OVERVEIW OF THESIS
This thesis explored the subjective quality of life of people with MS, their partners,
and a control comparison group. The unique combination of physical symptoms and
the unpredictable, fluctuating disease course of MS provided an interesting context
within which to explore subjective quality of life. Also, given the limited availability
of treatments for MS, research with such a focus was needed to aid people with MS
to live full and satisfying lives within the unique constraints of their condition. An
overview of the thesis is now presented. Implications for future research and
interventions are discussed, as are the strengths and limitations of the thesis. Finally
conclusions are drawn.
Background to the Thesis
A model proposed to explain the process whereby psychological factors maintain
subjective quality of life in an adaptive range was generated and tested by the thesis.
This model was based on the presupposition that life circumstances exert only a
limited influence on subjective quality of life (Argyle, in press). This is evidenced by
the consistent of life satisfaction evaluations across a range of life situations that have
resulted in the estimation of a life satisfaction ‘gold standard’ at 75+/-2.5 per cent SM
In the first five chapters of the thesis literature was reviewed that linked
psychological variables to subjective quality of life. Cummins and Cahill (in press)
have proposed a model that links personality, positive cognitive biases of self-esteem,
optimism and primary control, and subjective quality of life. The thesis additionally
integrated perceived social support and secondary control into this model. It was
hypothesised that these two variables contribute to subjective quality of life in a
consistent and predictable way. The inclusion of perceived social support into the
model was based on previous research that has consistently demonstrated a positive
association between perceived social support and subjective quality of life (Abbey et
al., 1985; Holahan et al., 1996; Revenson et al, 1991; Schaefer, Coyne, & Lazarus,
1981). Furthermore, the effect of perceived social support was hypothesised to be
subject to a threshold effect such that people under stress feel the absence of social
support more acutely (Cohen & Wills, 1985). Secondary control was hypothesised to
mediate between positive cognitive bias and subjective quality of life. The inclusion
of a mediating role of secondary control in the model was based on Wiesz et al.’s
(1984) definition of secondary control as accepting or adjusting to one’s situation. It
was proposed that life situations can challenge positive cognitive biases and that
secondary control acts to accommodate life situations and maintain subjective quality
of life. It was hypothesised that the model is valid for all types of people, including
those with MS, but that the variables of the model interact differently between
different groups of people. This model was illustrated in Figure 5.2 in Chapter 5.
Summary of Studies
The model was explored through two linked studies presented in chapter 7 and 8. In
the Study One a technique called ‘photo voice’ was utilised to obtain qualitative
information to inform the materials of the second study. This study used photographs
taken by the participants to explore, in an interview, the factors that effected their
quality of life and the ways in which they coped with everyday difficulties. People
took photographs of both positive and negative influences on their quality of life
including family members, pets, homes, wheelchairs and medications. They also
reported a wide range of ways they coped with everyday difficulties including two
novel approaches: ‘ventilation’, which included yelling and crying, and ‘give up’.
The qualitative information obtained from Study One valuably informed the materials
for Study Two. Specifically, two new items were added to the Primary and
Secondary Control Scale, a new domain was added to the Comprehensive Quality of
Life Scale and potentially biased items were identified in the NEO-Five Factor
Inventory and the Rosenberg Self-Esteem Inventory. The Perceived Social Support
Scale was also modified to include varied sources of support, and a subscale was
designed to assess support from pets.
The second study employed a quantitative methodology comprising a self-report
questionnaire designed to measure the variables of the model. This study involved a
larger group of people with MS, their partners and a control comparison group. The
findings of Study Two supported the proposed model for each group. The variables of
the model accounted for a significant portion of the subjective quality of life for each
group (45 % for controls, 59% for people with MS and 58% for partners of people
with MS). Personality, specifically neuroticism, predicted subjective quality of life.
Positive cognitive bias and perceived social support accounted for a significant
portion of subjective quality of life after personality was controlled. Also, secondary
control mediated between positive cognitive bias and subjective quality of life for
each of the three groups.
Consistent with predictions of the thesis, the variables of the model interacted
differently between the three groups and people with MS reported a lower subjective
quality of life than the other two groups. The subjective quality of life of the three
groups was similarly negatively effected by neuroticism. The three positive cognitive
biases differently predicted the subjective quality of life of the three groups. Self-
esteem was highly positively associated with the subjective quality of life of all three
groups. However, people with MS reported a low self-esteem. Additionally,
optimism positively predicted the subjective quality of life of controls, and primary
control negatively predicted the subjective quality of life of people with MS and
positively predicted for partners.
As hypothesised, perceived social support consistently positively predicted the
subjective quality of life of the three groups and was subject to a threshold effect.
People with MS felt the absence of support more acutely than the other two groups.
