If baby-boomers and subsequent generations are to work for
longer it will be necessary to improve our understanding of how work
environments and conditions influence health across the life course,
particularly given the marked changes to the nature of work and the
social context in which it occurs (Costa et al. 2004). This will need to
occur in tandem with the development, implementation and evaluation of
workplace health promotion and prevention initiatives if we are to
create working environments capable of facilitating better health and
improving and extending participation.
Acknowledgements
This research was supported by an Australian Research Council
Linkage grant (Grant No. LP0990065). Title: Australia's Baby-boomer
Generation, Obesity and Work--Patterns, Causes and Implications.
References
ABS (Australian Bureau of Statistics) (2003) South Australia's
Baby Boomers: A Profile, Cat. No. 4149.4.55.001
--(2006) Health Literacy, Australia, Cat. No. 4233.0, Canberra.
--(2008) Retirement and Retirement intentions, Australia July 2006
to June 2007, Cat. No. 62380.
--(2010) Australian Social Trends, Cat. no. 4102.0.
--(2011) Retirement and Retirement Intentions, Australia July 2010
to June 2011, Cat. No. 62380.
Access Economics Pty Ltd. (2001) Population ageing and the economy,
Canberra, Publications Production Unit (Public Affairs, Parliamentary
and Access Branch) Commonwealth Department of Health and Aged Care.
Adams, R., Tucker, G., Hugo, G., Hill, C. & Wilson, D. (2008)
'Projected future trends of hospital service use for selected
obesity-related conditions.' Obesity Research and Clinical
Practice, 2133-141.
AIHW (Australian Institute of Health and Welfare) (2006) Life
Expectancy and Disability in Australia 1988 to 2003. Cat. No. DIS 47.
--(2008) Australia's Health 2008, Cat. No. AUS 99.
--(2009) Chronic Disease and Participation in Work, Cat. No. PHE
109.
-- (2010) Australia's Health 2010, Cat. No. AUS 122.
Anderson, L.M., Quinn, T.A., Glanz, K., Ramirez, G., Kahwati, L.C.,
Johnson, D.B., Buchanan, L.R., Archer, W.R., Chattopadhyay, S., Kalra,
G.P., Katz, D.L. & Task Force on Community Preventive Services
(2009) 'The effectiveness of worksite nutrition and physical
activity interventions for controlling employee overweight and
obesity', American Journal of Preventive Medicine, 37 (4), 340.
Aquino, J., Russell, D., Cutronoa, C. & Altmaier, E. (1996)
'Employment status, social support, and life-satisfaction among the
elderly', Journal of Counseling Psychology, 43 (4), 480-89.
Atlantis, E., Lange, K. & Wittert, G. (2009) 'Chronic
disease trends due to excess body weight in Australia', Obesity
Reviews, 10 (5), 543-53.
Brown, H.S., Pagan, J.A. & Bastida, E, (2005) 'The impact
of diabetes on employment: genetic IVs in a bivariate probit',
Health Economics, 14 (5), 537-44.
Brown, P. & Vickerstaff, S. (2011) 'Health subjectivities
and labor market participation', Research on Aging, 33 (,5), 529-50
Buckley, J. (2008) 'Baby boomers, obesity, and social
change', Obesity Research & Clinical Practice, 2 (20), 73-82.
Buckley, J. (2011) Ageing in the 21st century -are baby boomers
prepared? A study of preparation for later life in a context of social
change, PhD thesis, University of Adelaide (unpublished)
http://library.adelaide.edu.au/ item/1740389
Buckley, J. (2012) Ageing baby boomers: tsunami or windfall, Fact
Sheet No. 1/Media Event, prepared on behalf of the Project Team,
Australia's Baby Boomer Generation: Obesity and Work--Patterns,
Causes, and Implications, Science Exchange, 26 September
http://www.adelaide.edu.au/ apmrc/research/
Burnham, K. P. & Anderson, D. R. (2004) 'Multimodel
inference: understanding AIC and BIC in Model Selection',
Sociological Methods and Research, 33(2), 261-304.
