When asked what he feared most, then Prime Minister Harold MacMillan is said to have responded: "Events dear boy, events". This article examines particular types of events that may occur at different stages of life and may have consequences for the wellbeing of those involved, such as serious personal injury/illness, serious injury/illness or death of a family member, relationship changes, and changes related to financial matters. The full list of life events covered is discussed further below.

Events spark change, but change is also an unavoidable part of life. Some key events are associated with developmental milestones that at a broad level cover infancy, toddlerhood, middle childhood, adolescence, youth, and middle and old age. For example, in many societies, a child enters primary school, progresses to secondary school, and then perhaps engages in further training or enters tertiary education, hopefully finding productive employment as a result of such endeavours.¹

The achievement of these various milestones is generally associated with further transitions, such as leaving the family home and forming an intimate live-in relationship. Despite increasing delays in partnership formation in the post-World War II era, most young adults eventually become parents, typically within a marriage. A small minority of Australian women have one or more children without having lived with the child's father. Some couples separate, and some separated individuals find new partners.

Leaving the parental home, partnering, separating, and having children are typically associated with other events, such as changes in housing. Across the life course, many individuals will have experienced a period in which they have lived alone, with a partner only, with a partner and child(ren), and/or with child(ren) only. Throughout the course of their lives, they may experience a variety of other unsettling events, including serious personal injury/illnesses. They are also likely to be unsettled by events that occur directly to other people, such as a friend or family member suffering a serious injury/illness or dying.

Whether or not they are largely benign or beneficial, life events alter the status quo. Many spark a series of other events that are also likely to have implications for personal wellbeing. Furthermore, as Moloney, Weston, Qu, and Hayes (2012) pointed out, the effects of any anticipated and realised "life event" are likely to be influenced by personal and contextual circumstances, including personal appraisal of the event, coping skills, access to effective support networks, socio-economic resources, and recent and current exposure to other difficult or favourable events or circumstances.

The experience of stressful life events - such as job loss, accidents or illnesses, and relationship breakdown - often sparks the need for and use of services that fall within the portfolio of the Australian Government Department of Human Services (the department). Such services include Medicare, Centrelink and Child Support. The department is in the process of undertaking a series of service delivery reforms that are aimed at significantly improving the effectiveness of the services provided.

To help inform this process, the department commissioned the Australian Institute of Family Studies (AIFS) to undertake research on the experiences and effects of life events. The research comprised a review of the literature on life events, including underlying theories and factors related to their effects on families, and issues for survey delivery (see Moloney et al., 2012), and analyses of two large-scale national longitudinal datasets that tap the experience of various life events - the Household, Income and Labour Dynamics in Australia (HILDA) survey and Growing Up in Australia: The Longitudinal Study of Australian Children (LSAC) (see Qu, Baxter, Weston, Moloney, & Hayes, in press).² This article outlines some of the key results of the HILDA and LSAC analyses.

The two datasets differ in a number of ways. For example, HILDA allows a focus on the whole of the adult population - young and old, men and women. Respondents are presented with a list of life events and asked to indicate which event, if any, they had experienced in the previous 12 months. LSAC is narrower in scope in that it focuses exclusively on families with relatively young children, thereby allowing a closer examination of life events that occur during this stage of life. In this analysis, we used the information provided in LSAC by primary carers. Unlike HILDA, which focuses on the events experienced by individual respondents, primary carers in LSAC who are partnered are asked to indicate whether they or their partner had experienced any of the events listed during the previous 12 months.

The results outlined below are presented in three sections, covering:

• the overall prevalence of specific life events experiences and multiple life events;
• personal or family characteristics of those who are most likely to experience particular life events; and
• links between the experiences of life events and personal (subjective) wellbeing.

Within each of these sections, the results that emerged for the broader (i.e., HILDA) sample are discussed before attention is given to those based on the narrower (i.e., LSAC) sample.

Overall prevalence of life events

Results from the HILDA survey

The HILDA survey, which began in 2001, follows the members of an initial national sample of 7,682 households on an annual basis, along with other individuals on a continuing or temporary basis. All members of the initial household sample are considered to be continuing sample members (CSMs), as are any subsequent children they have or adopt and the other parent of these children. These CSMs are followed up indefinitely. Unless they have a child with a CSM, other household members are interviewed only for as long as they continue to live with the CSM. All those in the household aged 15 years and older are eligible for interview. Nearly 14,000 household members participated in Wave 1 of the survey, while just over 13,500 members participated in Wave 10 (Melbourne Institute of Applied Economic and Social Research, 2011).

