Pluralized life courses? An exploration of the life trajectories of individuals with psychiatric disorders

Abstract

Background: Most of the existing research relating to the life courses of people with psychiatric symptoms focuses on the occurrence and the impact of non-normative events on the onsets of crises; it usually disregards the more regular dimensions of life, such as work, family and intimate partnerships that may be related to the timing and seriousness of psychiatric problems. An additional reason for empirically addressing life trajectories of individuals with psychiatric problems relates to recent changes of family and occupational trajectories in relation to societal trends such as individualization and pluralization of life courses.

Aim: This paper explores the life trajectories of 86 individuals under clinical supervision and proposes a typology of their occupational, co-residence and intimacy trajectories. The results are discussed in light of the life-course paradigm.

Method: A multidimensional optimal matching analysis was performed on a sample of 86 individuals under clinical supervision to create a typology of trajectories. The influence of these trajectories on psychiatric disorders, evaluated using a SCL-90-R questionnaire, was then assessed using linear regression modelling.

Results: The typologies of trajectories showed that the patients developed a diversity of life trajectories. Individuals who have developed a standard life course with few institutionalization periods reported more symptoms and distress than individuals with an institutionalized life trajectory.

Conclusion: The results of this study stress that psychiatric patients are social actors who are influenced by society at large and its ongoing process of change. Therefore, it is essential to take into account the diversity of occupational and family trajectories when dealing with individuals in therapeutic settings.

Keywords

life-course paradigm life trajectories mental health optimal matching sequence analysis

Introduction

One underdeveloped area of research on individuals with psychiatric symptoms concerns their long-term family and occupational trajectories. Although an interest in the effects of non-normative events on the onset of psychiatric disorders has developed in social psychiatry and in sociology (Becker, 1986; Goodyer, 1996; Turner and Lloyd, 1995; Wheaton, 1990), occupational and family trajectories of individuals with psychiatric disorders are not well known. Accounting for such trajectories is necessary, as they are related to key dimensions of physical and mental health in non-clinical samples (Dannefer, 2003; Ferraro et al., 2006; Ross and Wu, 1996; Willson et al., 2007). Indeed, research shows that individuals with erratic occupational and family trajectories and with a variety of non-normative transitions have a higher likelihood of experiencing depression and having a lower self-esteem, as well as more somatic and non-somatic problems (Gilman et al., 2003; Heim and Nemeroff, 2001; Kendler et al., 1999; Kessler, 1997; Penza et al., 2003; Ross and Mirowsky, 1999).

An additional reason for empirically addressing life trajectories of individuals with psychiatric problems relates to recent changes of family and occupational trajectories in relation to societal trends such as individualization, pluralization and de-standardization (Beck and Beck-Gernsheim, 2002; Brückner and Mayer, 2005; Widmer and Ritschard, 2009). These trends are likely to have had repercussions for individuals with a clinical record. This paper explores the pluralization of life trajectories of individuals with psychiatric problems based on two non-random samples of individuals supervised in clinical settings. It describes their family, occupational and intimacy trajectories and estimates the interrelations existing between those trajectories and psychiatric symptoms. Using an optimal matching method (Abbott and Tsay, 2000), first a typology of individual life courses is built by taking into account their occupational, intimate, and co-residential trajectories. Second is an estimation of the link existing between the discovered types of life courses and psychiatric symptoms, as they were measured using the SCL-90-R symptom scale (Derogatis, 1983; Derogatis et al., 1973). The results are discussed from a life course perspective (Elder et al., 2003).

Mental health and the life course

Since it was first formalized by Holmes and Rahe (1967), research on the link between life events and illness has thrived. Numerous studies have underlined the importance of non-normative events that precipitate the onset of psychiatric crises. Among these studies, relationships between mental illness and childhood abuse (Heim and Nemeroff, 2001; Penza et al., 2003), parental divorce (Cherlin et al., 1998; Mirowsky, 1999) or family disruption and residential instability during childhood (Gilman et al., 2003) have been underlined.

Trajectories of individuals with mental illness within institutions have also been analysed using the ‘career’ concept (Pavalko et al., 2007) in order to assess the links between institution trajectories and mental health. An important finding is that the trajectories of individuals with severe psychiatric problems are very diverse and have widely different outcomes (Harding, 1998).

Although the importance of such results cannot be disregarded, the issue of the interconnections between psychiatric problems and the life course should not be limited to unique and non-normative events but should take extended life trajectories into account. Barrett (2000) made a remarkable step in this direction by using the marital trajectory as a predictor of mental illness and by underlining a significant influence of this trajectory on mental health. Willits et al. (2004) and Stimpson et al. (2006) also found a significant influence of partnership; for example, individuals with long first partnerships are more likely to have good mental health, while splits are associated with poorer mental health. Regarding the relation between work and mental illness, Scheid and Anderson (1995) have emphasized that having a job is important for individuals with mental illness, but that it can also be a major source of stress. Indeed, the recovery approach using strengths (Davidson et al., 2009; Rapp, 2006) stresses that rehabilitation is sometimes associated with increased symptomatology due to the higher stress created by demanding job and family situations.

One should not restrict the analysis to just one of the multiple trajectories embedded in the life of an individual (Elder, 1999). Indeed, in their summative article, Cook et al. (1997) emphasized the importance of dealing with mental illness by including a variety of dimensions of life, such as family support, integration in local community, and work. They also stressed the differential experience of psychiatric frailty that individuals develop, depending on their birth cohorts, gender, social class, ethnicity and the life stage in which the psychiatric disease appears. Unfortunately, few studies exist that describe life trajectories of individuals with psychiatric disorders (for an attempt, see Singer et al., 1998).