Butterworth, P., Gill, S. C., Rodgers, B., Anstey, K.J., Villamil,
E. & Melzer, D. (2006) 'Retirement and mental health: analysis
of the Australian national survey of mental health and well-being.'
Social Science & Medicine, 62 (5), 1179-91.
Cai, L. & Cong, C. (2009) 'Effects of health and chronic
diseases on labour force participation of older working-age
Australians', Australian Economic Papers, 48 (2), 166-82.
Cai, L. & Kalb, G. (2004) 'Health status and labour force
participation: Evidence from the HILDA data', Melbourne Institute
Working Paper No. 4/04, Melbourne Institute of Applied Economic and
Social Research.
Carson, E. & Kerr, L. (2003) "'Stakeholder
welfare" and the "Pivot Generation": the challenge of
policy shifts and intergenerational dependencies for Australian
baby-boomers', Just Policy, 29, 3-14.
Caruso, C. (2006) 'Possible broad impacts of long work
hours', Industrial Health, 44 (4), 531-36.
Casey, B., Oxley, H., Whitehouse, E., Antolin, P., Duval, R. &
Leibfritz, W. (2003) 'Policies for an ageing society: recent
measures and areas for further reform', OECD Economics Department
Working Papers, No. 369, Paris, OECD Publishing.
Cornman, J. & Kingson, E. (1996) 'Trends, issues,
perspectives and values for the aging of the baby boom cohorts',
Gerontologist, 36 (1) 15-26.
Costa, G., Akerstedt, T., Nachreiner, E, Baltieri, E, Carvalhais,
J., Folkard, S., Dresen, M.F., Gadbois, C., Gartner, J., Sukalo, H.G.,
Harma, M., Kandolin, I., Sartori, S. & Silverio, J. (2004)
'Flexible working hours, health, and well-being in Europe: Some
considerations from a SALTSA project', Chronobiology International,
21 (6), 831-44.
Costa, G., Sartori, S. & Akerstedt, T. (2006) 'Influence
of flexibility and variability of working hours on health and
well-being', Chronobiology International, 23 (6), 1125-37.
Commonwealth of Australia. (2008) Ageing and Aged Care, Canberra,
Commonwealth of Australia, [Online]: http://www.health.gov.au/internet/
main/publishing.nsf/Content/BFE-46F21A3241ECBCA2574BE001A6E06/$
File/Ageing_and_Aged_Care.pdf
--(2009) Secure and Sustainable Pensions, Canberra, Commonwealth of
Australia http://www.aph.gov.au/budget/2009-10/content/glossy/pension/
download/pensions_overview.pdf
Daley, J., McGannon, C. & Ginnivan, L. (2012) Game-changers:
Economic reform priorities for Australia, Melbourne, Grattan Institute
http://grattan. edu.au/static/files/assets/bc719f82/Game_Changers_Web.pdf
Dave, D., Rashad, R. & Spasojevic, J. (2008) 'The effects
of retirement on physical and mental health outcomes', Southern
Economic Journal, 75 (2), 497-523.
Dawe, S. (2009) 'Older "workers and VET', Series: At
a Glance, Canberra, National Centre for Vocational Education Research.
De Luna, X., Stenberg, A. & Westerlund, O. (2010) Can adult
education delay retirement from the labour market? Working Paper 2010:
2, Uppsala, Institute for Labour Market Policy Evaluation.
Dugdill, L., Brettle, A., Hulme, C., McCluskey, S. & Long, A.E
(2008) 'Workplace physical activity interventions: a systematic
review', International Journal of Workplace Health Management, 1
(1), 20-40.
Everingham, C., Warner-Smith, P. & Byles, J. (2007)
'Transforming retirement: re-thinking models of retirement to
accommodate the experiences of women', Women's Studies
International Forum, 30 (6), 512-22.