The value of support from pets for people with MS suggested by Study One was also
The adaptive outcome of primary control was in accordance with the Discrimination
Model of control that claims that the use of primary control is adaptive when the
situation is controllable (Folkman 1984; Thompson et al., 1998). No support was
found for either the Primacy/Back-Up model that claims that primary control is more
adaptive than secondary control regardless of the controllability of the situation
(Heckhausen & Schulz, 1995), or Heep’s (2000) assertion that balanced levels of high
primary and secondary control are positively associated with well-being.
The effect of secondary control on subjective quality of life varied between the three
groups. The mediating effect of secondary control was most for people with MS, less
for partners, and least of all for controls. People with MS utilised a high level of
avoidant secondary control that was negatively associated with subjective quality of
life. Thus, people with MS were using secondary control in maladaptive ways.
The adaptive outcome of secondary control evident in the study did not support any
of the three models of control. ‘Ventilation’ commonly negatively predicted
subjective quality of life. Thus, the utility of the inclusion of this novel form of
secondary control was confirmed. However, the effect of the other four types varied
between the groups. Most notably, the five types of secondary control were only
negatively associated with the subjective quality of life of people with MS, had a
mixed effect for partners, and a negligible impact for controls. Thus, secondary
control efforts had the most impact on well-being in the most stressful life
The analysis of the adaptive outcome of secondary control highlighted the value of
analysing the five type secondary control separately. This suggested limitations to
the current models of control (i.e. The Discrimination Model (Folkman, 1984;
Thompson et al., 1998), the Primacy/Back-Up Model (Heckhausen & Schulz, 1995))
that compare only average primary and secondary control and therefore do not
accurately represent the varied functions and outcomes of secondary control.
Implications of the Findings
The implications of the findings of the thesis for interventions aimed at improving
subjective quality of life and for future research have been discussed previously and
are now summarised.
Implications for Interventions
The thesis aimed to inform interventions aimed at improving the subjective quality of
life of people with MS and their partners. The following recommendations were
made based on the findings of the thesis:
Extraversion and neuroticism are traits with a strong genetic link (Tellegen et al.,
1998), as such they are not likely to be appropriate targets for intervention. However,
other variables and processes identified by the model and confirmed by this thesis are
potentially malleable and could be effective targets for interventions
The findings suggested that improvement to self-esteem will likely benefit subjective
quality of life for people with MS, partners and controls. This is especially relevant
to people with MS who reported lowered self-esteem. Previous research hypothesises
that positive cognitive biases are maintained by failure experiences (Cummins &
Nistico, in press). Thus, interventions could focus on reducing the experience of
failure, or conversely increasing successful experiences of people with MS.
Additionally, this research found that secondary control mediates between positive
cognitive bias and subjective quality of life. Secondary control was proposed to
maintain the lower boundary of positive cognitive bias. However, as the adaptive
outcome of secondary control could not be determined by this thesis, no interventions
regarding secondary control for people with MS were made. However, situation bias
was positively associated with the subjective quality of life of partners. Thus,
partners would likely benefit from assistance to see the positive in their situation.
Interventions may include counselling and the promotion of the positive aspects of
MS. Further research needs to explore potential positive aspects of MS, these may
include the opportunity for a renewed perspective on life or the sense of achievement
gained through caring for a partner.
Primary control may also be an effective target for interventions. Partners would
benefit from increased feelings of personal competence. For example, interventions
could provide partners with access to resources, such as life skills training or
educational materials, so that they feel better equipped to deal with potential
stressors. If these resources already exist, then interventions could focus on raising
awareness of them to partners. Because, as Thompson (1981) asserts, control does
not need to be exercised for it to be effective and control does not need to be real, just
perceived, for it to influence the aversiveness of a stressful encounter (Thompson,
While primary control positively effected the quality of life of partners, it negatively
effected that of people with MS. Therefore, people with MS could be assisted to
decrease levels of primary control. For example, it may be more adaptive for people
with MS to relinquish personal control to other people such as partners or health-care
professionals. However, previous researchers (Rothbaum et al., 1982) suggest that
this may not be an easy task as control is so valued that the quest for it is rarely
abandoned. Individuals are instead likely to shift from one method of control to
another. Therefore, interventions may be more effective if they assisted people with
MS to re-focus their primary control on controllable situations. Interventions could
provide people with MS opportunities for primary control. For example, people with
MS could be taught new skills in order to develop feelings of personal mastery.
Further research needs to explore this possibility.
Improvements in perceived social support will also likely benefit the subjective
quality of life of both people with MS, partners and controls. This is especially
important to people with MS who felt the absence of social support more acutely and
to partners who reported low perceived support in Study Two. As discussed, when
designing and implementing interventions both the positive and problematic aspects
of social support need to be considered (Holahan et al., 1996; Revenson et al, 1991).