It is necessary to study empirically the life trajectories of individuals with psychiatric disorders as great changes of the social organization of life courses have happened since the end of the 1960s in most Western countries (Shanahan, 2000). Following a secular trend, occupational and intimate trajectories then achieved a high level of uniformity. At that time, a large majority of individuals went through an identified set of ordered and age-graded family and occupational stages with very few of them getting out of sequence or skipping transitions (Kohli, 1986; Modell et al., 1976; Shanahan, 2000). This move towards standardization was to some extent replaced in the late 1960s by an inverse tendency toward a pluralization of both occupational and family trajectories. This trend towards greater complexity and diversity of life paths was presented by individualization theorists as a process that concerned the majority of personal lives and as representing one of the most profound changes affecting societies in late modernity (Beck, 1992; Beck and Beck-Gernsheim, 2002; Sennett, 2000). Recently, the hypothesis that there is a pervasive trend of pluralization of life courses in late modernity has been critically examined in various empirical analyses (Brückner and Mayer, 2005; Elzinga and Liefbroer, 2007; Widmer and Ritschard, 2009). The hypothesized changes were defined as pluralization processes, with life stages and transitions characterizing a smaller part of the population or occurring at increasingly dispersed chronological ages in younger cohorts.

Overall, researchers found important differences regarding pluralization of life trajectories across countries, genders and social classes. The empirical evidence suggested that rather than being a general trend that concerned all individuals and all life domains uniformly, pluralization took distinct shapes and followed distinct paces in various social groups and in various domains. These changes across and within cohorts in the ordering and temporality of life sequences may have had consequences for the life trajectories of individuals with psychiatric disorders. The weakening institutional constraints imposed on psychiatric patients, the increasing complexity of the organizations and programmes (Pavalko et al., 2007; Stimpson et al., 2006; Willits et al., 2004) as well as the pluralization process, may have led to higher complexity and a pluralization of the life trajectories of individuals with psychiatric problems.

Although some researchers have highlighted the relationship between family ties and mental health (Barrett, 2000), little research empirically addresses the variety of trajectories experienced by individuals with severe psychiatric problems. The overall societal trend towards a pluralization of the life course, however, strongly suggests that clinical populations have also achieved a greater variability of trajectories in recent decades, but the methodological tools and the data necessary to access this variability empirically and to make sense of it have been lacking.

Method

The data for this study comes from two distinct facilities in which systematic data collection about family relationships, psychological health and life trajectories occurred over a two-year period (Widmer et al., 2008a, 2008b, 2010). The first subsample includes 54 individuals with psychiatric disorders who were undergoing treatment at the rehabilitation unit of a psychiatric facility in the French-speaking part of Switzerland (subsample REHAB). The second sample includes 42 individuals who underwent psychotherapy in a private facility in the same area (subsample THERAP).

Individuals who did not answer the retrospective questionnaire concerning life trajectories were removed. The total sample comprised 86 individuals (49% women), 48 of them coming from the REHAB sample (35.4% women) and 38 from the THERAP (65.8% women). Individuals who were interviewed had a variety of severe mental disorders with significant differences between the two subsamples: patients suffering mainly from schizophrenia and from disorders diagnosed during childhood or adolescence were overrepresented in the REHAB sample, while patients suffering from depressive and anxiety disorders, and from personality disorders were more frequent in the THERAP sample (Table 1).

Table 1.

Main diagnoses established in the two subsamples (number of subjects)

	REHAB	THERAP
Schizophrenia and other psychotic disorders	22	6
Personality disorders	11	21
Disorders diagnosed during childhood or adolescence, often associated with mild retardation	9	0
Depressive and anxious disorders	6	11

Measurement of mental health

Each individual’s mental health was measured using the SCL-90-R (Derogatis, 1983; Derogatis et al., 1973), a multidimensional self-reported inventory commonly used by both researchers and clinicians. The answers to the 90 items in the questionnaire, rated on a five-point Likert-scale (ranging from 0 = ‘not at all’ to 4 = ‘extremely’), can be combined subsequently into nine dimensions: Somatization, Obsessive-Compulsive, Interpersonal Sensitivity, Depression, Anxiety, Anger-Hostility, Phobic Anxiety, Paranoid Ideation and Psychoticism. In addition, three indices can be computed, providing global measures of overall mental health: the Global Severity Index (GSI), which is a sum of all symptoms, the Positive Symptom Total (PST), which indicates the diversity of symptoms while considering only their presence or absence), and the Positive Symptom Distress Index (PSDI), which is the GSI divided by the PST, an index of psychological distress.

The mean values and reliability of the subscales and the GSI for this sample were compared with results obtained by Franke (1992) on a community reference sample constituted in the framework of a German validation study of the SCL-90-R. The internal consistency of the nine subscales was good for the whole sample, with Cronbach’s α ranging from 0.69 to 0.87. The reliability of the subscales was also good for the two subsamples separately, except for the Psychoticism subscale in the THERAP sample. As expected, on most of the subscales, the mean scores of the clinical samples (taken together or separately) were clearly higher than those of the reference sample, which is a result commonly found (Schmitz et al., 2000).