We focused on 14 life events (or sets of events) experienced between late 2009 and late 2010, as reported by respondents in the 2010 (Wave 10) survey. The events are shown in Figure 1, along with the percentage of all respondents who had experienced them in the 12 months prior to the 2010 survey.

The most commonly occurring life events were: a close relative/family member experiencing a serious injury/illness (14%) and changing residence (13%), followed by the death of a close friend (12%), and the death of a close relative or family member (11%). The least commonly occurring events were the death of one's spouse or child, detention of the respondent or a family member in jail, the respondent becoming a victim of physical violence, and the respondent forming a live-in relationship. These events were each reported by fewer than 2% of respondents.

Over the 12-month period, around one-half of respondents (47%) indicated that they had not experienced any of the events examined. Just under one-third (32%) said they had encountered only one of the events, 14% mentioned experiencing two such events, and 7% had encountered at least three of the events.

Figure 1: Prevalence of life events in the 12 months prior to 2010 (HILDA)

Source: HILDA, Wave 10

Results from LSAC

LSAC focuses on two age cohorts of children and their families, with each cohort initially comprising approximately 5,000 children. The first survey wave was conducted in 2004, when children in the younger cohort were infants (called the "B cohort") and those in the older cohort were 4-5 years old (called the "K cohort"). These two cohorts and their families are followed up every two years, with the most recent data that are available for analysis having been collected in 2010 (Wave 4), when the B and K cohorts were 6-7 and 10-11 years old respectively. Information was collected from around 83-84% of the original sample in Wave 4, representing more than 8,000 families. The results concerning the experience of life events are based on Wave 4 data.

In each main wave of LSAC, the "primary carer" (i.e., the parent or guardian who indicated that they knew the child best - almost always the mother) completes a questionnaire that includes a list of life events. As noted above, respondents are asked to indicate any of those that either "happened to" them or their partner during the previous 12 months. While it was not possible to identify which events occurred more directly to the respondent than partner, events occurring to one partner are likely to have repercussions for the other partner, and some events listed were obviously "family events". Attention was directed to the prevalence of life events experienced by those with children in either cohort. That is, the data provided by primary carers in the two cohorts were combined. Given that virtually all respondents were parents (or step-parents) of the children in the study, they will frequently be referred to as "the parents" in the discussion below.

The LSAC life events (or sets of events) examined in the present study are shown in Figure 2. A number of these cover much the same experiences as those assessed in HILDA, although the combination of events within a set tends to differ. Two examples of matters asked only of LSAC respondents - living in a drought-affected area and living with someone with an alcohol or drug problem - refer to ongoing circumstances that may have commenced years earlier, rather than events that only commenced within the previous 12 months.³

The death of a close family friend or another relative was the most commonly experienced event, with one quarter of respondents reporting that this had occurred to them or to their partner in the previous 12 months (Figure 2). The next most commonly experienced event was that a close relative had become seriously ill or had been injured or assaulted (20% of respondents). On a similar theme, 12% of respondents indicated that they or their partner had suffered a serious illness, injury or assault. A slightly higher proportion of respondents (17%) reported that they had moved house. A major financial crisis was also reported by 12%. For some families this may have been related to the respondent or partner having lost his or her job - reported by 9% of respondents.

Relationship changes were less commonly experienced than those mentioned above: 6% referred to separation due to relationship or marital difficulties; 4% reported the break-up of a steady romantic relationship; and 2% indicated that they had started living with a new partner or spouse. The other seven life events examined were experienced by 3-8% of respondents.

In total, 70% of respondents indicated that at least one event had occurred to either them or their partner. This includes 30% who had experienced only one of the events listed, approximately 20% who nominated two events, 10% who referred to three events, and 10% who experienced four or more events.