Gilfillan, G. & Andrews, L. (2010) Labour force participation
of women over 45, Productivity Commission Staff Working Paper, Canberra.
Gong, H. & McNamara, J. (2011)Workforce participation and
non-participation among baby-boomers in Australia: a profile from HILDA
data, Canberra, NATSEM, University of Canberra.
Grant, J., Chittleborough, C., Taylor, A., Dal Grande, E., Wilson,
D., Phillips, A., Cheek, J., Price, K., Gill, T. & Ruffin, R. (2006)
'The North West Adelaide Health study: detailed methods and
baseline segmentation of a cohort along a chronic disease
continuum', Epidemiologic Perspectives & Innovations, 3 (4),
1-14.
Grant, J. F., Taylor, A. W., Ruffin, R.E., Wilson, D.H., Phillips,
P.J., Adams, R.J.T., Price, K. & the North West Adelaide Health
Study Team (2009) 'Cohort profile: the North West Adelaide health
study (NWAHS)', International Journal of Epidemiology, 38 (6),
1479-86.
Gruszin, S. & Szuster, F. (2010) Literature review: review of
health status and labour force productivity and participation data with
regard to chronic disease, Adelaide, Public Health Information
Development Unit, the University of Adelaide.
Holden, L., Scuffham, P., Hilton, M., Vecchio, N. & Whiteford,
H. (2010) 'Psychological distress is associated with a range of
conditions affecting working Australians', Australian and New
Zealand Journal of Public Health, 34 (3), 304-10.
Hosmer, D. & Lemeshow, S. (2000) Applied Logistic Regression,
New York, John Wiley & Sons.
Jackson, N., Walter, M. & Felmingham, B. (2006), 'Will
Australia's baby boomers change their retirement plans in line with
government wishes?', paper presented to the 13th Biennial
Australian Population Conference 'Population, Policy and
Australia's Destiny, Adelaide, 5-8 December.
Kelly, S. (2009) Don't stop thinking about tomorrow: the
changing face of Retirement--the past, the present and the future,
Income and Wealth Report No. 24
http://www.natsem.canberra.edu.au/files/download?id=1011
Knox, G. (2003) 'Retirement intentions of mature age
workers', paper presented to the Australian Social Policy
Conference, University of New South Wales, Sydney, 9-11 July.
Lawton Smith, H. (2009) 'Skill shortages, demographic aging,
and training implications for skill-based economies', The
Professional Geographer, 61 (1), 59-69.
Magee, C., Caputi, P., Stefanic, N. & Iverson, D. (2010)
'Occupational factors associated with 4-year weight gain in
Australian adults', Journal of Occupational & Environmental
Medicine, 52 (10), 977-81.
Mathers, C. (2007) 'The health of older Australians'. In
A. Borowski, S. Encel & E. Ozanne (eds), Longevity and Social Change
in Australia, Sydney, UNSW Press.
Maurer, J., Klein, R. & Vella, E (2007) Subjective health
assessments and active labor market participation of older men: evidence
from a semiparametric binary choice model with nonadditive correlated
individual-specific effects, IZA Discussion Paper No. 3257, Bonn, The
Institute for the Study of Labor.
McDonald, P. (2011) 'Employment at older ages in Australia:
determinants and trends'. In T. Griffin & F. Beddie (eds.)
Older Workers: Research Readings, Adelaide, NCVER.
Michie, S. & Williams, S. (2003) 'Reducing work related
psychological ill health and sickness absence: a systematic literature
review', Occupational and Environmental Medicine, 60(1), 3-9.
NWAHS (North West Adelaide Health Study) (2011) NWAHS Stage 2
database, 2004-06. Adelaide, University of Adelaide, [unpublished data
analysed by Buckley].