Measurements of life trajectories

Participants in the study answered a retrospective questionnaire on their occupational trajectory, their intimacy trajectory, and their co-residence trajectory. They were asked to provide detailed information about the sequences of jobs and cohabitations that they went through from the age of 18 onwards. As more detailed information was available for early intimate relationships, intimate trajectories began at age 16 and include sequences of partners. The data from the retrospective survey is in an ‘episode’ format: each respondent gave a date for the beginning and the end of every distinct period of partnership, occupation and cohabitation that he or she had experienced. On this basis, a sequential representation was created that indicated, for each individual at each age, the state the individual was experiencing. There are six distinct possible states in the occupational trajectory: education employment, internship, insurance, institutionalization, or taken care of by a relative. For intimacy trajectories, the states represent the order number of each partner that the person has had at each age, or 10 if he/she had no partner at that time. Starting from age 16, an intimacy trajectory written as ‘10-10-1-1-2-10-3’ means that the person had no partner until he/she was 18, then stayed with his or her first partner for two years, found another one for one year, again had no partner at 21 and then found a new partner at 22. Sequences end at the age the individuals were at the time of the interview, but at the latest when the individual reached age 34. This transformation of episode data into sequences of statuses is necessary when using the optimal matching method.

Optimal matching analysis was used to explore complex life trajectories and uncover typical life patterns; this is a sequence analysis technique, which helps researchers to categorize life trajectories empirically into distinct types (Abbott and Hrycak, 1990; Abbott and Tsay, 2000, Gauthier et al., 2009). The optimal matching analysis originated in the biological sciences and was recently imported to the social sciences for the study of life courses (Abbott and Tsay, 2000). It enables researchers to take into account the order, chronology and duration of states during a life trajectory. This technique is implemented as a dynamic programming algorithm that computes a distance between two sequences of states. This distance corresponds to the number of operations needed to transform one of the sequences into the other one. The distance computed by this algorithm can then be used with a clustering method in order to group sequences that are the most similar (i.e. the least distant). This method has already been used with success in several studies on non-clinical populations (Abbott and Hrycak, 1990; Elzinga and Liefbroer, 2007; McVicar and Anyadike-Danes, 2002; Müller et al., 2008; Pollock, 2007; Stovel and Bolan, 2004; Widmer et al., 2003). The distances are inputted in a hierarchical cluster analysis (Kaufmann and Rousseeuw, 1990) in order to create a typology of life trajectories.

Multiple life dimensions are combined with one another using a multi-channel approach of optimal matching (Gauthier et al., 2010). This methodological approach is designed to reveal the overall structure associating various life trajectories (e.g. co-residence, occupational and intimacy trajectories) without inferring a causal order, which is difficult to set a priori, as those dimensions belong to interacting fields, that is, phenomena that are interconnected rather than occurring independently. Analyses have been done using the TraMineR package for the R environment (Gabadinho et al., 2008).

Results

A multidimensional sequence analysis (Gauthier et al., in press; Pollock, 2007), which provides a clustering that takes the interactions of the three life trajectories into account, was used. An agglomerative hierarchical clustering method using the Ward criterion produced three distinct clusters. Each is characterized by a combination of the three trajectories of interest.

The clusters are represented by tables with the mean number of years spent in each state (Table 2 to Table 4) and by chronological bar plots. These plots represent, in each cluster, the distribution of the individuals in each state for each year in the trajectory (Figure 1 to Figure 3). Some states were merged in order to facilitate the graphical representation of the trajectories. For the activity trajectory, the employment and the internship states were merged into a unique employment state, and the ‘insurance’, ‘institutionalization’ and ‘taken care of by a relative’ states form a unique ‘Institution/Relatives’ state. The intimacy trajectory was recoded in three states: first and second partner, third and over, and alone. The co-residence trajectory was recoded in four states: the ‘Parents’ state groups all the states in which a parent is cited, the ‘Partner’ state is the merging of the ‘With a partner’, ‘With a partner and a biological child’ and ‘With a partner and the partner’s child’, the ‘Alone’ state represents the ‘Alone’ and ‘Alone with a child’ states, and the ‘Other’ state is the union of the ‘With friends’ and ‘Others’ states. This recoding scheme is only effective for the graphical representation; the sequence analysis is done on the full set of states.

Table 2.

Occupation trajectories, mean number of years in each state

	Institution (SD)	Unstable (SD)	Standard (SD)	Total (SD)
Training	2.81 (2.47)	5.48 (4.4)	1.58 (2.44)	3.15 (3.41)
Employment	1.14 (2.03)	4.09 (3.94)	10.77 (5.1)	4.84 (5.5)
Internsip	0.08 (0.36)	0.13 (0.46)	0.15 (0.61)	0.12 (0.47)
Insurance	0.81 (1.27)	0.74 (1.57)	0.88 (1.42)	0.81 (1.38)
Institution	3.62 (3.32)	0.39 (0.99)	0.69 (1.62)	1.87 (2.84)
Relative	0.16 (0.99)	0.74 (2.18)	0.85 (2.13)	0.52 (1.75)

Table 3.

Intimacy trajectories, mean number of years in each state by order of partners

Partners	Institution (SD)	Unstable (SD)	Standard (SD)	Total (SD)
1st	1.65 (1.44)	3 (2.88)	6.35 (5.99)	3.43 (4.2)
2nd	1.35 (1.49)	1.22 (1.44)	3.04 (3.38)	1.83 (2.34)
3rd	1.08 (1.89)	1.17 (2.98)	1.92 (2.74)	1.36 (2.48)
4th	0.54 (1.5)	0.39 (1.03)	1.42 (2.9)	0.77 (1.97)
5th	0.16 (0.55)	0.22 (0.67)	0.85 (2.17)	0.38 (1.31)
6th	0.14 (0.54)	0.04 (0.21)	0.15 (0.46)	0.12 (0.45)
7th	0 (0)	0 (0)	0.04 (0.2)	0.01 (0.11)
8th	0.03 (0.16)	0 (0)	0 (0)	0.01 (0.11)
9th	0.11 (0.66)	0 (0)	0.08 (0.39)	0.07 (0.48)
alone	6.46 (3.67)	7.74 (4.41)	3.42 (2.67)	5.88 (3.97)

Table 4.