Figure 2: Prevalence of life events in the 12 months prior to 2010 (LSAC)

Source: LSAC, B and K cohorts, Wave 4

Summary

Overall, a higher percentage of respondents in LSAC than HILDA referred to the experience of one or more life events in the previous 12 months (70% vs 50%). This trend may be partly explained by the fact that the experience of many of the events listed in each survey would be age-related (an issue examined below) and more generally feasible for LSAC than HILDA sample members given their common stage in family life. There are also differences in the life events listed. In addition, the LSAC results refer to experiences of two people if the respondent has a partner, whereas those pertaining to HILDA focus on the experiences of the respondent alone.

Despite these differences, some consistency in the findings from the two studies emerge. In each study, the life events that were relatively common included having an illness or injury occur to a close relative, having a close friend or other relative die, suffering a serious personal injury or illness, and changing residence.

In the next section, we outline some of the ways in which the experience of life events varied according to a selection of personal and family characteristics, including age and financial circumstances.

Prevalence of life events according to personal and family characteristics

In order to assess the prevalence of life events among people with different personal and family characteristics, we focused on the personal and family characteristics of respondents that were captured in the survey wave preceding the one from which the life events data that we used were derived. That is, for HILDA respondents we focused on their 2009 (Wave 9) personal and family characteristics, along with their experiences of life events reported in 2010 (Wave 10), while for LSAC respondents, we used their personal and family characteristics recorded in 2008 (Wave 3) and their experiences of life events reported in 2010 (Wave 4). Although there is a two-year interval between LSAC survey waves, like HILDA respondents, those in LSAC are asked to identify the life events that had taken place in the previous 12 months. In other words, we examined the nature of socio-demographic factors that might be associated with the subsequent experience of particular life events.

Analyses of HILDA

The experience of some of the life events examined would be impossible or highly improbable for some individuals, given their personal circumstances. For example, separation from a partner requires one to have been partnered in the first place, and pregnancy or childbirth requires women to be in their reproductive years (notwithstanding modern reproductive technology). For this reason, we narrowed our focus of some events to certain groups, which we explain in the discussion of the results. However, we did not impose such restrictions in the analysis of age-related differences (outlined first).

Age differences

Figure 3, which focuses on the more commonly occurring life events, shows that their experiences varied with age in obvious ways, with partnership formation and having a new child occurring in young adulthood, and retirement from the workforce occurring later in life. The experience of a serious personal injury or illness tended to increase with age, and was most common among those aged 75 years or more. Encountering the death of a close friend increased almost progressively with age. However, the death of a close relative (other than a spouse or child), and serious injury to a family member, were only marginally less commonly experienced by people under 25 years old compared with older age groups. A change in residence most commonly occurred to respondents under 35 years old (especially to those aged 25-34 years) and became progressively less common for older age groups.

Figure 3: Prevalence of selected life events in the previous 12 months, by age and gender, 2010 (HILDA)

Note: Age was measured at 2009 (Wave 9) and the experience of life events (over the previous 12 months) were reported in 2010 (Wave 10).

Source: HILDA, Wave 9 and 10

Gender and age differences

Overall trends were similar for men and women within the same age cohort. The largest gender differences, representing at least five percentage points in the proportions of men and women reporting the experience, were apparent for the following issues:

• Among those aged 35-74 years (four age cohorts), women were more inclined than men to report a serious injury or illness of a family member (18-21% vs 12-13% respectively).
• Among those aged 65-74 years, a lower proportion of women than men reported a serious personal injury or illness (11% vs 15% respectively).
• Among those aged 15-24 years, a higher proportion of women than men indicated that they had moved residence (25% vs 20% respectively).

Other characteristics were explored in the detailed analyses. Key results of how the prevalence of life events varied with socio-economic circumstances, relationship status and family composition, and residential location are described below.

Socio-economic circumstances

The two family formation events - forming a live-in relationship, and pregnancy (self or partner) or birth/adoption of a new child - were more commonly experienced by those with relatively high rather than low socio-economic status. For example, using education as an indicator of socio-economic status:

• Of single men and women under 65 years old, 12-14% who had a degree or higher qualification had "formed a live-in relationship", compared with 5-7% who did not have a post-school qualification.
• Of all men under 55 years old, 12% who had a degree or higher qualification indicated that their partner had become pregnant or that they and their partner had had a child together, compared with 6% without a post-school qualification.
• Of all women under 45 years old, 17% with a degree or higher qualification indicated that they had become pregnant or given birth to a child, compared with 10% of those with no post-school qualification.

