«Yona Lunsky (University of Toronto) and Paul Bramston (University of Southern Queensland) Approximate word count: 4468 words Keywords: Intellectual ...»
A preliminary study of perceived stress in adults with intellectual
disabilities according to self-report and informant ratings1
Yona Lunsky (University of Toronto) and
Paul Bramston (University of Southern Queensland)
Approximate word count: 4468 words
Keywords: Intellectual disability, stress
Address for correspondence: Yona Lunsky, Ph.D., Centre for Addiction and Mental
Health, 1001 Queen Street West, Toronto, Ontario CANADA, M6J 1H4, Email:
firstname.lastname@example.org 1 This is the authors‟ final corrected version of Lunsky, Yona and Bramston, Paul (2006) A preliminary study of perceived stress in adults with intellectual disabilities according to self-report and informant ratings. Journal of Intellectual and Developmental Disability, 31 (1). pp. 20-27. ISSN 1366- 8250 DOI: 10.1080/13668250500488660 Lifestress Informant 2 Abstract Background. Stress is a major risk factor for mental health problems in individuals with intellectual disabilities, but few studies on stress have been conducted that take both the perspective of the person with the disability and the caregiver into account. The present study evaluated an informant version of the Lifestress Inventory, and compared it to the self-report version.. Method. Seventy pairs of individuals with intellectual disability and their caregivers completed the Lifestress Inventory, the Inventory of Negative Social Interactions and the Birleson Depression Scale. Results. Informant and self-report ratings on the Lifestress Inventory were internally reliable, showed modest agreement with each other and correlated with the negative Social Interactions and Depression measures. The most troublesome stressors reported by informants and self-reports differed, however, and families tended to agree more with self-reports than staff informants. Conclusions.
The informant version of the Lifestress Inventory is a suitable parallel instrument but not a replacement for self reports.
2 Lifestress Informant 3 Introduction Research suggests people with intellectual disabilities (ID) experience at least as much or more stress than the general population (Hatton & Emerson, 2004), yet they have fewer resources (internal and external) to manage their stress (Lunsky & Neely, 2000). Potential stressors arise from the greater risk they face of being marginalised within our society (Annison, 2000), which manifests itself in prejudice and discrimination. So, not only are they faced with the personal challenges and demands of their disability, but also with the restricting environmental factors that others without a disability do not have to face In order to conceptualise the impact of psychological and situational pressures on an individual with ID, it is helpful to briefly review the Lazarus model of stress (Lazarus & Folkman, 1984). This model holds that if the individual perceives an event as being overly challenging and beyond his/her ability to cope, then he or she is likely to experience stress. In cognitively appraising an event as manageable or unmanageable, the individual is thought to call upon: 1) previous experiences; 2) environmental supports; and 3) personal beliefs. The research on those with intellectual disabilities suggests that: 1) People with ID often have previous coping experiences that are less than positive (Hastings, Hatton, Taylor, & Maddison, 2004);
2) Those with ID have limited environmental supports (Lunsky & Benson, 2001;
Reiss & Benson, 1984) and; 3) Those with ID have low beliefs in their own abilities (Jahoda, Cattermole, & Markova, 1988). The combination of these three factors when appraising a situation would greatly increase the likelihood of stress occurring.
Stress has long been recognised as a major risk factor for mental and physical health problems, thus it is surprising that so little research on stress has been carried out with people with ID, who often have multiple medical and psychiatric issues. One of the reasons so little research has been published is the absence of reliable and valid measures. Few existing measures of stress for the ID population consider both major life stressors as well as minor events or daily hassles (Bramston et al., 1999). In addition, existing measures of stress employ either exclusively informant views or exclusively self-reports. Below we review the limited research on stress according to informant and self report perspectives.
Only one measure of stress faced by individuals with ID has been developed specifically for caregivers/informants: Groden and colleagues (2001) developed an informant measure of stress for individuals with autism, the Stress Survey Schedule.
