Depression or anxiety symptoms associated with occupational role transitions in Brazilian adults with a traumatic spinal cord injury: A multivariate analysis

Abstract

BACKGROUND:

Psychological morbidity is commonly experienced by people with a spinal cord injury (SCI), but whether it is associated with occupational role transitions in is unknown.

OBJECTIVE:

To analyze whether anxiety or depression symptoms are independently associated to increased likelihoods of role transitions in adults with SCI.

METHODS:

Cross-sectional study; multivariate analysis using a heteroscedastic Dirichlet regression. Participants: Thirty persons with traumatic SCI. Measures: Role Checklist (e.g. role transitions: dependent variables) and Beck’s Depression Inventory and State-Trait Anxiety Inventory (independent variables), adjusted for socio-demographic, functional, and injury-level confounders.

RESULTS:

Greater depression symptoms independently increased the likelihood of occupational role transitions, either for roles loss [adjusted Odds Ratio (AOR): 1.04; 95% confidence interval (CI):1.009–1.080] or roles gain [AOR: 1.07; 95% CI:1.02–1.13], as opposed to continued occupational roles. Higher anxiety as a trait, in turn, independently reduced the likelihood of occupational roles gain [AOR: 0.93; 95% CI: 0.869–0.992]. The “worker” role was the one lost more frequently (83%).

CONCLUSION:

Psychological morbidity is associated to occupational role transitions, as opposed to continued roles. Further research (e.g. with larger samples, longitudinal design, using structural equation modelling) should elucidate on the intricate relationships between mental health status and occupational role transitions in people experiencing SCI.

Keywords

Mental health role performance worker

1 Background

Psychological morbidity is commonly experienced by people with a spinal cord injury (SCI), and more so than by the general population [1 –3]. For instance, studies have showed that, at any given time, nearly half of the persons with SCI experience at least mild depressive symptoms or mood disturbance [2, 4]. In mainland China, for instance, a recent study found that 35%, 29%, and 27%of the people that experienced a SCI exceeded a clinical cutoff for post-traumatic stress disorder, anxiety, and depression, respectively [5]. In Taiwan, a population-based coh-ort study showed that persons with traumatic SCI had a high risk of anxiety or depression post-discharge, compared with persons with other health conditions [6]. Finally, according to a recent meta-analysis, the prevalence of anxiety in people experiencing SCI ranged from 15 to 32%, depending on the individual self-report measures used [7].

In people experiencing SCI, psychological morbidity has been associated to a myriad of somatic and non-somatic determinants, such as, but not limited to pain [8, 9], lower self-efficacy to manage the effects of SCI [10], lower body image [11], passive (e.g. idealization) coping responses or defense mechanisms [12], financial strain and poor social relationships [13], and fewer rewarding activities [10]. In turn, psychological morbidity has been found to be a predictor of worsened aggregative psychosocial outcomes, such as subjective well-being, life sat-isfaction, and health-related quality of life [8 , 14–16]. Finally, a myriad of personal resources, adaptive coping responses, and ‘positive psychology’ variables (e.g. post-traumatic growth) have been found to predict better long-term psychosocial outcomes in people experiencing SCI [3 , 17–20].

In short, there is a complex interplay among mental health variables, their determinants, and aggregative psychosocial outcomes in people experiencing a SCI, which leads to unique trajectories on the individual’s psychosocial adjustment [16, 19]. However, despite these advances, much less is known about the effects of psychosocial variables and morbidities in occupational role changes among people experiencing a SCI.

While a number of studies have been addressing employment pathways or outcomes in people experiencing SCI, often in the context of vocational rehabilitation [21 –25], the broader spectrum of occupational role changes in people experiencing SCI have not been examined, especially when associated to anxiety or depression symptomatology. Hence, this study aims to examine:

The prevalence of anxiety and depression symptoms in adults experiencing SCI for over six months as well as receiving home-based services in a sub-urban, resource-poor location in Brazil;

Whether and which occupational role changes occurred in this population from the pre-injury period, including the examination of the meaningfulness of those occupational roles; and

Whether the presence of anxiety and depression symptoms are independently associated to an increased likelihood of occupational role transitions in this population.

2 Methods

2.1 Study design

Cross-sectional, quantitative study with a multi-va-riate analysis accounting for different occupational role transitions (dependent variables) and both dep-ression and anxiety (independent variables), adjusted for socio-demographic, functional, and injury-related factors.

2.2 Settings and participants

People experiencing a traumatic SCI that also: were adults (≥18 years), had over six months post injury (i.e. not undergoing an acute or post-acute rehabilitation), did not experience a pediatric-onset of the SCI (e.g. which could interfere with pre-injury roles), lived in the community, and received home-based care services from the only public provider in São José do Rio Preto, São Paulo, Brazil. The service covered suburban, socio-economically deprived locations.