Onyx, J. & Baker, E. (2006) 'Retirement expectations:
gender differences and partner effects in an Australian employer-funded
sample', Australasian Journal on Ageing, 25 (2), 80-3.
Park, J. (2010) 'Health factors and early retirement among
older workers', Statistics Canada, Cat no. 75-001-X.
Pocock, B. (2006) The Labour Market Ate My Babies. Work, Children
and a Sustainable Future, Sydney, Federation Press.
Pocock, B., Skinner, N. & Williams, P. (2(107) 'Work-life
in Australia: outcomes from the Australian Work and Life Index (AWALI)
', Adelaide, Centre for Work + Life, University of South Australia.
Poulos, S. & Nightingale, D.S. (1997) The Aging Baby Boom:
Implications for Employment and Training Programs, prepared for the U.S.
Department of Labor, Employment and Training Administration, Washington
DC, The Urban Institute.
Price, R., Nam Choi, J. & Vinokur, A. (2002) 'Links in the
chain of adversity following job loss: How financial strain and loss of
personal control lead to depression, impaired functioning, and poor
health', Journal of Occupational Health Psychology, 7 (4), 302-12.
Productivity Commission (2005) Economic Implications of an Ageing
Australia, Research Report, Canberra.
Quine, S., Bernard, D. & Kendig, H. (2006) 'Understanding
baby-boomers' expectations and plans for their retirement: findings
from a qualitative study', Australasian Journal on Ageing, 25 (3),
145-50.
Quine, S., Wells, Y., de Vaus, D. & Kendig, H. (2007)
'When choice in retirement decisions is missing: Qualitative and
quantitative findings of impact on well-being', Australasian
Journal on Ageing, 26 (4), 173-9.
Ryder, N. (1997) 'The cohort as a concept in the study of
social change'. In M. Hardy (ed.), Studying Aging and Social
Change: Conceptual and Methodological Issues, Beverly Hills and London,
Sage Publications. Safework SA (n.d.) Work life balance,
http://www.safework.sa.gov.au/ worklifebalance/wlb_home.jsp?id=100001
Schofield, D.J., Passey, M.E., Earnest, A., Gloor, I.C. &
Shrestha, R. (2007) 'Are we getting healthier as we grow
older?', Annals of the New York Academy of Sciences, 1114 (1),
230-40.
Schofield, D.J., Shrestha, R., Passey, M., Earnest, A. &
Fletcher, S. (2008) 'Chronic disease and labour force participation
among older Australians', Medical Journal of Australia, 1819 (8),
447-50.
Siegrist, J., Benach, J., McKnight, A. & Goldblatt, P. (2009)
Employment arrangements, work conditions and health inequalities,
London, Marmot Report.
Skinner, N. & Pocock, B. (2008) Work, life and workplace
culture: the Australian Work and Life Index 2008, Adelaide, Centre for
Work + Life, Hawke Institute, University of South Australia.
Treasury (2002) Intergenerational Report 2002-03, Canberra
,Department of Treasury; Australian Government.
--(2010a) Intergenerational Report 2010--Australia to 2050. Future
Challenges, Canberra, Department of Treasury, Australian Government.
Warner-Smith, P., Everingham, C. & Ford, J. (2006)
'Mid-age women's experiences of work and expectations of
retirement', Just Policy, 40, 4553.
Winefield, A.H. (1995) 'Unemployment: its psychological
costs'. In C.L. Cooper & I.T. Robertson (eds) International
Review of Industrial and Organizational Psychology 1995, Volume 10,
London, John Wiley & Sons Ltd.
Winefield, A.H., Montgomery, B., Gault, U., Muller, J.,
O'Gorman, J., Reser, J. & Roland, D. (2002) 'The
psychology of work and unemployment in Australia today: an Australian
psychological society discussion paper', Australian Psychologist,
37 (1), 1-9.
Zhang, X., Zhao, X. & Harris, A. (2009) 'Chronic diseases
and labour force participation in Australia', Journal of Health
Economics, 28(1), 91-108.