Thus, interventions need to focus not only on increasing social contact, but also on
improving the quality of social contact. Thus, potential interventions include
organised social activities, support groups, and relationship and individual
counselling. Interventions could also focus on support received from pets.
Interventions may include promoting the value of pets as important sources of
support and assistance to people in the care of their pet, such as the provision of
volunteers to exercise pets, or respite care for pets during times of hospitalization.
The interventions recommended have so far focussed on individuals, however
interventions aimed at improving the subjective quality of life of people with MS and
their partners may also need to extend beyond the individual to the context
surrounding the individual. The capacity of the individual to change their subjective
quality of life may be limited by life circumstances. For example, the
maladaptiveness of primary control evident in this study may lie not in the individual,
but in a world that results in uncontrollable situations for people with disabilities to
such an extent that reduced subjective quality of life is inevitable. Thus,
interventions could focus on providing a more controllable world for people with MS
and other disabilities. For example, by lobbying local government to ensure that new
buildings and infrastructure are disability friendly.
Implications for Future Research
The findings of the study and consideration of the application of these findings to
interventions aimed at improving subjective quality of life have highlighted some
avenues for future research. These are now summarised.
Further research is needed to inform the proposed interventions recommended to
improve the subjective quality of life of people with MS and their partners. For
example, research is needed to explore the support needs of people with MS, and the
reasons why partners perceived less support than people with MS. Research needs
also to determine ways that partners can feel more personally competent and how
people with MS can develop more self-esteem. Research also needs to consider the
contextual issues regarding subjective quality of life to determine to what extent the
subjective quality of life of partners and people with MS is restrained by life
circumstances. For example, while beliefs of personal competence are beneficial to
partners, there may be very little capacity for development of this belief in partner’s
lives. Also, their ability to maintain satisfying social support networks may be
constrained by their caregiving responsibilities.
The inconsistency of some of the findings of the thesis with previous research also
warrants further investigation. In particular, further research needs to explore the
previously asserted stable nature of personality (Tellegen et al., 1998) especially in
relation to people with chronic illness. In the meanwhile, this thesis recommends that
any further research exploring personality and chronic illness undertake a process,
similar to that employed in this thesis, in which measures are assessed for suitability
to the population in question.
The utility of the inclusion of self-esteem in a model to explain subjective quality of
life was questioned by the findings of this thesis. Self-esteem was highly correlated
with the subjective quality of life of all three groups and served to weaken statistical
analysis. The inclusion of self-esteem in the model may be superfluous and needs to
be research further. The relationship between self-esteem and subjective quality of
life needs further investigation.
The magnitude of optimism in people with MS in the second study was unexpected.
The unpredictable and fluctuating disease course of MS would intuitively result in
lowered optimistic. However, Robinson and Ryff (1999) claim that conditions of
uncertainty and high motivation, likely present in MS, are ideal conditions for
exaggerated optimism. This thesis proposed that a combination of positive and
negative influences serve to maintain optimism at normal levels for people with MS.
Further research needs to explore the influences on the optimistic bias especially in
relation to chronic illness.
A new secondary control item ‘ventilation’ was identified in Study One and
confirmed in Study Two. It was proposed that this represents a new type of
secondary control. Future research needs to confirm the utility of this type of control
and to determine whether any similar responses to stressful life circumstances exist.
The findings of the thesis suggested that current models of control that focus on
average primary and secondary control (i.e. The Discrimination Model (Folkman,
1984; Thompson et al., 1998), the Primacy/Back-Up Model (Heckhausen & Schulz,
1995), and Heep’s (2000) assertion regarding relative levels of control) are too
simplistic and do not adequately describe the function or adaptive outcome of
secondary control. Therefore, research is needed to explore, and generate models
regarding, the adaptive outcome of secondary control. This could be achieved
through a meta-analysis of the many studies that have explored the function of
secondary control in different groups of people (e.g. Aikens et al., 1997; Chipperfield
et al., 1999; Pakenham, 1999; Warren et al., 1991).
Further research needs to explore the possibility that the adaptiveness of primary
control may be constrained by situational factors. This thesis did not explicitly
measure the controllability of life circumstances for each group. Rather it was
assumed, based on a review of literature regarding the psychosocial impact of MS
(e.g. Larsen, 1990; Pakenham, 1998, Schofield et al., 1997), that MS provides
frequent uncontrollable situations to people with MS and, to a lesser extent, partners.
The findings of the two studies were interpreted based on this assumption regarding
the uncontrollability of life for people with MS, partners and controls. Future
research would benefit from assessments of the frequency with which people
experience life stressors and uncontrollable situations. These assessments could then
be related to the functioning of primary and secondary control.