Cohabitation trajectories, mean number of years in each state

Living with	Institution (SD)	Unstable (SD)	Standard (SD)	Total (SD)
Both parents	1.27 (2.19)	4.83 (3.37)	1.77 (1.95)	2.37 (2.89)
One parent	1.7 (2.74)	0.26 (0.75)	0.96 (2.31)	1.09 (2.29)
One parent with partner	0.03 (0.16)	0.17 (0.83)	0.04 (0.2)	0.07 (0.45)
Alone	2.57 (2.88)	2.61 (3.92)	2.38 (2.38)	2.52 (3.03)
Partner without children	0.19 (0.57)	2.09 (3.64)	4.73 (4.71)	2.07 (3.72)
Partner with biological children	0.24 (1.09)	0 (0)	3.69 (5.07)	1.22 (3.28)
Partner with partner’s children	0 (0)	0 (0)	0 (0)	0 (0)
Alone without children	0 (0)	0 (0)	0.27 (0.83)	0.08 (0.47)
Friends	0.32 (1.08)	0.48 (1.2)	0.58 (1.5)	0.44 (1.24)
Other	1.84 (3.07)	0.17 (0.65)	0.23 (0.65)	0.91 (2.21)

Figure 1.

Institutionalized cluster

Figure 2.

Standardized Life Course

Figure 3.

Unstable Life Course

Cluster 1: Institutionalized Life Course

The first cluster contains 43% of the sample (37 individuals) and is characterized by two main features: a lack of community integration in the fields of work, intimacy and cohabitation, and a prevalence of time spent in institutions. The lack of integration in the occupational trajectory is revealed by a low number of years in employment (1.14) versus a high number of years in institutions (3.62 years) (Table 2). The average number of years in an institution is significantly higher than in the other two clusters (Kruskal-Wallis χ² p < 0.001). Figure 1 shows that few individuals in this cluster managed to go from a ‘education’ state to employment, and that most of them ended up in the ‘Institution/Relatives’ state.

The lack of integration in the intimacy trajectories is denoted by a high mean number of years without a partner and a high number of different partners (Table 3). This is a clear indication of difficulties in establishing long-term relationships. This is also confirmed by the co-residence trajectories. In these trajectories, the average number of years with a partner is clearly lower than in the other two clusters. Figure 1 shows noticeably that some of these individuals stay with their parents or alone for a long time, and that almost 40% of them live alone at the age of 34. The high average number of years in the ‘other’ state is related to their frequent stays in medical institutions. Overall, this cluster includes individuals who did not integrate in social fields such as work and the family: they are not active in the labour market, they have numerous and short intimate relationships and have not moved in with a partner. Neither do they have children.

Cluster 2: Standardized Life Course

This Life Course represents 30% of the sample and is called Standardized because it contains individuals who have trajectories that are close to what has been designated as a ‘standard’ life course (Kohli, 1986): departure from the parental home in the early twenties, a stable partnership, children, and a long-term participation in the labour market. Trajectories in this cluster are more stable that those in the previous cluster. They are similar to those found in non-clinical samples, having a close match with the tripartition of occupational trajectories and the family life cycle (Widmer and Ritschard, 2009).

In the occupational trajectory, the main state is ‘employment’ as underlined by the mean number of years doing paid work (10.7). Individuals in this cluster are the most active in the labour market. Their intimacy trajectory is also characterized by a propensity for stability. Indeed, members of this cluster are more involved in long-term relationships, as their mean number of years with their first partner (Table 3) and the stability of the intimacy trajectory plot (Figure 2) indicates. This high mean number of years spent with their first partner, along with the low mean number of years spent without a partner, differentiates clearly the individuals belonging to this cluster from those belonging to the other two. The cohabitation trajectory is very close to those found in non-clinical samples (Widmer and Ritschard, 2009); starting from age 24, the co-residence trajectory seems very stable (Figure 2). It is also noteworthy that this is the only cluster in which respondents live with children.

Cluster 3: Unstable Life Course

The last type of Life Course is made up of 27% of the sample. Individuals belonging to this cluster are more often in employment than individuals in the Institutionalized life course and are almost never institutionalized. The rate of employment is lower compared to the Standardized cluster but it is balanced by a high mean number of years in the Training state. Regarding the intimacy trajectories, this cluster is situated midway between the other two. Indeed, individuals in this cluster had a longer relationship with their first partner than individuals in the Institutionalized cluster, but they also spent more time ‘alone’ than those from the other two clusters. Regarding their relatively long training periods (with a mean number of 5.48 years), individuals in this cluster spent long periods of time living with their parents and left the parental home later than individuals in the other clusters. The major features of this cluster are a lack of stability similar to those in the Institutionalized cluster but without the institutionalization associated with a long period of vocational training and almost no periods of employment. This instability is clearly represented in the chronological bar plots and their lack of structure (Figure 3) compared to the plots of the other two clusters.

Overall, the results show a diversity in life trajectories of individuals with psychiatric troubles, which is encapsulated in three distinct multidimensional trajectory types. None of these types is preponderant. While 70% of the respondents belong to types not found in non-clinical samples (Widmer and Ritschard, 2009), 30% of the respondents have followed a ‘standard’ life course, which implies an early departure from the parental home, a regular participation in the labour market, and cohabitation with a partner and sometimes children. The results of a logistic regression using the birth cohort as an explaining variable and the type of trajectories as a response variable show that the probability of belonging to the Standardized type instead of being in one of the other two types is about 5 times higher for individuals born before 1960 and about 7.5 times higher for individuals born between 1960 and 1970 than for individuals born after 1970.