On the other hand, for men and women, experiencing a separation varied only marginally with socio-economic status.

Relationship status and family composition

Separation from a partner was more commonly experienced by those in a cohabiting rather than married relationship, and by those in step-families rather than those not in step-families:

• Of partnered men and women under 65 years, 6-8% who were cohabiting and 2% who were married subsequently indicated that they had separated from their partner or spouse.
• Partnered women in households with step-children were marginally more likely than their counterparts of the same sex to subsequently report that they had separated from their partner (6% vs 3%). Among partnered men, 4% of those in step-families and 2% of others subsequently reported that they had separated.
• Moving house was more commonly experienced by men and women who were single or cohabiting (18-23%), than by those who were married (9%).

Residential location

Two aspects of residential location were examined: regional location (major city, inner region or outer/remote region) and the level of socio-economic advantage or disadvantage of the area in which respondents lived:⁴

• A slightly higher proportion of men (under 55 years) and women (under 45 years) who lived in outer/remote regions than those of the same gender who lived in inner regions or major cities reported the conception or birth of a child (men: 14% vs 7-9%; women: 17% vs 10-13%).
• Men and women who lived in a major city were marginally less likely than men and women who lived elsewhere to indicate that a close friend had died (10% vs 14-15%).
• Slightly higher proportions of men and women who lived in disadvantaged areas in 2009 than those of the same sex who lived in advantaged areas subsequently indicated the following events:
• separation from a spouse or long-term partner (for partnered respondents under 65 years) (3-4% vs 1%);
• the experience of a personal injury or illness (11-12% vs 9%); and
• the death of a close friend (14-15% vs 9-10%).

Although such locational differences in experiences were small, the direction they took was consistent for men and women.

Analyses of LSAC

Some key results regarding the strength of links between LSAC parents' experiences of life events and their family composition, socio-economic circumstances and residential location are listed below. As noted earlier, these results refer to life events experiences over the previous 12 months reported in 2010 (Wave 4), and personal and family characteristics recorded in 2008 (Wave 3).

Relationship status, family form and age

The following trends emerged concerning relationship status and family form:

• Cohabiting parents were more likely than married parents to subsequently report that they experienced separation due to relationship or marital difficulties (11% vs 4%), and cohabiting parents were the most likely to experience a pregnancy or to have had a child (13%), followed by single parents (9%) and then married parents (6%).
• Single parents were the most likely to indicate that somebody other than a new baby or partner moved into the household (13% vs 6-7% of married and cohabiting parents). And, of course, single parents were the most likely of the three parental groups to indicate that they had commenced living with a new partner or spouse (8% vs 1-2%).

Given that all families in LSAC had relatively young children (with the younger of the two cohorts being 4-5 years old and the elder cohort being 8-9 years old in 2008), we would not expect a great deal of variation in the experience of life events according to the age of youngest child in the family. Of all life events examined, pregnancy or birth of a child varied the most according to this factor.

A pregnancy or birth of a child was reported by 11% of parents whose youngest child was 0-4 years old, 4% of those whose youngest child was 5-7 years old, and only 2% of those whose youngest child was 8-9 years old.

Links between life events experiences and the age of the primary carer were also assessed. For this analysis, primary carers (here called "parents") were classified as "younger", "mid-age" or "older". The most notable findings from these analyses were that parents in the younger group were the most likely of all three groups of parents to indicate that they had experienced the following:

• the death of a close friend or relative (30% of younger parents, 23% each of mid-age and older parents);
• relationship breakdown or separation (9% of younger parents, 3% of mid-age and 4% of older parents);
• pregnancy or birth of a child (15% of younger parents, 5% of mid-age and 2% of older parents);
• moving house (24% of younger parents, 15% of mid-age and 11% of older parents); and
• a person (other than partner or child) joining the household (10% of younger parents, 6% each of mid-age and older parents).