Strengths and Limitations of the Thesis
The thesis was primarily limited by the characteristics of the sample. The sample of
people with MS and partners was relatively small in both the first and second study
despite extensive recruitment efforts. Also, people with MS seemed to be
experiencing illness remission rather than exacerbation. A larger sample may have
enabled more in-depth comparisons between the different types of MS and
exploration of intra-couple processes. While recruitment difficulties are likely
inherent to research with chronically ill people that may find it difficult to take part in
research due to their illness, nevertheless future research would benefit from a larger
and more representative sample of people with MS. This may be obtained through
more active recruitment. For example, rather than relying on passive means of
recruitment, researchers could attend self-help groups and other places were people
with MS attend and minimise the effort required by participants to take part. The
complex methodology, especially of the first study, may have precluded the
involvement of some individuals.
The sample excluded people with medical conditions other than MS. While, it was
necessary to obtain a relatively homogenous sample for analyses, this nevertheless
limits the generalizability and interpretability of findings. The research is unable to
determine whether the effects of MS on subjective quality of life demonstrated in the
studies is due specifically to MS or is instead characteristic of all chronic illness. In
addition, the findings cannot be generalised to people with other chronic illness.
Another shortcoming of the thesis is that only people with partners were included.
Inclusion of people without partners would have enabled further exploration of the
role of intimate relationships in subjective quality of life.
Further research could therefore improve on this thesis by utilising a longitudinal
methodology so that changes over time and causal relationships between the variables
of the model could be explored. Also, a larger sample would permit the use of more
sophisticated data analysis techniques and the analysis of within group hypotheses
regarding the effect of illness characteristics on subjective quality of life and the
adaptiveness of couple coping.
The major strength of the thesis was the use of a model for the maintenance of
subjective quality of life that was valid for different groups of people. Utilisation of
such a generic model enabled comparisons between different groups of people. The
thesis benefited greatly from the use of three comparison groups. As discussed
previously, many of the findings of the study were only apparent in comparisons
between groups. Also, the control group provided a baseline comparison so that
deficits in the other two groups could be identified.
Another strength of the thesis was the combination of qualitative and quantitative
methodology employed. The quantitative second study was greatly benefited by the
qualitative first study. This information obtained in the qualitative study ensured that
the materials in the second study were valid for people with MS and their partners.
Validation of the questionnaire was important as the questionnaires were designed for
use with a ‘normal’ healthy population. The validated questionnaires provided some
findings that would not have been otherwise obtained. For example, the role of
perceived social support from pets would not have been detected.
Findings from the two studies support the utility of a generic model for the
maintenance of subjective quality of life. A model was generated based on previous
research and then tested with three comparison groups: people with MS, partners of
people with MS and controls. The model was supported for all three groups. This
model was based on Cummins and Cahill’s (in press) proposition that personality and
positive cognitive bias influence subjective quality of life, and on newly hypothesised
links between these variables and perceived social support and secondary control.
Specifically, perceived social support was found to exert a consistent and strong
influence on subjective quality of life, that was further subject to a threshold effect.
Secondary control was found to mediate between positive cognitive bias and
subjective quality of life.
The adaptive outcome of primary control was supportive of the Discrimination Model
of Control (Folkman, 1984; Thompson) that claims the primary control is adaptive
when situations are controllable. However, the adaptive outcome of secondary
control was not supportive of any of the models of control and varied across the three
groups and between the five types. The findings suggest that secondary control had
the most impact of subjective quality of life in stressful life conditions. This is
consistent with the stress-reactive role of secondary control proposed by the thesis.
The subjective quality of life of people with MS was less than both partners and
controls. Consistent with their lowered subjective quality of life they reported
lowered self-esteem, an exaggerated need for perceived social support, and the
maladaptive use of primary and secondary control. Partners did not report a lowered
subjective quality of life. They reported a lowered perceived social support that was
strongly related to their subjective quality of life however, this was likely
counterbalanced by the positive impact of primary control and situation bias
The findings of the thesis have implications for interventions aimed at improving the
subjective quality of life of people with MS and their partners. Specifically, it was
recommended that the self-esteem of people with MS be improved. Also, the quality
of social support, including that from pets, would be an effective target for
interventions for people with MS and their partners. Partners would benefit from
improved feelings of personal competence and assistance in seeing the positive in
The findings of the thesis also have implications for future research. Specifically
future research could explore the stability of personality, the utility of self-esteem in
models predicting subjective quality of life, the magnitude of optimism for people
with chronic illness and the adaptive outcome of the five types of secondary control.
Further research would benefit from a longitudinal design that could assess causality
between variables, and from the use of comparison groups. This thesis benefited
from the use of comparison groups, a generic model for subjective quality of life, and
from the combination of qualitative and quantitative research techniques.