Life trajectories and psychiatric symptoms

The mean differences shown in the symptom scales among types of occupation, intimacy or cohabitation trajectories were first tested using the Kruskal Wallis non-parametric test (Kruskal and Wallis, 1952). Despite the small sample size, several mean differences were significant at the 0.1 level. However, this type of analysis does not allow to control for factors such as gender or the sample of origin. Therefore, linear regression models were used, enabling the estimation of the impact of belonging to a certain trajectory while controlling for gender, sample of origin and year of birth. One model by symptom scale was fitted, using the multi-channel typology mentioned above as a predictor (Table 5).

Table 5.

Linear regression models for each symptom

	GSI	PSDI	Somatization	Obsessive compulsive	Interpersonal sensitivity	Depression
Birth year	1.263*	0.00917	0.190	0.114	0.0954	0.149
	(0.715)	(0.0074)	(0.118)	(0.114)	(0.108)	(0.155)
Women	28.43***	0.0993	2.255	4.238**	3.863**	6.745***
	(10.72)	(0.111)	(1.774)	(1.711)	(1.626)	(2.320)
THERAP	8.662	0.288	5.961*	−0.434	−1.581	2.276
	(19.53)	(0.202)	(3.230)	(3.115)	(2.960)	(4.225)
Institution	−35.08**	−0.338**	−4.251*	−1.680	−4.047*	−4.808
	(15.14)	(0.156)	(2.504)	(2.415)	(2.295)	(3.275)
Unstable	−23.69	−0.211	−4.228*	0.440	−0.436	−2.341
	(14.46)	(0.149)	(2.392)	(2.307)	(2.192)	(3.129)
Standard(ref)	0	0	0	0	0	0
Constant	−2398*	−16.09	−366.1	−214.5	−177.5	−279.1
	(1414)	(14.61)	(233.9)	(225.6)	(214.4)	(306.0)
R²	0.169	0.155	0.165	0.093	0.113	0.153
	Anxiety	Hostility	Phobia	Paranoia	Psychoticism	Other
Birth year	0.272***	0.0843	0.0524	0.126*	0.0665	0.114*
	(0.102)	(0.0678)	(0.0734)	(0.0704)	(0.0860)	(0.0636)
Women	1.780	0.846	1.895*	1.955*	2.083	2.769***
	(1.535)	(1.017)	(1.100)	(1.055)	(1.289)	(0.954)
THERAP	4.301	0.895	−0.178	0.406	−3.077	0.0938
	(2.796)	(1.851)	(2.003)	(1.921)	(2.348)	(1.737)
Institution	−4.790**	−3.010**	−2.730*	−3.908**	−2.366	−3.487**
	(2.168)	(1.435)	(1.553)	(1.490)	(1.820)	(1.347)
Unstable	−4.171**	−2.473*	−1.638	−3.948***	−2.226	−2.670**
	(2.071)	(1.371)	(1.484)	(1.423)	(1.739)	(1.287)
Standard(ref)	0	0	0	0	0	0
Constant	−525.6**	−160.2	−97.63	−238.4*	−121.6	−217.3*
	(202.5)	(134.1)	(145.1)	(139.2)	(170.1)	(125.8)
R²	0.151	0.087	0.080	0.165	0.113	0.192

Standard errors in parentheses

***

p < 0.01, ** p < 0.05, * p < 0.1

The reference category for the effect of the life trajectory is the Standardized cluster. Controlling for gender, sample of origin and birth year, there was a significant impact of trajectory types on the majority of symptom scales. All other things being equal, individuals in the Institutionalized cluster reported systematically and significantly lower scores on symptom scales than individuals from the Standardized cluster. The two global indices, GSI and PSDI, measuring respectively the sum of all symptoms and psychological distress, were lower for the Institutionalized cluster than for the Standardized cluster, but not significantly lower for the Unstable cluster, although the coefficient was negative. Individuals belonging to the Institutionalized cluster also had a significantly lower score than individuals belonging to the Standardized cluster on the Somatization, Interpersonal Sensitivity, Anxiety, Hostility, Phobia, Paranoia and Other scales. Individuals from the Unstable cluster had lower scores than individuals from the Standardized cluster on the Somatization, Anxiety, Hostility, Paranoia and Other scales. Differences between the Institutionalized and Unstable clusters were tested and were found to be not significant for all symptom scales (results not presented). The sole symptom scales that were not significantly associated with life trajectories are the Obsessive-Compulsive, Depression and Psychoticism scales. Overall, individuals belonging to the Standardized cluster reported clearly higher scores on symptoms than the others, even when controlling for major confounding variables.

Regarding the control variables, younger individuals were found to experience more symptoms as the influence of the birth year had a significant effect on the sum of all symptoms (GSI). The sample of origin had no significant influence on symptoms, except for the Somatization scale for which patients from the THERAP sample had a higher score. Many epidemiological studies show higher scores for women regarding self-reported distress, particularly regarding depression and anxiety (e.g. Turner et al., 1995) and this same effect had been previously found by Schmitz et al. (2000) in a German validation study of the SCL-90-R, although the differences were generally small. This effect was confirmed by the regression models: women reported more symptoms than men and had higher scores on several symptom scales. The highest difference was for depressive symptoms.