2.3 Procedures

Clinical charts from the home-based service were reviewed for the previous five years to identify eligible participants. All those identified as eligible were invited by phone to participate in the study. For those accepting, one research author (AP) collected the data in single encounters in each participants’ home. All the participants signed informed consents. The study was approved by Research Ethics Committee of the Nursing School of Ribeirão Preto.

2.4 Measures

The Role Checklist is a self-administered tool that assesses performance in, and value attributed to, ten different occupational roles;[26] we used the version from Cordeiro et al. [27] adapted to the Brazilian Portuguese. In the performance subscale, participants classify whether each role has been conducted in the past (i.e. pre-injury), present time, and is intended to be conducted in the future. According to the classification schema (see Fig. 1), these ratings allow to determine which occupational roles have been continuous or absent all across the time periods analyzed, or entail a form of occupational transition: loss, gain, or change. These three latter are our result (i.e. dependent) variables.

Fig. 1

Diagram with the classification of the occupational roles according to the past, present and future tenses.

Depression and anxiety measures were, in turn, our independent or predictor variables.

For depression, we used the Beck’s Depression Inventory (BDI), a widely used measure of depression in clinical research, including SCI research [28], composed of 21 items quantifying the intensity of depressive symptoms [29, 30]. We used the version and cutoffs validated for the Brazilian population [31].

For anxiety, we used the State-Trait Anxiety Inventory (STAI) [32]; the version translated and validated for the Brazilian population [33]. The STAI is a measure to quantify subjective components related to anxiety [34], used for example with people with SCI in Latin America [35 –37]. The Inventory has two different scales: one for identifying trait, another for the anxiety state. We use only the trait subscale, i.e. referring to more stable anxiety symptoms [38, 39], whose scores range from asymptomatic/mild to severe anxiety.

As possible confounders, we collected data on de-mographics (age, gender, marital status, educational level, and employment status both pre- and post-in-jury), injury characteristics (lumbar/thoracic versus cervical injury; time since injury), and finally about functional status. The latter refers to a single item on dependency versus independency on the activities of daily living, determined by the interviewer, occupational therapist by background, based on a brief, semi-structured interview with the patient, complemented by observation as needed.

Finally, data was also collected on socio-economic variables: individual’s income before and after injury, and current household income. For the latter, we used the Brazilian Economic Classification Criteria (BECC), a 10-items measure classifying household income in one of 6 economic grades [40].

The choice of these possible confounders, i.e. capable of interfering with both the dependent and independent variables, was grounded in the literature. For people experiencing SCI, evidence shows that age, educational level, gender, and time since injury can interfere with both mood or anxiety [10 , 42]. Evidence also shows that financial strain has been associated to poorer mental health [13]. Finally, impairment levels arguably can interfere with mental health, occupational roles, or both.

All the collected data was typed twice in Excel spreadsheets and verified for typing consistency before transference to the “R” program for statistical analysis.

2.5 Statistical analysis

The outcomes from the Role Checklist (occupation role gain, loss, etc.) were treated as compositional data: quantitative descriptions of the parts of some whole (i.e. a proportion), thus conveying relative information [43]. In this context, a suitable approach for the multi-variate analysis is a Dirichlet Regression Model, applied through the “R” program, version 3.5.3, with the DirichletReg package. Dirichlet regression models can be used to analyze a set of variables lying in a bounded interval that sum up to a constant (e.g., proportions, rates, compositions, etc.) exhibiting skewness and heteroscedasticity, without having to transform the data [43 –45].

Concretely, within the Dirichlet regressions, we apply 3 different equations for each type of occupational transition (i.e. occupational role “loss”, “change”, and “gain”), each one against the “continuous” occupational roles; the neutral or reference category - i.e. no change in occupational roles. Of note, the occupational roles that were “absent” were not considered [44].

Within these options, we applied three different statistical models: 1) including all the possible confounders theoretically selected; 2) including only confounders selected by Likelihood test (i.e. the most significant in empirical terms); and finally 3) a model accounting for the dispersion (i.e. heteroscedasticity) of the data. Likelihood-ratio tests compared the nested models (i.e. all the theoretically-selected confounders measured versus the empirically-selected confounders, and then the empirically-selected confounders versus the model accounting for the data dispersion). The model accounting for the dispersion (i.e. heteroscedasticity) fitted better with data, thereby it was finally used for the analysis.

3 Results

Of the 33 eligible persons identified, 30 partici-pated in the study; two refused and one was unavailable.

Table 1 provides their socio-demographics. The st-udy engaged 30 persons (male: n = 26; 86.7%) with traumatic SCI. The ages ranged between 19 and 72 years (mean: 40.8±14.4). The time since injury ranged between 6 months and 40 years (mean: 09.4±11.0), and most (73.3%) had household income of a relatively low level.