The scale is completed by the caregiver and considers anticipation/uncertainty, changes & threats, unpleasant events, pleasant events, sensory/personal contact, food related activity, social/environmental activities and ritual related stress -- all factors particularly relevant to individuals with autistic spectrum disorders. Autism has significant overlap with intellectual disability, with 75% of individuals with autism having ID and up to 30% of people with ID having autism (see Bradley & Lofchy for review, 2005). As a result, some of the items on the Groden et al. scale might have relevance to the broader population with ID, but the measure has yet to be researched amongst this group.
3 Lifestress Informant 4
The impact of significant life events (major stressors) and their relationship to mental health crises in those with ID has been well-documented (e.g., Stack et al., 1987; Ghazziudin, 1988; Hatton & Emerson, 2004; Nadarajah, Roy, Harris, & Corbett, 1995). One recent study (Hastings, Hatton, Taylor, & Maddison, 2004) summarized information on 17 major life events for 1155 individuals with ID according to informants, using the PAS-ADD checklist (Moss et al., 1998) and reported a relationship between significant life events and psychopathology, particularly in the affective domain. Another recent study (Owen et al., 2004) on 93 individuals with ID in an institutional setting used the Life Events List (LEL), a 20item informant completed list of negative life events thought to be particularly relevant to individuals with ID in a long term residential setting. This study reported that the most common stressful life events experienced were staffing and residence changes, conflict, family bereavements and relationships and illness or injury, and individuals exposed to such life events were rated as displaying more aggression and at higher risk for affective disorders. Although both of the scales (PAS-ADD checklist and LEL) employed in these two studies provide information on overall frequency of stressful life events, they do not assess the impact of such events and also do not address forms of stress other than major life events (e.g., daily hassles).
Only two measures relating to stress faced by individuals with ID have been developed specifically as a self-report. A self-report measure of anxiety, the Glasgow Anxiety Scale (Mindham & Espie, 2003), was recently published. It appears to successfully tap self-reported anxiety in people with an intellectual disability using a 35-item inventory with a 3-point Likert response scale (never, sometimes, always).
While overlapping with the construct of stress, the content of this scale incorporates more anxiety-specific symptoms (such as panic attacks) and is primarily a measure for individuals with mild intellectual disability, with no informant version. The other selfreport scale was developed by Bramston and his colleagues, who conducted a series of focus groups amongst Australians with ID and from this developed a list of life events that typically bother, upset and frustrate individuals with ID. These were then refined into a self-report stress scale (the Lifestress Inventory) with accompanying normative self-report data from 459 people with mild or moderate intellectual disability (Bramston, Fogarty & Cummins, 1999). Subsequent use of the inventory confirmed that the subjective experience of the person with ID can be reliably and validly obtained, that it is predictive of depression and that this perspective is important to solicit (Esbensen, 2004; Hartley & MacLean., in press).
Previous research regarding perceived stress or negative life events in the ID population has focused on either exclusively on the perspective of caregivers or exclusively on individuals with ID. No studies thus far have obtained both perspectives. Whether the informant perspective can add further information beyond what we can learn from self-report has yet to be studied. Thus, the purpose of this study was to determine whether an informant version of the Lifestress Inventory could be a valuable parallel assessment tool to the existing self-report version of the Lifestress Inventory developed by Bramston and colleagues. Specifically, we sought to measure how informants (staff or family) perceive stress in people with intellectual
disability, how well informant perceptions correspond to self-reported stress, and how well an informant measure of stress is correlated with other informant measures of problematic life events and distress/depressive symptoms in the ID population.
Method Participants Ninety-nine adults with ID in Southern Ontario, Canada completed a study looking into “things in life that bother you”. For this study, each participant was asked to nominate someone “who knows you the best who might be willing to answer some questions about the same kinds of things we spoke about.” Seventy participants nominated an individual who agreed to participate and returned questionnaires (see Measures below). Findings here are based on these 70 self-report/informant pairs.