Discussion

Life patterns of individuals with psychiatric problems found in this study are distinct from those found in non-clinical samples (Widmer and Ritschard, 2009; Widmer et al., 2003). When compared with individuals from identical cohorts and gender from non-clinical samples in Switzerland, individuals who participated in this study had distinct profiles of occupation, cohabitation and intimacy trajectories. The normative sequencing of transitions characterizing life courses in non-clinical samples was significantly lower in this sample, with many individuals going back and forth between situations that are usually considered irreversible (independence from parents, finding stable employment, and so on). In addition, for many individuals from the two subsamples, some key transitions, such as leaving the parental nest, getting a job, forming a stable partnership, or having children, did not occur. The independence provided by having a separate household was in some cases not secured as individuals continued to live with their parents or resided in public facilities.

However, life courses of individuals with psychiatric problems are diverse. Indeed, the Standardized type represents only a minority of life courses. The ‘unstable’ life trajectories are more prevalent among younger cohorts of patients, thus providing hints that the social processes of pluralization of life courses observed in non-clinical life courses (Brückner and Mayer, 2005; Elzinga and Liefbroer, 2007; Widmer and Ritschard, 2009; Widmer et al., 2003) are also present among individuals with psychiatric problems. This pluralization of life courses implies that life courses of patients cannot be summarized by an ‘average’ life course, which is a misleading indicator of the variety of patients’ lives. Results also show that individuals with psychiatric symptoms who follow the standard model have significantly higher symptom and distress levels than individuals from the two other types.

This result leads to different but not mutually exclusive interpretations. According to the recovery approach using the strengths model (Davidson et al., 2009; Rapp, 2006), one must consider symptomatology as merely one dimension of mental health. For example, the progress of an individual is not measured by the disappearance of symptoms, but by his or her ability to deal with them. As Rapp (2006) underlined it, recovering people experience symptoms, but still, they can work and hence follow what we called a ‘standard’ life trajectory.

Another explanation stems from the link between life trajectories and social capital. Previous research on this sample (Widmer et al., 2008b) highlighted a strong relationship between one’s life trajectory and one’s family ties; indeed, life trajectories including a single partner and a regular work activity, as in the Standardized type, lead to a restricted nuclear family configuration, which implies less social capital and thus less social and emotional support (Widmer et al., 2008b). Individuals in the Standardized type are certainly lacking these kinds of support, which are useful to cope with mental illness (Cohen and Wills, 1985; Corrigan and Phelan, 2004).

Overall, the standard trajectory is close to life trajectories found in non-clinical samples, especially among older birth cohorts. The two other types should be understood both as expressions of the difficulty of individuals with psychiatric disorders to overcome normative transitions of the life course, and as the result of the pluralization processes occurring in late modernity. Indeed, one should stress that psychiatric patients are social actors who participate, in their own ways, in society at large and its ongoing process of change. Therefore, it is essential to take into account societal trends such as the pluralization of occupational and family trajectories in recent cohorts (Widmer and Ritschard, 2009) when dealing with individuals in therapeutic settings. Considering trajectories of patients in their social and historical context may help psychiatrists and health professionals to understand the social and cultural constraints set on their patients’ lives.

Limitations

This research has several limitations that should be noted. First, it proposes no more than an exploration of the ways in which individuals with psychiatric disorders have built their family and occupational trajectories. As such, the small sample size might be a problem if one wishes to infer conclusions from the sample to the whole population of individuals with psychiatric disorders. The primary goal of this study was to raise an interest in the variety of ways in which individuals with psychiatric problems may evolve throughout their lives in fundamental domains such as family, work and intimate relationships. Indeed, such a detailed assessment as that proposed here is very difficult to reach in standard epidemiological studies that do not collect information on all dimensions of the patients’ lives.

Conclusion

The results of this study should be considered as an enticement to know more rather than as a confirmatory study about how work, family, intimacy and psychiatric disorders interact with each other. The link between trajectories and psychiatric symptoms requires additional analyses of larger samples. In particular, the issue of the causal mechanisms relating life trajectories and psychiatric symptoms should be carefully considered, as the various dimensions constituting one’s life course are deeply intertwined.

References

Abbott

Hrycak

(1990) ‘Measuring Resemblance in Sequence Data: An Optimal Matching Analysis of Musicians’ Careers.’ American Journal of Sociology 96: 144–185.

Abbott

Tsay

(2000) ‘Sequence Analysis and Optimal Matching Methods in Sociology.’ Sociological Methods Research 29: 3–33.

Barrett

A.E.

(2000) ‘Marital Trajectories and Mental Health.’ Journal of Health and Social Behavior 41: 451–464.

Beck

(1992) Risk Society: Towards a New Modernity. Thousand Oaks, CA: Sage.

Beck

Beck-Gernsheim

(2002) Individualization: Institutionalized Individualism and its Social and Political Consequences. Thousand Oaks, CA: Sage.

Becker

(1986) ‘Life Events as Predisposing and Precipitating Factors for Mental Disorders’, in Katschnig

(ed.) Life Events and Psychiatric Disorders: Controversial Issues. Cambridge: Cambridge University Press.

Brückner

Mayer

K.U.

(2005) ‘The De-Standardization of the Life Course: What It Might Mean? And if It Means Anything, Whether It Actually Took Place?’, in Macmillan

(ed.) Advances in Life Course Research, Vol. 9, The Structure of the Life Course: Standardized? Individualized? Differentiated?. Amsterdam: Elsevier.

Cherlin

A.J.

Chase-Lansdale

P.L.

McRae

(1998) ‘Effects of Parental Divorce on Mental Health throughout the Life Course.’ American Sociological Review 63: 239–249.