Endnotes
(1) 1961 is sometimes used as the cut-off date because this is the
year in which fertility peaked. However, we have used the dates adopted
by the Australian Bureau of Statistics (ABS) which fits with census
collection years and thus facilitates data comparison (Buckley 2010:
12-15; ABS 2003)
Table 1: Changes in participation rates of older workers, 1970-2006
Per cent participation
Age group May 1970 June 1999 January 2006
Males Females Males Females Males Females
55-59 91.5 28.7 72.9 44.1 72.7 54.0
60-64 79.2 14.9 46.9 17.6 54.6 31.9
65+ 23.2 3.6 9.7 3.2 10.8 3.7
Source: ABS Labour Force Surveys (1970; 1999; 2006)
Table 2: Employment status of males and females in each baby boomer
age cohort
Males
Employment status (a) 1946-55 1956-65
n=236 n=277 P value
% %
Full-time 62.8 86.0 .000
Part-time 7.2 3.4 .055
Casual 5.2 5.6 .827
Unemployed 3.7 3.2 .751
Retired 13.3 .3 .000
Home Duties .0 .8 .502
Unable to Work 7.1 1.1 .000
Females All baby
Employment status (a) 1946-55 1956-65 boomers
n=259 n-285 P value 1946-65
n=1058
% % %
Full-time 32.0 44.3 .003 43.7
Part-time 25.4 29.4 .288 16.6
Casual 8.4 9.6 .630 7.3
Unemployed 2.6 1.1 .204 2.6
Retired 17.5 .6 .000 7.5
Home Duties 8.2 11.3 .227 5.3
Unable to Work 5.1 3.8 .474 4.1
(a) Volunteers (n=5) and students (n=3) excluded due to low numbers.
Source: NWAHS TFU Survey 2 (CATI), 2007.
Table 3: Hours per week in paid work by age cohort and gender
Age cohort Gender All
baby
boomers
Hours in paid
work 1946-55 1956-65 Males Females 1946-1965
0 hours worked 8.2 5.5 6.5 6.7 6.6
1 hour or more 91.7 94.5 93.5 93.3 93.4
Total 342 494 435 401 836
1-15 8.9 7.2 3.7 12.4 7.9
16-24 10.6 9.2 2.8 17.3 9.7
25-34 15.9 14.5 7.6 23.1 15.1
35-39 19.6 20.7 20.7 19.8 20.3
40 hours 17-5 15.6 19.8 12.7 16.4
41-49 10.8 15.4 18.4 8.2 13.6
50 hours or more 16.7 17.4 27.0 6.5 17.1
Total n 314 464 404 374 778
Source: NWAHS TFU Survey 2 (CATI), 2007.