Table 1
Socio-demographic and injury data of people experiencing a SCI (n = 30)

Variable Frequency Mean/SD

Gender

Woman 4 (13.3%)

Men 26 (86.7%)

Age

19–29 8 (26.7%)

29–39 10 (33.3%) 40.8±14.4

> 40 12 (40%)

Marital status

Married 13 (43.3%)

Singles/divorced 17 (56.7%)

Economic classification criteria (A=-higher income level)^*

B1/B2 8 (26.7%)

C1/C2 21 (70%)

D-E 1 (3.3%)

Income

Previous R$ 2390,00±1659,321^

Current R$ 1642,32±936,325^

Education level

No formal education 1 (3.3%)

Elementary/high 26 (86.7%)

Technical/incomplete graduation 3 (10%)

Professional occupation

Before injury (no) 1 (3.3%)

Before injury (yes) 29 (96.7%)

Time/level of injury

<1 year 2 (6.6%)

1–5 years 14 (46.7%) 09.4±11.0

> 5 years 14 (46.7)

Lumbar 1 (3.3%)

Cervical 12 (40%)

Thoracic 17 (56.7%)

Dependence: activities of daily living.

Yes 6 (20%)

No 24 (80%)

Variable	Frequency	Mean/SD
Gender
Woman	4 (13.3%)
Men	26 (86.7%)
Age
19–29	8 (26.7%)
29–39	10 (33.3%)	40.8±14.4
> 40	12 (40%)
Marital status
Married	13 (43.3%)
Singles/divorced	17 (56.7%)
Economic classification criteria (A=-higher income level)^*
B1/B2	8 (26.7%)
C1/C2	21 (70%)
D-E	1 (3.3%)
Income
Previous		R$ 2390,00±1659,321^**
Current		R$ 1642,32±936,325^**
Education level
No formal education	1 (3.3%)
Elementary/high	26 (86.7%)
Technical/incomplete graduation	3 (10%)
Professional occupation
Before injury (no)	1 (3.3%)
Before injury (yes)	29 (96.7%)
Time/level of injury
<1 year	2 (6.6%)
1–5 years	14 (46.7%)	09.4±11.0
> 5 years	14 (46.7)
Lumbar	1 (3.3%)
Cervical	12 (40%)
Thoracic	17 (56.7%)
Dependence: activities of daily living.
Yes	6 (20%)
No	24 (80%)

Class A: 45-100 points: Annual family income R$ 20.888 or higher; Class B1:38-44 points: Family income R$ 9.254; Class B2:29-37 points: Family income R$ 4.852; Class C1:23-28 points: Family income R$ 2.705; Class C2:17-22 points: Family income R$ 1.625; Class D-E: 0-16 points: Family income R$ 768 (ABEP, 2016).^** R$- Brazilian currency, ^*ADLs.

Finally, nearly all participants were employed pre-injury (n = 29; 96.7%), but only four (13.3%) remained employed at the time of the interview; another four participants (13.3%) have reached the retirement age meanwhile.

3.1 Prevalence of depression and anxiety symptomatology

Table 2 shows that most persons (n = 16; 53.3%) had some level of depression symptoms, but many (n = 11) only mild/moderate symptoms. In turn, nearly two-thirds of the participants (n = 19; 63.4%) had moderate to severe anxiety as a trait.

Table 2
Depression and anxiety level of people experiencing a SCI (n = 30)

Depression Frequency Anxiety-trait Frequency

No/minimal depression (0 to 9) 14 (46.7%) Asymptomatic/mild (< 33) 11 (36.7%)

Light/moderate (10 to 18) 11 (36.7%) Moderate (33 to 49) 17 (56.6%)

Moderate/severe (19 to 29) 4 (13.3%) Severe (> 49) 2 (6.7%)

Severe (30 to 63) 1 (3.3%)

Total Median 10.5 Total Median 39.0

Total Mean 11.8±7.8 Total Mean 37.7±7.9

Depression	Frequency	Anxiety-trait	Frequency
No/minimal depression (0 to 9)	14 (46.7%)	Asymptomatic/mild (< 33)	11 (36.7%)
Light/moderate (10 to 18)	11 (36.7%)	Moderate (33 to 49)	17 (56.6%)
Moderate/severe (19 to 29)	4 (13.3%)	Severe (> 49)	2 (6.7%)
Severe (30 to 63)	1 (3.3%)
Total Median 10.5		Total Median 39.0
Total Mean 11.8±7.8		Total Mean 37.7±7.9

3.2 Occupational roles –frequent changes and their meaningfulness

Table 3 shows that occupational loss occurred frequently (n = 25, 83%) for the “worker” role - which is of “much importance” to 70%of the participants, and to a lesser degree (33.3%) for the “Hobbyist/amateur” role - which is of “much importance” to 60%.