Cohen

Wills

T.A.

(1985) ‘Stress, Social Support and the Buffering Hypothesis.’ Psychological Bulletin 98: 310–357.

10.

Cook

J.A.

Cohler

B.J.

Pickett

S.A.

Beeler

J.A.

(1997) ‘Life Course and Severe Mental Illness: Implications for Caregiving within the Family of Later Life.’ Family Relations 46: 427–436.

11.

Corrigan

P.W.

Phelan

S.M.

(2004) ‘Social Support and Recovery in People with Serious Mental Illnesses.’ Community Mental Health Journal 40: 513–523.

12.

Dannefer

(2003) ‘Cumulative Advantage/Disadvantage and the Life Course: Cross-Fertilizing Age and Social Science Theory.’ Journal of Gerontology Series B: Psychological Sciences and Social Sciences 58: S327–S337.

13.

Davidson

Tondora

Staeheli Lawless

O’Connell

M.J.

Rowe

(2009) A Practical Guide to Recovery-Oriented Practice. Tools for Transforming Mental Health Care. New York, NY: Oxford University Press.

14.

Derogatis

L.R.

(1983) Administration, Scoring and Procedures (2nd edn). Baltimore, MD: Clinical Psychometric Research.

15.

Derogatis

L.R.

Lipman

R.S.

Covi

(1973) ‘SCL-90: An Outpatient Psychiatric Rating Scale: Preliminary Report.’ Psychopharmacological Bulletin 9: 13–28.

16.

Elder

G.H.

(1999) Children of the Great Depression (25th Anniversary Edition). Boulder, CO: Westview Press.

17.

Elder

G.H.

Johnson.

M.K.

Crosnoe

(2003) ‘The Emergence and Development of Life Course Theory’, in Mortimer

J.T.

Shanahan

M.J.

(eds) Handbook of the Life Course. New York, NY: Kluwer Academic.

18.

Elzinga

C.H.

Liefbroer

A.C.

(2007) ‘De-Standardization of Family-Life Trajectories of Young Adults: A Cross-National Comparison using Sequence Analysis.’ European Journal of Population 23: 225–250.

19.

Ferraro

K.F.

Thorpe

R.J.

McCabe

G.P.

Kelley-Moore

J.A.

Jiang

(2006) ‘The Color of Hospitalization over the Adult Life Course: Cumulative Disadvantage in Black and White?’ Journal of Gerontology Series B: Psychological Sciences and Social Sciences 61: S299–S306.

20.

Franke

G.H.

(1992) SCL-90-R. Die Symptom Checklist-90 Revised. Minneapolis, MN: NCS Assessment.

21.

Gabadinho

Ritschard

Studer

Müller

N.S.

(2008) Mining Sequence Data in R With the TraMineR Package: A User’s Guide. Geneva: University of Geneva.

22.

Gauthier

J.-A.

Widmer

E.D.

Bucher

Notredame

(2009) ‘How Much Does It Cost? Optimization of Costs in Sequence Analysis of Social Science Data.’ Sociological Methods & Research 38(1): 197–231.

23.

Gauthier

J.-A.

Widmer

E.D.

Bucher

Notredame

(2010) ’Multichannel Sequence Analysis Applied to Social Science Data.’ Sociological Methodology 40(1): 1–38.

24.

Gilman

S.E.

Kawachi

Fitzmaurice

Buka

(2003) ‘Socioeconomic Status, Family Disruption and Residential Stability in Childhood: Relation to Onset, Recurrence and Remission of Major Depression.’ Psychological Medicine 33: 1341–1355.

25.

Goodyer

I.M.

(1996) ‘Recent Undesirable Life Events: Their Influence on Subsequent Psychopathology.’ European Child & Adolescent Psychiatry 5: 33–37.

26.

Harding

C.M.

(1988) ‘Course Types in Schizophrenia: An Analysis of European and American Studies.’ Schizophrenia Bulletin 14: 633–643.

27.

Heim

Nemeroff

C.B.

(2001) ‘The Role of Childhood Trauma in the Neurobiology of Mood and Anxiety Disorders: Preclinical and Clinical Studies.’ Biological Psychiatry 49: 1023–1039.

28.

Holmes

T.H.

Rahe

R.H.

(1967) ‘The Social Readjustment Rating Scale.’ Journal of Psychosomatic Research 11: 213–218.

29.

Kaufmann

Rousseeuw

P.J.

(1990) Finding Groups in Data: An Introduction to Cluster Analysis. New York, NY: John Wiley and Sons.

30.

Kendler

K.S.

Karkowski

L.M.

Prescott

C.A.

(1999) ‘Causal Relations between Stressful Life Events and the Onset of Major Depression.’ American Journal of Psychiatry 156: 837–841.

31.

Kessler

R.C.

(1997) ‘The Effects of Stressful Life Events on Depression.’ Annual Review of Psychology 48: 191–214.

32.

Kohli

(1986) ‘The World We Forgot: A Historical View of the Life Course’, in Marshall

(ed.) Later Life: The Social Psychology of Aging. London: Sage.

33.

Kruskal

W.H.

Wallis

W.A.

(1952) ‘Use of Ranks in One-Criterion Variance Analysis.’ Journal of the American Statistical Association 47: 583–621.

34.

McVicar

Anyadike-Danes

(2002) ‘Predicting Successful and Unsuccessful Transitions from School to Work by using Sequence Methods.’ Journal of the Royal Statistical Society: Series A (Statistics in Society) 165: 317–334.

35.

Mirowsky

(1999) ‘Analyzing Associations between Mental Health and Social Circumstances’, in Aneshensel

Carol S.