Socio-economic circumstances

Total household income represented one of the indicators of socio-economic circumstances used in these analyses. Incomes were classified as "lower", "mid-range" or "higher", based on how each household's income compared to other household incomes in the sample. This measure was related to the following life events:

• a major financial crisis (18% of those with lower household incomes, 10% with mid-range incomes and 6% with higher incomes);
• pregnancy or the birth of a child (10% of those with lower household incomes, 7% with mid-range or 6% with higher incomes); and
• moving house (18% of those with lower household incomes, 16% each for those with mid-range or higher incomes).

The results pertaining to the other indicators of socio-economic circumstances included in the analyses (main source of income, parental employment, the primary carer's educational attainment level, and housing tenure) were fairly consistent with those found in relation to household income. That is, those with poorer socio-economic circumstances generally had the higher rates of experiencing the above life events.

Residential location

The main difference in the experiences of those who lived in different locations by level of remoteness was the expected one - that the more remote the area in which the families lived, the greater was their likelihood of living in a drought-affected area, or having their home or local area being affected by bushfire, flooding or severe storm. Note, though, that the families in LSAC who were living in remote or very remote areas of Australia were not representative of all families living in these regions, given that these areas were excluded when the LSAC sample was initially selected.

Summary

The results based on the HILDA dataset highlighted the extent to which experiences of certain life events vary according to a person's age and gender. Age-related trends were, of course, less apparent in the LSAC results, given that LSAC currently focuses on families with young children, and the vast majority of primary carers (on which the analyses were based) were mothers.

Two of the life events examined - partnership formation and separation - are entirely (or almost entirely) dependent on relationship status. That is, having a partner is a prerequisite for undergoing relationship separation, and being single is almost always a prerequisite for forming a couple relationship, though some partnered individuals may move in with a new partner without encountering an intervening period of being single.

We found that people who were cohabiting had a somewhat greater chance than those who were married of experiencing relationship separation and moving house. The cohabiting respondents in LSAC were also more likely than their married counterparts to experience a pregnancy or the birth of a new child during the 12-month period.

The experiences of some events also varied according to socio-economic status. These included relationship status changes and moving house. Given differences between the two populations, partnership formation was more commonly reported by HILDA respondents with high socio-economic status, many of whom would not have been parents, while this event was more commonly reported by LSAC primary carers with low socio-economic status (virtually all of whom would have been single mothers).

In relation to circumstances that may lead individuals or families to seek additional services or supports from the Department of Human Services, or indeed from other places, the results suggest that certain information may be useful for identifying those at considerable risk of experiencing certain life events. For example, some life events are linked to socio-economic disadvantage or to regions within which people live, while others are necessarily linked to relationship status, as noted above.

Links between life events and changes in personal wellbeing

An important reason for undertaking this research on life events was to improve understanding of ways in which the experience of life events might affect individual or family wellbeing. A range of measures of wellbeing in HILDA, and one in LSAC (personal distress) was used to assess such links. We compared each respondent's level of wellbeing, recorded before and after the "life events period" that we focused upon; that is, we compared the wellbeing of HILDA respondents apparent in 2009 (Wave 9) with that apparent in 2010 (Wave 10), and the personal distress of LSAC respondents apparent in 2008 (Wave 3) with that apparent in 2010 (Wave 4).

It is important to point out that with only two reference points (2009 and 2010 for HILDA, and 2008 and 2010 for LSAC), it is not possible to determine whether any change in wellbeing experienced by those who encountered adverse or beneficial events was a function of these events or whether the change formed part of a longer term trend, perhaps intensified by the events. For example, unhappiness, loneliness and so on may contribute to relationship breakdown and eventual separation, resulting in further deterioration of morale. Similarly, those who are content with their lives may be more likely to attract certain positive events, such as finding a partner and eventually moving in with him or her, the immediate effects of which may enhance morale. For these reasons, we refer to "apparent" effects of events on wellbeing.

We also acknowledge that wellbeing will be a function of a range of personal and family characteristics that are quite unrelated to relatively recent experiences of life events. For example, putting aside whether or not any life events occurred in the past year, wellbeing may be lower for single parents or for people living in jobless households. Such associations are clearly important, but have not been reported on in this paper.

Analyses of HILDA

This article presents the findings pertaining to two of the wellbeing measures that we assessed in HILDA: sense of isolation and overall wellbeing (see Box 1).