Table 3
Time and importance of occupational roles of people experiencing a SCI (n = 30)

Role time and importance Student Worker Volunteer Caregiver Home maintainer Friend Family member Religious participant Hobbyst/amateur Participant in organization

Loss1 5 (16.7%) 3 (10%) 6 (20%) 4 (13.3%) 3 (10%) 1 (3.3%) 3 (10%) 4 (13.3%)

Loss 2 20 (66.7%) 4 (13.3%) 3 (10%) 5 (16.7%) 7 (23.3%) 2 (6.7%)

Gain 1

Gain 2 2 (6.7%) 3 (10%) 4 (13.3%) 2 (6.7%) 1 (3.3%) 1 (3.3%) 1 (3.3%)

Change 9 (30%) 1 (3.3%) 11 (36.7%) 4 (13.3%) 1 (3.3%) 1 (3.3%) 4 (13.3%) 3 (10%) 1 (3.3%)

Absent 19 (63.3%) 7 (23.3%) 6 (20%) 6 (20%) 1 (3.3%) 1 (3.3%) 22 (73.3%) 4 (13.3%) 19 (63.3%)

Continuous 1 1 (3.3%)

Continuous 2 4 (13.3%) 2 (6.7%) 6 (20%) 12 (40%) 25 (83.4%) 28 (93.4%) 3 (10%) 12 (40%) 3 (10%)

No importance 8 (26.7%) 2 (6.7%) 5 (16.7%) 2 (6.7%) 1 (3.3%) 8 (26.7%)

Such importance 13 (43.3%) 9 (30%) 19 (63.3%) 12 (40%) 16 (53.3%) 5 (16.7%) 1 (3.3%) 7 (23.3%) 11 (36.7%) 18 (60%)

Much importance 9 (30%) 21 (70%) 11 (36.7%) 16 (53.3%) 9 (30%) 25 (83.3%) 29 (96.7%) 21 (70%) 18 (60%) 4 (13.3%)

Role time and importance	Student	Worker	Volunteer	Caregiver	Home maintainer	Friend	Family member	Religious participant	Hobbyst/amateur	Participant in organization
Loss1		5 (16.7%)	3 (10%)	6 (20%)	4 (13.3%)	3 (10%)	1 (3.3%)		3 (10%)	4 (13.3%)
Loss 2		20 (66.7%)	4 (13.3%)	3 (10%)	5 (16.7%)				7 (23.3%)	2 (6.7%)
Gain 1
Gain 2	2 (6.7%)		3 (10%)	4 (13.3%)	2 (6.7%)			1 (3.3%)	1 (3.3%)	1 (3.3%)
Change	9 (30%)	1 (3.3%)	11 (36.7%)	4 (13.3%)	1 (3.3%)	1 (3.3%)		4 (13.3%)	3 (10%)	1 (3.3%)
Absent	19 (63.3%)		7 (23.3%)	6 (20%)	6 (20%)	1 (3.3%)	1 (3.3%)	22 (73.3%)	4 (13.3%)	19 (63.3%)
Continuous 1			1 (3.3%)
Continuous 2		4 (13.3%)	2 (6.7%)	6 (20%)	12 (40%)	25 (83.4%)	28 (93.4%)	3 (10%)	12 (40%)	3 (10%)
No importance	8 (26.7%)			2 (6.7%)	5 (16.7%)			2 (6.7%)	1 (3.3%)	8 (26.7%)
Such importance	13 (43.3%)	9 (30%)	19 (63.3%)	12 (40%)	16 (53.3%)	5 (16.7%)	1 (3.3%)	7 (23.3%)	11 (36.7%)	18 (60%)
Much importance	9 (30%)	21 (70%)	11 (36.7%)	16 (53.3%)	9 (30%)	25 (83.3%)	29 (96.7%)	21 (70%)	18 (60%)	4 (13.3%)

Occupational gain was observed for 13%of the participants for the “caretaker” role - which is of “much importance” to 53.3%of the participants.

An intended change (i.e. role not performed yet but intended for the future) was mentioned by 36.7%of the participants for the “volunteer” role - which is of “some” or “much” importance to all the participants.

A “continuous” role performance, in turn, was common for being a “family member” (93.3%) and a “friend” (83.3%).These roles are of “much importance” for 96.7%and 83.3%of the participants, respectively.

3.3 Depression and anxiety symptoms as independent predictors of occupational role changes

Table 4 shows that depression symptomatology was significantly and independently associated to an increased likelihood of occupational role transitions either for occupational roles loss or for occupational roles gain [adjusted OR: 1.04; 95% confidence interval (CI): 1.009–1.080] or for occupational roles gain [adjusted OR: 1.07; 95% CI: 1.02–1.13], as opposed to continuous (i.e. maintained) occupational roles.