Table 4: Socio-economic characteristics by employment status
Variable Full-time Part-time Casual
Cohort
1946-55 Cohort 38.8 ** 47.4 142.6
195665 Cohort 61.2 ** 52.6 57.4
Total 596 174 74
Gender
Male 64.8 ** 14.0 ** 37.5 *
Female 35.2 ** 860 ** 62.5 *
Total 596 174 74
Living Arrangements
Family with children 50.3 55.1 * 46.0
Step family & children 3.3 .9 6.4 *
Sole parent/shared parent 5.0 4.8 7.3
Living alone 11.1 6.3 * 11.9
Living with partner only 26.2 * 30.7 19.4
Living with other adults 4.2 2.1 8.9
Total 595 174 74
Education
Secondary 39.1 ** 49.3 60.1 *
Trade/apprenticeship 19.3 ** 5.5 ** 10.5
Certificate/Diploma 24.0 31.1 * 20.0
Bachelor or higher 17.6 ** 14.1 9.4
Total 557 159 69
Marital Status
Married/Partner 80.1 86.0 * 73.7
Separated/Divorced 12.5 8.0 12.8
Widowed 1.3 * 3.4 3.8
Never Married 6.1 2.6 * 9.6
Total 595 174 74
Government Pension
Yes 3.6 ** 11.4 19.2
No 96.4 ** 88.6 80.8
Total 557 160 67
Income
Up to $20,000 1.1 ** 12.7 18.7 *
$20,001,440,000 17.4 * 19.2 32.0 *
$40,001-$60,000 28.2 29.6 24.4
$60,001,480,000 26.1 ** 21.2 12.5
$80,0014100,000 13.3 * 11.1 6.2
More than $100,000 13.9 ** 5.3 6.2
Total 549 153 69
Country of Birth
Australia 71.7 * 65.9 72.3
UK/Ireland 17.4 19.7 23.2
Other 10.9 14.3 4.6 *
Total 596 174 74
Housing Tenure
Mortgage 53.1 ** 44.7 141.9
Outright owner 37.7 ** 45.4 37.3
Renter 9.2 * 9.9 20.7 *
Total 596 174 74
Variable Un-employed Retired Home duties
Cohort
1946-55 Cohort 56.2 96.8 ** 37.5
195665 Cohort 43.8 3.2 ** 62.5
Total 27 79 54
Gender
Male 63.4 40.1 4.1 **
Female 36.6 59.9 95.9 **
Total 27 79 54
Living Arrangements
Family with children 32.7 21.6 ** 58.0
Step family & children 0.0 0.0 1.0
Sole parent/shared parent 30.7 1.3 * 14.5 *
Living alone 26.5 * 11.0 1.8 *
Living with partner only 23.1 61.5 ** 20.4
Living with other adults 9.0 4.7 5.2
Total 27 78 54
Education
Secondary 55.5 57.5 * 76.8 **
Trade/apprenticeship 14.8 14.8 2.2 *
Certificate/Diploma 24.0 18.2 13.6
Bachelor or higher 15.8 9.6 7.3
Total 23 73 46
Marital Status
Married/Partner 55.3 * 87.4 83.1
Separated/Divorced 23.3 5.9 4.2
Widowed 11.5 5.3 7.2 *
Never Married 20.0 * 1.4 5.4
Total 27 79 54
Government Pension
Yes 42.9 ** 39.5 ** 36.2 **
No 57.1 ** 60.5 ** 63.8 **
Total 25 73 45
Income
Up to $20,000 34.1 ** 19.9 * 14.4
$20,001,440,000 22.0 29.0 40.3 **
$40,001-$60,000 31.1 247 20.4
$60,001,480,000 11.5 12.6 11.1
$80,0014100,000 0.0 9.8 7.8
More than $100,000 1.3 4.1 6.1
Total 23 70 45
Country of Birth
Australia 71.5 55.6 * 73.1
UK/Ireland 10.8 30.4 * 5.3 *
Other 117.8 14.0 21.7 *
Total 27 79 54
Housing Tenure
Mortgage 47.8 9.5 ** 43.2
Outright owner 26.6 82.9 * 46.5
Renter 25.5 * 7.6 10.4
Total 27 78 54
Variable Unable to
work
Cohort
1946-55 Cohort 68.4 **
195665 Cohort 31.6 **
Total 44
Gender
Male 45.2
Female 54.8
Total 44
Living Arrangements
Family with children 30.2 *
Step family & children 1.2
Sole parent/shared parent 7.5
Living alone 29.2 **
Living with partner only 24.6
Living with other adults 7.3
Total 44
Education
Secondary 42.9
Trade/apprenticeship 20.5
Certificate/Diploma 20.2
Bachelor or higher 16.3
Total 39
Marital Status
Married/Partner 56.7 **
Separated/Divorced 28.9 **
Widowed 1.3
Never Married 13.1 *
Total 44
Government Pension
Yes 68.0 **
No 32.0 **
Total 40
Income
Up to $20,000 67.4 **
$20,001,440,000 9.4
$40,001-$60,000 13.5
$60,001,480,000 7.9 *
$80,0014100,000 1.8
More than $100,000 0.0 *
Total 37
Country of Birth
Australia 58.9
UK/Ireland 22.4
Other 18.8
Total 44
Housing Tenure
Mortgage 30.8 *
Outright owner 32.8
Renter 36.4 **
Total 44
*p<.05; **p<.001 Tests examine differences between observed
and expected cell proportions.