The events with apparent negative effects on personal wellbeing included: separation from spouse or a long-term partner, serious personal injury/illness, a major worsening of financial circumstances, and being a victim of physical violence. Generally, men and women who experienced such events over a 12-month period had a decline in their overall wellbeing and were more likely to have a high sense of social isolation after the event (i.e., in 2010) than before its occurrence (i.e., in 2009).⁵ There were, however, some gender differences. For example, Figure 4 shows that, of men who separated from their spouse or partner, the proportions classified as having a high sense of social isolation increased from 7% before the event to 23% after the event. Of women who experienced this event, the proportions with a high sense of isolation were stable (14-15% before and after the event), but higher than those derived for men and women who had not experienced this event (6-7%). Women who separated were more likely than their male counterparts to report a high sense of social isolation before the event (15% vs 7%) but less likely than men to indicate this after separation (14% vs 23%).

Figure 4: Prevalence of high sense of isolation, by experience of selected life events and gender, 2009 and 2010 (HILDA)

Note: The analyses for "separated from spouse or long-term partner" only included those who were partnered and aged less than 65 years age in Wave 9.

Source: HILDA, Wave 9 and 10

The other life event that had quite substantial associations with having a high sense of isolation was being a victim of physical violence. Although those who experienced this event were already more likely than others to report a high sense of isolation, the experience of this event was associated with a subsequent substantial increase in the chance of reporting a high sense of isolation (from 11% to 27% for men, and from 19% to 28% for women).

The findings for overall wellbeing are shown in Figure 5. The pattern of results pertaining to relationship separation are similar to, but not as strong as, those emerging for sense of social isolation. Specifically, men and women who experienced separation were more likely than those who remained partnered to indicate low overall wellbeing both in 2009 and 2010 (men: 32-39% vs 22-23%; women: 40-44% vs 22-23% respectively). Secondly, separated women were more likely than separated men to report low wellbeing before this event took place (44% vs 32%). Thirdly, the likelihood of separated men reporting low wellbeing increased after separation (from 32% to 39%) while that for women remained stable (40-44%). The lower wellbeing of women than men prior to separation has been observed in other studies and helps explain the fact that women are more likely to initiate the separation than men (e.g., Weston, 1986).

Figure 5 also shows that the percentages indicating low wellbeing increased among men and women who experienced the other life events presented. Furthermore, the 2009 rates of low wellbeing were greater for those who went on to experience the event than for those who did not. There are at least two possible explanations for this finding. Firstly, relatively low wellbeing may contribute to some negative events. Secondly, the event may represent an outcome of, or stage in, a process that had already been well advanced at the time the 2009 survey took place.⁶ This is one reason for the event itself not necessarily resulting in a further decline in wellbeing.⁷

Figure 5: Prevalence of relative low wellbeing, by experience of selected life events and gender, 2009 and 2010 (HILDA)

Note: The analyses for "separated from spouse or long-term partner" only included those who were partnered and aged less than 65 years age at Wave 9.

Source: HILDA, Wave 9 and 10

The following events did not have a significant association with personal wellbeing: serious injury/illness to family member; death of close relative/family member; and death of a close friend. For many individuals, it is possible that such events are sufficiently public and noticeable as to attract a range of supportive responses. It is also possible that many individuals who experienced these specific events had lowered wellbeing for a period, but had readjusted to these events within the 12-month period examined.⁸

The experience of unsettling multiple life events appeared to accentuate any negative effects on personal wellbeing. This became apparent when people reported three or more such events over the 12-month period.⁹ For instance, Figure 6 and Figure 7 show that men who experienced three or more life events were more likely to indicate a high sense of social isolation and low overall wellbeing in 2010 than in 2009. Among women, this effect was apparent for low overall wellbeing but not for high sense of social isolation.

Consistent with the earlier findings, the likelihood of expressing low overall wellbeing in 2009 (pre-events) was greatest for men and women who subsequently experienced three or more events. In fact, for women in particular, this likelihood increased progressively with increases in the number of events subsequently experienced.

Figure 6: Prevalence of high sense of isolation, by number of life events experienced in the 12 months prior to 2010 and gender, 2009 and 2010 (HILDA)

Source: HILDA, Wave 9 and 10

Figure 7: Prevalence of relative low wellbeing, by number of life events experienced in the 12 months prior to 2010 and gender, 2009 and 2010 (HILDA)

Source: HILDA, Wave 9 and 10

Analyses of LSAC

The LSAC analyses focused exclusively on links between the experience of life events and levels of non-specific emotional distress apparent in 2008 (Wave 3) and 2010 (Wave 4). For simplicity, attention focused on mothers' reports only.