Table 4
Whether and by how much depression and anxiety are independent predictors of occupational role transitions using a heteroscedastic Dirichlet regression model, adjusted for relevant confounders (n = 30)

Loss Estimate SE z value Pr(> |z|) OR 95%CI (lower) 95%CI (upper)

Depression 0.043 0.0175 2.4594 0.0139^* 1.044 1.0088 1.0804

Anxiet-trait 0.0478 0.0245 1.9519 0.0509 1.049 0.9998 1.1006

Gain

Depression 0.0702 0.0281 2.4988 0.0125^* 1.0727 1.0153 1.1335

Anxiety-trait –0.0745 0.0339 –2.1966 0.0280^* 0.9282 0.8685 0.992

Change

Depression 0.0229 0.0259 0.8846 0.3764 1.0232 0.9725 1.0765

Anxiety-trait –0.0112 0.0284 –0.395 0.6928 0.9888 0.9352 1.0455

Loss	Estimate	SE	z value	Pr(> \|z\|)	OR	95%CI (lower)	95%CI (upper)
Depression	0.043	0.0175	2.4594	0.0139^*	1.044	1.0088	1.0804
Anxiet-trait	0.0478	0.0245	1.9519	0.0509	1.049	0.9998	1.1006
Gain
Depression	0.0702	0.0281	2.4988	0.0125^*	1.0727	1.0153	1.1335
Anxiety-trait	–0.0745	0.0339	–2.1966	0.0280^*	0.9282	0.8685	0.992
Change
Depression	0.0229	0.0259	0.8846	0.3764	1.0232	0.9725	1.0765
Anxiety-trait	–0.0112	0.0284	–0.395	0.6928	0.9888	0.9352	1.0455

Legend: ^*Statistically significant relationship with a 95%confidence level.

Lower anxiety as a trait, in turn, was an independent, negative predictor (i.e. decreased the likelihood) of occupational roles gain. It was not independent predictor of occupational roles loss, albeit marginally related (p = 0.051).

No significant, independent associations were found on the effect of depression symptoms or anxiety as a trait on occupational roles change, i.e. a role not played in the past or present but intended for the future.

4 Discussion

This study is, to our knowledge, the first to analyze whether and by how much depression and anxiety symptoms are independently associated to occupational role transitions. Even though we had a relatively small sample, i.e. 30 participants, and hence a suboptimal statistical power, we found a few significant, independent associations between depression symptoms and anxiety as a trait with occupational role transitions, adjusted for relevant cofounders.

For instance, we found that depression sympto-matology was independently and significantly associated to occupational roles loss (a 4%higher likelihood of occupational role loss for each additional point in the BDI), and that anxiety as a trait was independently associated to occupational roles gain (a 7%reduced likelihood of occupational role gain by each additional point in the STAIT, notably the trait subscale). Hence, as relatively expected, we found a negative effect of depression or anxiety symptoms on occupational role transitions, either for higher occupational role losses or lower occupational role gains. Such an negative effect on occupational roles is an additional, occupational reason to assess and address mental health issues in people experiencing a SCI [10 , 14].

Apparently as a counter-intuitive finding, we ob-served that depression symptoms not only were associated to key occupational roles losses but also to some occupational role gains. This can be explained by the acquisition of roles such as those of a “caregiver”, “home maintainer”, or “volunteer” likely as a result of losses in other occupational roles, such as those of “worker” or “hobbyist/amateur”. Hence, for this sample of people experiencing a SCI, and in the context of depressive symptomology, this finding may reflect negatively appraised changes in occupational patterns from the pre- to the post-injury time period. For example, it is possible that newly acquired occupational roles were gained not because they were much wanted or much valued (e.g. not arising from one’s volition), but possibly perceived as a less significant alternative to the lost roles. Working on assessing and developing the volition for and meaningfulness of new occupational roles can be important tasks for rehabilitation providers (e.g. occupational therapists) working with people likely experiencing occupational roles transitions after SCI, and especially those with or at risking of experiencing depression symptoms.

The relationship between mental health issues and occupational roles transitions in the context of long-term physical impairments is likely an intricate one. Not only can psychological morbidity affect occupational role transitions but also the other way around. The loss or replacement of previous and valued occupational roles after a disrupting life event, such as the experience of SCI, can arguably generate stress, anxiety and depressive symptoms. By the same token, psychological morbidity likely interferes with occupational role transitions and their personal significance. While we studied the effect mental health issues on occupational roles transitions, we did not studied the effect of occupational role transitions as a mental health determinant, i.e. the issue of occupational spin-off issue [46, 47]. In clinical practice, mental health and occupational issues in people experiencing SCI would need to be considered in tandem. For example, clinicians may consider, for each patient, whether any implied occupational role transitions is likely to affect mental health states, the same ways that any psychological morbidity may negatively affect the success of occupational role transitions, when not both at the same time.

In research, longitudinal studies and those using structural equation models, ideally applied over larger samples, would be important to disentangle the complex relationships between mental health issues and occupational roles transitions. In this relatively small sample, anxiety as a trait was not significantly, independently associated to occupational role loss, but for a marginal value (p = 0.051). In study with a larger sample, and hence a larger statistical power, this association would possibly be different. All accounted, our preliminary findings seem to warrant further investigation, with greater means.