The average age of the 70 participants with ID was 38.7 years (range 20yrs) and 51% were male. The majority of individuals were between the ages of 36 and 50 (18 women and 13 men), 28 (18 men and 10 women) were 35yrs or below, and 11 individuals (5 men and 6 women) were older than 50. Participants came from a range of residential settings: 23% were living with family, 33% lived independently, 28% lived in supported independent living, and 15% lived in group homes. Twenty percent of the sample was working in community settings, 28% were in sheltered workshops, 33% attended educational or recreational day programs, and 19% had no daytime structured activities. On average, informants knew the participant for 14.6 years with 59% of the informants describing themselves as staff members (16 residential staff, 16 case managers or counsellors and 9 day program staff) and 41% identifying themselves as family (26 individuals) or unpaid caregiver/friend (3 individuals).
A standardised screening measure of intelligence (Wechsler Abbreviated Scale of Intelligence) was individually administered to each participant by the researcher (registered psychologist) so that any relationship between cognitive ability and performance on the dependent measures could be accounted for. Of the 65 individuals who completed the matrices and vocabulary WASI subscales, six individuals (1 woman and 5 men) had IQ screen scores between 76 and 86; 50 individuals had standardized scores between 56 and 75 (suggesting mild ID), and 9 individuals received a standardized score of 55 or below (the lowest score obtainable on the test, suggesting moderate ID). As WASI scores were not found to be related to scores on the dependent measures, and all individuals interviewed were receiving ID services, the six people who may have been functioning in the borderline range of intelligence were included in the study. (Correlations did not change significantly when these six pairs were excluded.) Measures Informants were asked to complete the Lifestress Inventory for informants (Lifestress-Inf), which is a new inventory compiled specifically for this study and designed to enable informants to quantify the number of potentially stressful life events being experienced by people with ID (see Appendix A). The Lifestress-Inf was
5 Lifestress Informant 6
derived from the Lifestress Inventory (Bramston, Fogarty & Cummins, 1999) which is a self-report stress scale comprised of a list of 30 life events which people with ID reported (in focus groups) to most often bother or upset them (for details on scale construction, see Bramston, Bostock and Tehan, 1993). The original Lifestress Inventory (self-report) has been shown to be reasonably reliable, valid and factorially stable (Bramston, Fogarty, & Cummins, 1997) and clinically useful in understanding the pressures experienced by young adults moving into community accommodation (Bramston & Cummins, 1998). For comparative purposes with the new Lifestress-Inf, the Lifestress Inventory (self-report) was also administered in this study.
The Lifestress-Inf slightly adapts the wording of items from the original Lifestress Inventory (self-report) to suit the perspective of a rater. For example, the self-report item “Do you get to choose things that are important to you?” was changed to “Does he/she get to choose things that are important to him/her?”. Informants were asked to score each item on a 0-2 scale Likert scale with 0 = „No, not at all‟; 1 = „Sometimes or somewhat‟; and 2 = „Yes, most definitely‟. This three-point scale was also used for the self-report Lifestress Inventory in this study in order to facilitate comparisons between the two scales. The original Lifestress Inventory (self-report) uses a 4-point Likert Scale to measure stress impact, but this was considered unsuitable for the current study because informants cannot reasonably be expected to so finely quantify the personal impact on another of experiencing various life events.
It was anticipated that this change would not alter the number of reported life events (frequency score) but would restrict the range of Impact totals. To minimise response sets, 17 of the 30 items on both scales are worded positively and require reversing prior to analysis so that higher scores reflect more stress.
On both the self-report and the informant scales a frequency score represents the total number of life events that have happened recently in the individual‟s life (Frequency score: maximum of 30 events). The impact of those events on the individual is estimated by adding the ratings (0, 1, 2) on the endorsed items (Impact score: maximum score of 60). These scoring procedures are in accordance with the processes set out in the original Lifestress Inventory (Bramston, Cummins, & Fogarty, 1999).