The Kessler 6 scale was used to measure non-specific psychological distress (Kessler et al., 2002). This six-item measure has been widely used and validated in many epidemiological studies. Respondents are asked to rate how often in the previous four weeks they have experienced certain symptoms of distress (e.g., felt nervous, hopeless, worthless), using a 5-point scale ranging from 0 ("none of the time") to 4 ("all of the time"). Responses are summed to produce a possible score between 0 and 24. A higher score indicates a higher level of psychological distress. Those with scores of 10 or more are classified as distressed.

An important point about these analyses is that the LSAC life events questions ask about whether the respondent or partner experienced the various life events. Thus, we assessed the extent to which the experience of a life event that occurred to one or both parents (if partnered) was associated with a change in the mother's level of distress. While some of the events were clearly family-level events (such as experiencing a financial crisis), and events such as job loss would have financial ramifications for both partners, it seems reasonable to generally assume that any disturbing event occurring to one partner would be likely to affect the other partner. As with the HILDA analyses, we examined whether respondents who experienced each life event had higher distress not only after the event had occurred but also beforehand.

Figure 8 focuses on a selection of the life events that were more strongly related to mothers' wellbeing. For example, among the mothers who, in 2010, reported that they had suffered a serious illness, injury or assault during the previous 12 months, 10% were already psychologically distressed in 2008 (before the event took place). After the event took place, 15% were distressed. In comparison, among those who had not experienced such events, 7% indicated high distress in both 2008 and 2010. The relatively high percentage of mothers who were classified as distressed in 2008 (9-19%) is apparent for all the life events shown in Figure 8. For all comparisons, those who did not experience the event in question were less likely to indicate high distress in both periods (6-8%).

There was also evidence of distress levels increasing following the experience of particular life events. Figure 8 shows increases in the percentages of mothers classified as distressed for mothers (and/or partners) who experienced:

• a major financial crisis;
• breaking off a steady romantic relationship;
• separation due to relationship or marital difficulties;
• suffering an illness, injury or assault;
• having someone new move into the household; and
• a serious illness, injury or assault happening to a close relative.

While not shown in this figure, increases in distress among mothers were also apparent for those who experienced:

• loss of a job not from choice;
• having something of value lost or stolen; and
• having their home or local area affected by bushfire, flooding or severe storm.

Not surprisingly, the more events that mothers reported, the higher was their level of distress (Figure 9). Levels of distress were relatively high even before these life events occurred, but there was some indication of an increase in distress after the events had taken place, especially for those who experienced four or more life events.

Figure 8: Prevalence of mothers' distress, by selected life events occurring in the 12 months prior to 2010, 2008 and 2010 (LSAC)

Source: LSAC, B and K cohorts, Waves 3 and 4

Figure 9: Prevalence of mothers' distress, by number of life events experienced in the 12 months prior to 2010, 2008 and 2010 (LSAC)

Source: LSAC, B and K cohorts, Waves 3 and 4

Summary

From both the HILDA and LSAC analyses two key findings stand out:

• those who go on to experience life events already had lower wellbeing prior to the life event occurring; and
• experiences of particular life events are associated with some declines in wellbeing.

These are generalisations and do not apply to all life events. Clearly, some life events (such as experiencing a major financial crisis) are likely to have greater implications for wellbeing than others (such as moving house). In both HILDA and LSAC, strong associations with wellbeing were apparent for experiencing a major financial crisis. HILDA analyses also showed strong associations with being a victim of serious physical assault. Again, in both HILDA and LSAC, experiencing a relationship separation or breaking off a serious romantic relationship was also linked with having poorer wellbeing.

Multiple life events also seem to pose difficulties for people, with the number of life events experienced being quite strongly related to the wellbeing measures explored here.