5 Study limitations

The study has the following limitations. First, it uses a cross-sectional design and has a small number of participants (n = 30). That impedes causal inference on one hand, and higher statistical power on the other. That could work, for example, toward detecting more significant relationships between the independent and dependent variables, or even for performing sub-group analyses. Second, the sample was mostly composed of males and of people living in a peripherical, sub-urban area with lower household income for the Brazil (a country with known socio-economic inequalities). While the data can be representative of the local context (i.e. nearly all of those attending the service became participants), it might not be fully representative of every environment, even in Brazil. Indeed, characteristics of the built and social environment such as that of the neighborhoods can interfere with community-based SCI outcomes [48, 49]. Also, differences exist across nations on how occupational roles are both perceived and exemplified [50]. So, any implication for other contexts must be taken with caution.

Third, we did not use a formal scale, with a gradient, to measure the performance in the activities of daily living as a confounding factor, but rather a single classification item of dependency versus independency following semi-structured questions.

Fourth, we do not match occupational role transitions with the value such role has for each participant in particular (i.e. individual-level analysis), the latter potentially with greater relevancy for an individualized, person-centered clinical practice [51].

Fifth, we apply the classical version of the Roles Checklist, the one translated and validated for the Brazilian population, not the more recently modified version [52] or even the Revised Role Checklist published in 2019 [53].

Sixth, we do not consider other types of psychological morbidity with relevance to traumatic SCI, such as post-traumatic stress disorders [5]. Similarly, we did not assessed issues around sexuality, which could arguably affect the mental state in the studied population.

Seventh, we focused on symptomatology (i.e. negative psychological variables) and their association to occupational changes, and for instance did not account for effect of variables such as personal resources, social skills, social support or relationships, as well as positive psychology variables (e.g. post-traumatic growth). Any of these, in theory, can interfere with (e.g. buffer, revert) any negative impact of psychological morbidity in the occupational roles transitions.

6 Conclusion

This cross-sectional study, using a relatively small sample (n = 30), found that depression symptoms and anxiety as a state were significant, independently associated to occupational role transitions. Larger studies, in other contexts, with a longitudinal timespan, with additional variables (e.g. from the positive psychology), and with other analytical approaches (e.g. using structural equation models) could further elucidate on the likely intricate relationships between psychological morbidity and occupational role transitions.

Conflict of interest

None to report.

References

Craig

, Perry

, Guest

, Tran

, MClinPsych

, Middleton

. Prospective study of the occurrence of psychological disorders and comorbidities after spinal cord injury. Archives of Physical Medicine and Rehabilitation. 2015;96(8):1426–34.

Craig

, Guest

, Tran

, Middleton

. Depressive mood in adults with spinal cord injury as they transition from an inpatient to a community setting: secondary analyses from a clinical trial. Spinal Cord. 2017;55(10):926–34.

Post

, van Leeuwen

. Psychosocial issues in spinal cord injury: a review. Spinal Cord. 2012;50(5):382–9.

MacDonald

, Nielson

, Cameron

. Depression and activity patterns of spinal cord injured persons living in the community. Archives of Physical Medicine and Rehabilitation. 1987;68(6):339–43.

Wang

, Xie

, Zhao

. Psychological morbidities and positive psychological outcomes in people with traumatic spinal cord injury in Mainland China. Spinal Cord. 2018;56(7):704–11.

Lim

, Shiue

, Ho

, Yu

, Kao

, Wang

, Kuo

. Anxiety and Depression in Patients with Traumatic Spinal Cord Injury: A Nationwide Population-Based Cohort Study.PloS One. 2017;12(1):e0169623.

, Dorstyn

. Anxiety prevalence following spinal cord injury: a meta-analysis. Spinal Cord. 2016;54(8):570–8.

Hartoonian

, Hoffman

, Kalpakjian

, Taylor

, Krause

, Bombardier

. Evaluating a spinal cord injury-specific model of depression and quality of life. Archives of Physical Medicine and Rehabilitation. 2014;95(3):455–65.

Tran

, Dorstyn

, Burke

. Psychosocial aspects of spinal cord injury pain: a meta-analysis. Spinal Cord. 2016;54(9):640–8.

10.

Bombardier

, Fann

, Tate

, Richards

, Wilson

, Warren

, Temkin

, Heinemann

. An exploration of modifiable risk factors for depression after spinal cord injury: which factors should we target? Archives of Physical Medicine and Rehabilitation. 2012;93(5):775–81.

11.

van Diemen

, van Leeuwen

, van Nes

, Geertzen

, Post

. Body Image in Patients With Spinal Cord Injury During Inpatient Rehabilitation. Archives of Physical Medicine and Rehabilitation. 2017;98(6):1126–31.

12.

Ghazwin

, Tavakoli

SAH

, Latifi

, Saberi

, Derakhshanrad

, Yekaninejad

, Sadeghi

, Razavi

SHE

, Javidan

, Ghodsi

. Psychological defense mechanisms among individuals with SCI with adjustment disorder. The Journal of Spinal Cord Medicine. 2017;40(5):538–47.