Source: NWAHS Stage 2,2004-06; NWAHS TFU Survey 2 (CATI), 2007
Table 5: Percentage of each employment status category with
selected chronic conditions, baby boomers 1946-65
Chronic condition(CC) % of
sample Total Full-time Part-time
with CC n % %
Diabetes 5.9 0.973 4.2 * 4.5
Arthritis 19.0 1021 14.3 ** 15.1
Cardio 4.0 1018 3.6 2.0
Depression 12.9 1021 9.0 ** 12.8
COPD 6.3 972 4.0 ** 6.2
Asthma 15.0 679 13.9 117.3
Kidney Disease 5.5 959 4.8 6,.3
Fair/Poor SRH 20.1 1049 * 14.6 ** 16.2
CCs [greater than 4.4 928 1.9 ** 9.8
or equal to] 3 (a)
Chronic condition(CC) Home
Casual Un- Retired duties
% employed % %
Diabetes 6.7 14.9 11.3 * 12.0
Arthritis 31.0 ** 28.8 40.3 ** 15.6
Cardio 7.6 0 14.1 3.3
Depression 22.7 ** 28.71 * 12.9 15.5
COPD 10.1 0 16.2 ** 8.4
Asthma 20.5 21.4 12.1 10.2
Kidney Disease 8.2 2.8 7.7 6.5
Fair/Poor SRH 28.4 31.4 34.4 ** 23.3
CCs [greater than 8.0 3.0 9.2 ** 3.9
or equal to] 3 (a)
Chronic condition(CC) Unable to
work
%
Diabetes 18.1 *
Arthritis 40.5 **
Cardio 14.5 *
Depression 39.7 **
COPD 15.8 **
Asthma 17.6
Kidney Disease 3.1
Fair/Poor SRH 60.5 **
CCs [greater than 20.3 **
or equal to] 3 (a)
(a) Multiple CCs variable includes: diabetes, asthma, cardio,
COPD, depression, kidney disease, arthritis.
* p value<.05; ** p value<.001
Source: NWAHS TFU Survey 2 (CATI), 2007; NWAHS Stage 2,2004-06.
Table 6: Sex adjusted odds ratios for reporting not in labour
force by chronic condition and age cohort
1946-55
Chronic (1) ILF (2) NILF P Val ORs
condition
n % n %
Depression 339 7.8 121 17.6 0.007 2.36(1.27-4.38)
Arthritis 339 22.1 121 38.3 0.002 2.04(1.30-3.21)
Diabetes 339 6.9 121 14.4 0.009 2.43(1.24-0.75)
Asthma 339 17.6 121 12.5 0.283 070(0.37-1.34)
Cardio 337 6.1 120 6.7 0.356 1.55(061-3.89)
Kidney
disease 334 7.4 121 8.2 0.972 0.99(0.45-2.151
COPD 339 5.7 121 128 0.015 2.46(1.19-5.05)
1956-65
Chronic (1) ILF (2) NILF P Val ORs
condition
n % n %
Depression 470 13.2 44 15.9 0.859 1.08(0.45-2.61)
Arthritis 470 11.1 44 15.4 0.332 1.57(0.63-3.88)
Diabetes 470 2.8 44 13.7 <0.001 8.74(2.67-28.61)
Asthma 470 13.7 44 11.3 0.343 0.58(0.19-1.79)
Cardio 470 1.6 44 5.9 0.043 6.38(1.05-39.43)
Kidney
disease 457 3.9 43 12.5 0.441 0.46(0.06-3.32)
COPD 470 4.3 44 13 0.034 3.10(1.09-8.8)
Total
Chronic (1) ILF (2) NILF P Val ORs
condition
n % n %
Depression 809 11.0 164 171 0.008 2.01(1.21-3.36)