Generally, the findings were apparent for both the measures presented from HILDA and from the measure of mothers' wellbeing from LSAC. Further evidence is provided, using HILDA, in the full report. Of course, wellbeing can always be measured in more extensive ways. In particular, with regard to family wellbeing and the analyses in LSAC, it is important to acknowledge that life events may affect families in ways other than were examined in these analyses, with the wellbeing of other family members, including children, potentially affected by families' experiences of life events.

The findings of associations between life events and wellbeing will, to some extent, reflect that wellbeing is likely to be dependent on a range of factors beyond those captured by experiences of specific life events. In the previous section of this paper we saw that certain groups within the population are more likely to experience life events than others, and the characteristics that are linked to a higher likelihood of experiencing life events (such as being socio-economically disadvantaged) may also be linked with a higher likelihood of having poorer wellbeing. That is, while life events do seem to matter to wellbeing, they should be viewed in the broader context of what may be important to the wellbeing of individuals and families.

A limitation of these analyses is that we have not taken account of the fact that some of the events examined may have been chronic, or may represent stages in a long-term process. Each of these longitudinal studies offers the potential to examine these processes across multiple waves of the study in future analyses.

Summary and conclusions

This article summarises the main results emerging from analyses of the experience of life events reported by participants in the HILDA and LSAC surveys. Analyses from HILDA provide a broad population view, as respondents included men and women aged 15 years or more. This dataset allows investigation of life events that occur across the adult life course; for example, forming relationships and having children, possibly separating, and retiring from the workforce. In contrast, the LSAC data focus on families with young children.

While these two sets of analyses have made use of quite different data, considerable consistency was apparent in some of the key findings that emerged. Not surprisingly, the analyses based on both datasets showed that many of the life events that were explored were quite uncommon when the focus was on a 12-month period, with the HILDA dataset highlighting the fact that the likelihood of experiencing certain events varies across one's life course. While some events may not be very likely at the population level covering a 12-month period, they may be quite common when the focus is on those in a particular stage of life. In addition, experiences of life events will, of course, accumulate, with some defining or contributing to the direction a person's life course takes.

Of course, some events are somewhat random, and others are influenced more directly by the decisions that people make (e.g., separation may be the result of one partner's decision). Some may flow from longer term experiences of difficulties and previous experiences of life events (e.g., families may experience ongoing difficulties associated with a family member having drug or alcohol problems).

The analyses of each of the datasets highlighted that some individuals or families appear more vulnerable to the experience of negative life events. For instance, the likelihood of experiencing some of the life events varied with a respondent's age or socio-economic circumstances.

In relation to the effects of life events on personal wellbeing, one finding that stood out was that those who subsequently experienced life events often already had poorer wellbeing at the outset, compared with those who did not subsequently encounter these events. Nevertheless, the experience of the life event was often accompanied by a further decline in wellbeing. The precise results differed according to the life events examined and the measure of wellbeing used.

The results also suggest that not only are vulnerable groups, such as those with limited socio-economic resources, more likely than others to experience the life events examined, the repercussions of the events are also more likely to have serious ("last straw") effects.

Put another way, adverse life events can be an important signal of the need for assistance, but the events themselves may be the culmination of an extended negative process that needs to be addressed. In some cases, an event such as relationship separation can provide relief from highly damaging circumstances; in other cases, it can dramatically worsen the plight of at least one family member. But in relation to such events as natural disasters, the negative effects are likely to dominate.

Such findings highlight the need for prevention and early intervention. Considerable damage can occur for partners and children before the event of separation takes place (e.g., see Amato & Booth, 1997; Cherlin, Chase-Lansdale, & McRae, 1998; and Pryor & Rodgers, 2001, regarding the negative effects of high levels of pre-separation conflict on children). At the same time, the data suggest that sole reliance on "life events" as indicators of the need for service provision would be unfortunate. The identification of individuals or families who are vulnerable to experiencing adverse events in the future is clearly important, but so too is the identification of families experiencing chronically destructive circumstances such as family violence.

Taken together, the present analyses indicate that events are more likely to differ at some life stages than at others, and that life events are more likely to be encountered by some people than others. Social address influences the likelihood of experiencing multiple potentially stressful life events and limits the capacity of individuals and their families to negotiate these successfully. The results also provide a basis for targeting services to those who are more likely to be placed at risk as a result of the load of life events that may befall them. As such, prevention and early intervention efforts are best targeted and tailored to those most at risk.