13.

Zurcher

, Tough

, Fekete

. Mental health in in-dividuals with spinal cord injury: The role of socioeconomic conditions and social relationships. PloS One. 2019;14(2):e0206069.

14.

LaVela

, Etingen

, Miskevics

, Heinemann

. What determines low satisfaction with life in individuals with spinal cord injury? The Journal of Spinal Cord Medicine. 2019;42(2):236–44.

15.

Kraft

, Dorstyn

. Psychosocial correlates of depression following spinal injury: A systematic review. The Journal of Spinal Cord Medicine. 2015;38(5):571–83.

16.

Pretz

, Kozlowski

, Chen

, Charlifue

, Heinemann

. Trajectories of Life Satisfaction After Spinal Cord Injury. Archives of Physical Medicine and Rehabilitation. 2016;97(10):1706–1713.e1701.

17.

Kennedy

, Kilvert

, Hasson

. A 21-year longitudinal analysis of impact, coping, and appraisals following spinal cord injury. Rehabilitation Psychology. 2016;61(1):92–101.

18.

Muller

, Peter

, Cieza

, Post

, Leeuwen

CMV

, Werner

, Geyh

. Social skills: a resource for more social support, lower depression levels, higher quality of life, and participation in individuals with spinal cord injury? Archives of Physical Medicine and Rehabilitation. 2015;96(3):447–55.

19.

Peter

, Muller

, Post

, van Leeuwen

, Werner

, Geyh

. Psychological resources, appraisals, and coping and their relationship to participation in spinal cord injury: a path analysis. Archives of Physical Medicine and Rehabilitation. 2014;95(9):1662–71.

20.

Geyh

, Kunz

, Muller

, Peter

. Describing functioning and health after spinal cord injury in the light of psychological-personal factors. Journal of Rehabilitation Medicine. 2016;48(2):219–34.

21.

Marti

, Escorpizo

, Schwegler

, Staubli

, Trezzini

. Employment pathways of individuals with spinal cord injury living in Switzerland: A qualitative study. Work (Reading, Mass). 2017;58(2):99–110.

22.

Hwang

, Zebracki

, Vogel

. Occupational characteristics of adults with pediatric-onset spinal cord injury. Topics in Spinal Cord Injury Rehabilitation. 2015;21(1):10–19.

23.

Hilton

, Unsworth

, Stuckey

, Murphy

. The experience of seeking, gaining and maintaining employment after traumatic spinal cord injury and the vocational pathways involved. Work (Reading, Mass). 2018;59(1):67–84.

24.

Holmlund

, Guidetti

, Eriksson

, Asaba

. Return to work in the context of everyday life 7-11 years after spinal cord injury - a follow-up study. Disability and Rehabilitation. 2018;40(24):2875–83.

25.

Nutzi

, Trezzini

, Staubli

, Ronca

, Schwegler

. An interdisciplinary approach to job matching: developing an occupation-specific job matching tool for reintegrating persons with spinal cord injury into the labor market. Disability and Rehabilitation. 2019:1-15.

26.

Oakley

, Kielhofner

, Barris

, Reichler

. The Role Checklist: Development and Empirical Assessment of Reliability. The Occupational Therapy Journal of Research. 1986;6(3):157–70.

27.

Cordeiro

, Camelier

, Oakley

, Jardim

. Cross-cultural reproducibility of the Brazilian Portuguese version of the Role Checklist for persons with chronic obstructive pulmonary disease. The American Journal of Occupational Therapy: Official Publication of The American Occupational Therapy Association. 2007;61(1):33–40.

28.

Sakakibara

, Miller

, Orenczuk

, Wolfe

. A systematic review of depression and anxiety measures used with individuals with spinal cord injury. Spinal Cord. 2009;47(12):841–51.

29.

Beck

, Ward

, Mendelson

, Mock

, Erbaugh

. An inventory for measuring depression. Archives of General Psychiatry. 1961;4:561–71.

30.

Beck

, Steer

, Carbin

. Psychometric properties of the Beck Depression Inventory: Twenty-five years of evaluation. Clinical Psychology Review. 1988;8(1):77–100.

31.

Gorenstein

, Andrade

LHSG

. [Beck’s Depression Inventory: psychometric properties of the Portuguese version] in Portuguese. Revista de Psiquiatria Clínica. 1998;25(5):245–50.

32.

Spielberger

, Gorsuch

, Lushene

. Manual for the State-Trait Anxiety Inventory. Palo Alto, CA.: Consulting Psychologists Press; 1970.

33.

Gorenstein

, Andrade

LHSG

. Validation of a Portuguese version of the Beck Depression Inventory and the State-Trait Anxiety Inventory in Brazilian subjects. Braz J Med Biol Res. 1996;29:453–7.

34.