Arthritis 809 15.7 165 32.3 0.029 1.61(1.05-2.46)
Diabetes 809 4.5 165 14.2 <0.001 3.06(1.64-5.69)
Asthma 809 13.4 165 10.8 0.218 0.70(0.39-1.24)
Cardio 806 3.5 164 6.5 0.190 1.79(0.75-4.25)
Kidney
disease 791 5.4 164 6.7 0.291 0.67(0.31-1.42)
COPD 809 4.9 165 12.9 0.001 2.95(1.57-5.54)
(1) In the labour force; (2) Not in the labour force
Source: NWAHS TFU Survey 2 (CATI), 2007; NWAHS Stage 2, 2004-06
Table 7: Age adjusted odds ratios for reporting not in labour
force by chronic condition and gender
Males
Chronic (1) ILF (2) NILF
condition
n % n % P Val OR
Depress 431 8.3 52 20.8 .000 5.20(2.06-13.1)
Arthritis 431 13.9 52 35.2 .004 2.87(1.41-5.84)
Diabetes 431 5.9 52 18.8 .067 2.31(0.94-5.65)
Asthma 431 14.1 52 10.3 .673 0.81(0.29-2.20)
Cardio 429 4.9 52 9.8 .959 1.03(0.35-3.07)
Kid Dis 422 3.2 52 4.4 .924 0.93(0.20-4.31)
COPD 431 3.9 52 10.4 .040 3.43(1.06-11.1)
Females
Chronic ILF (a) NILF (b) P Val OR
condition
n % n %
Depress 378 13.9 112 15.5 349 1.35(0.72-2.53)
Arthritis 378 17.7 112 30.9 .398 1.26(0.74-2.15)
Diabetes 378 2.9 112 12.1 .003 3.81(1.57-9.23)
Asthma 378 16.7 112 13.0 .221 0.65(0.32-1.30)
Cardio 378 1.8 112 5.0 .064 4.63(0.91-23.5)
Kid Dis 367 7.9 112 7.7 .286 0.63(0.27-1.48)
COPD 378 5.9 112 14.0 .008 2.73(1.29-5.75)
(a) In the labour force; (b) Not in the labour force
Source: NWAHS TFU Survey 2 (CATI), 2007; NWAHS Stage 2, 2004-06
Table 8: Multivariate logistic regression for labour force
participation in the baby boomer (1946-1965) cohort
95% CI P Value ORs 95% CI
Lower Upper
Age .000 1.0
55-59 (reference group)
38-39 .003 5.12 1.718 15.233
40-44 .000 7.00 3.783 12.950
45 49 .000 5.73 3.264 10.052
50-54 .000 3.45 2.104 5.666
Sex (female) .000 .28 .168 .470
Education .037 1.0
Bachelor degree or
higher (reference group)
Certificate/diploma .862 1.07 .523 2.170
Trade/Apprenticeship .151 .56 .253 1.235
Secondary education .077 .57 .308 1.063
Housing .006 1.0
Renter/other
(reference group)
Owner/joint owner .784 1.08 .609 1.928
Mortgage .017 2.11 1.140 3.905
Chronic conditions
Arthritis .007 .122 .027 .549
Diabetes .005 .39 .205 .742
Chronic Obstructive .012 .418 .215 .813
Pulmonary Disorder
Interactions
Sex by arthritis .019 2.930 1.198 7.166
(a) p value is based on Likelihood Ratio Tests
Source: NWAHS TFU Survey 2 (CAT), 2007; NWAHS Stage 2, 2004-06