Phelan

Jo C.

(eds) Handbook of the Sociology of Mental Health. New York, NY: Springer Science.

36.

Modell

Furstenberg

F.F.

Hershberg

(1976) ‘Social Change and Transitions to Adulthood in Historical Perspective.’ Journal of Family History 1: 7–32.

37.

Müller

N.S.

Gabadinho

Ritschard

Studer

(2008) ‘Extracting Knowledge from Life Courses: Clustering and Visualization’, in Song

I.-Y.

Eder

Nguyen

T.M.

(eds) Lectures Notes in Computer Sciences, Vol. 5182, Data Warehousing and Knowledge Discovery, Proceedings of DaWaK, Torino, Italy, September 1–5, 2008. Berlin: Springer.

38.

Pavalko

E.K.

Harding

C.M.

Pescosolido

B.A.

(2007) ‘Mental Illness Careers in an Era of Change.’ Social Problems 54: 504–544.

39.

Penza

K.M.

Heim

Nemeroff

C.B.

(2003) ‘Neurobiological Effects of Childhood Abuse: Implications for the Pathophysiology of Depression and Anxiety.’ Archives of Women’s Mental Health 6: 15–22.

40.

Pollock

(2007) ‘Holistic Trajectories: A Study of Combined Employment, Housing and Family Careers by using Multiple-Sequence Analysis.’ Journal of the Royal Statistical Society: Series A (Statistics in Society) 170: 167–183.

41.

Rapp

(2006) The Strengths Model: Case Management with People with Psychiatric Disabilities. New York, NY: Oxford University Press.

42.

Ross

C.E.

Mirowsky

(1999) ‘Parental Divorce, Life-Course Disruption and Adult Depression.’ Journal of Marriage and the Family 61: 1034–1045.

43.

Ross

C.E.

C.-L.

(1996) ‘Education, Age, and the Cumulative Advantage in Health.’ Journal of Health and Social Behavior 37: 104–120.

44.

Scheid

T.L.

Anderson

(1995) ‘Clinic Care Update. Living with Chronic Mental Illness: Understanding the Role of Work.’ Community Mental Health Journal 31: 163–176.

45.

Schmitz

Hartkamp

Kiuse

Franke

Reister

Tress

(2000) ‘The Symptom Check-List 90 (SCL-90-R): A German Validation Study.’ Quality of Life Research 9: 185–193.

46.

Sennett

(2000) The Corrosion of Character: The Personal Consequences of Work in the New Capitalism. New York, NY: W.W. Norton & Co.

47.

Shanahan

M.J.

(2000) ‘Pathways to Adulthood in Changing Societies: Variability and Mechanisms in Life Course Perspective.’ Annual Review of Sociology 26: 667–692.

48.

Singer

Ryff

C.D.

Carr

Magee

W.J.

(1998) ‘Linking Life Histories and Mental Health: A Person-Centered Strategy.’ Sociological Methodology 28: 1–51.

49.

Stimpson

J.P.

Peek

M.K.

Markides

K.S.

(2006) ‘Depression and Mental Health among Older Mexican American Spouses.’ Aging & Mental Health 10: 386–393.

50.

Stovel

Bolan

(2004) ‘Residential Trajectories: Using Optimal Alignment to Reveal the Structure of Residential Mobility.’ Sociological Methods and Research 32: 559–598.

51.

Turner

R.J.

Lloyd

D.A.

(1995) ‘Lifetime Traumas and Mental Health: The Significance of Cumulative Adversity.’ Journal of Health and Social Behavior 36: 360–376.

52.

Turner

R.J.

Wheaton

Lloyd

D.A.

(1995) ‘The Epidemiology of Social Stress.’ American Sociological Review 60: 104–125.

53.

Wheaton

(1990) ‘Life Transitions, Role Histories, and Mental Health.’ American Sociological Review 55: 209–223.

54.

Widmer

E.D.

Kempf

Galli-Carminati

(2010) ‘Do We See It the Same Way? Exploring Perceptions of their Family by Individuals with Intellectual Disability and Psychiatric Disorders.’ Families in Society 91(4): 378–384.

55.

Widmer

E.D.

Levy

Pollien

Hammer

(2003) ‘Entre Standardisation, Individualisation et Sexuation: Une Analyse des Trajectoires Personnelles en Suisse.’ Revue Suisse de Sociologie 29: 35–67.

56.

Widmer

E.D.

Kempf-Constantin

Robert-Tissot

Lanzi

Galli-Carminati

(2008a) ‘How Central and Connected Am I in my Family? Family-Based Social Capital of Individuals with Intellectual Disability.’ Research in Developmental Disabilities 29: 176–187.

57.

Widmer

E.D.

Orita

Gauthier

J.-A.

Sénac

Cucchiam

A.T.

Stekel

(2008b) ‘How Central and Connected Am I in my Family? Bridging and Bonding Social Capital in Family Configurations of Young Adults with Psychiatric Disorders.’ International Journal of Social Psychiatry 54: 550–561.

58.

Widmer

E.D.

Ritschard

(2009) ‘The De-Standardization of the Life Course: Are Men and Women Equal?’ Advances in Life Course Research 14: 28–39.

59.

Willits

Benzeval

Stansfeld

(2004) ‘Partnership History and Mental Health over Time.’ Journal of Epidemiology and Community Health 58: 53–55.

60.

Willson

A.E.

Shuey

K.M.

Elder

G.H.

(2007) ‘Cumulative Advantage Processes as Mechanisms of Inequality in Life Course Health.’ American Journal of Sociology 112: 1886–1924.