Keedwell

, Snaith

. What do anxiety scales measure? Acta psychiatrica Scandinavica. 1996;93(3):177–80.

35.

Coleman

, Harper

, Perrin

, Oliveira

, Perdomo

, Arango

, Arango-Lasprilla

. The relationship between physical and mental health variables in individuals with spinal cord injury from Latin America. PM & R: The Journal of Injury, Function, and Rehabilitation. 2015;7(1):9–16.

36.

Coleman

, Harper

, Perrin

, Landa

, Oliveira

, Perdomo

, Arango

, Arango-Lasprilla

. Mind and body: Mental health and health related quality of life in SCI caregivers from Neiva, Colombia. NeuroRehabilitation. 2015;36(2):223–32.

37.

Harper

, Coleman

, Perrin

, Oliveira

, Perdomo

, Arango

, Lasprilla

JCA

. Comparison of mental health between individuals with spinal cord injury and able-bodied controls in Neiva, Colombia. Journal of Rehabilitation Research and Development. 2014;51(1):127–36.

38.

Andrade

, Gorenstein

, Vieira Filho

, Tung

, Artes

. Psychometric properties of the Portuguese version of the State-Trait Anxiety Inventory applied to college students: factor analysis and relation to the Beck Depression Inventory. Braz J Med Biol Res. 2001;34(3):367–74.

39.

Craig

, Tran

, Lovas

, Middleton

. Spinal cord injury and its association with negative psychological states. Int J Psychosoc Rehabil. 2008;12:115–21.

40.

Portuguese BAoRCi. [Brazil Criterion 2016] in Portuguese. Available at: www.abep.org/criterio-brasil, Accessed May 19 2018.

41.

Arango-Lasprilla

, Ketchum

, Starkweather

, Nicholls

, Wilk

. Factors predicting depression among persons with spinal cord injury 1 to 5 years post injury. NeuroRehabilitation. 2011;29(1):9–21.

42.

Al-Owesie

, Moussa

, Robert

. Anxiety and depression among traumatic spinal cord injured patients. Neurosciences (Riyadh, Saudi Arabia). 2012;17(2):145–50.

43.

Tsagris

, Stewart

. A Dirichlet Regression Model for Compositional Data with Zeros. Lobachevskii Journal of Mathematics. 2018;39(3):398–412.

44.

Meier

. DirichletReg: Dirichlet Regression for Compositional Data in R. Research Report Series / Department of Statistics and Mathematics, 125. Vienna: WU Vienna University of Economics and Business; 2014.

45.

Ferrari

, Cribari-Neto

. Beta Regression for Modelling Rates and Proportions. Journal of Applied Statistics. 2004;31(7):799–815.

46.

Ekelman

, Allison

, Duvnjak

, DiMarino

, Jodzio

, Iannarelli

. A Wellness Program for Men With Spinal Cord Injury: Participation and Meaning. OTJR: Occupation, Participation and Health. 2017;37(1):30–9.

47.

Rebeiro

, Cook

. Opportunity, not prescription: an exploratory study of the experience of occupational engagement. Canadian Journal of Occupational Therapy Revue Canadienne d’ergotherapie. 1999;66(4):176–87.

48.

Botticello

, Tulsky

, Heinemann

, Charlifue

, Kalpakjian

, Slavin

, Byrne

, Rohrbach

. Contextualizing disability: a cross-sectional analysis of the association between the built environment and functioning among people living with spinal cord injury in the United States. Spinal Cord. 2019;57(2):100–9.

49.

Botticello

, Boninger

, Charlifue

, Chen

, Fyffe

, Heinemann

, Hoffman

, Jette

, Kalpakjian

, Rohrbach

. To What Extent Do Neighborhood Differences Mediate Racial Disparities in Participation After Spinal Cord Injury? Archives of Physical Medicine and Rehabilitation. 2016;97(10):1735–44.

50.

Bonsaksen

, Meidert

, Schuman

, Kvarsnes

, Haglund

, Prior

, Forsyth

, Yamada

, Scott

. Does the Role Checklist Measure Occupational Participation? The Open Journal of Occupational Therapy. 2015;3(3) https://doi.org/10.15453/12168-16408.11175.

51.

Jesus

, Bright

, Kayes

, Cott

. Person-centred rehabilitation: what exactly does it mean? Protocol for a scoping review with thematic analysis towards framing the concept and practice of person-centred rehabilitation. BMJ Open. 2016;6(7):e011959.

52.

Avrech Bar

, Rubin

, Gavrieal-Tyjchman

, Jarus

. The validity and reliability of the modified version of the Role Checklist (M-RCL). Scandinavian Journal of Occupational Therapy. 2013;20(6):454–62.

53.

Scott

, McKinney

, Perron

, Ruff

, Smiley

. The Revised Role Checklist: Improved Utility, Feasibility, and Reliability. OTJR: Occupation, Participation and Health. 2019;39(1):56–63.