Mental illness in singapore psychosocial aspects of caregiving 3

Part Study 2: To examine the role of the environmental protectors and their link with the environmental stressor (EE) 124 Chapter 10: 10.1 Method for Study Aim for Study The aim of study was as follows: To identify variables which may be affecting the level of EE within a particular family. In this study it is proposed that the relative’s appraisal, coping style, causal attributions and optimistic disposition will affect the level of EE. 10.2 1. Hypotheses for Study Appraisal of symptoms/behaviours will be associated with the caregiver’s appraisal of the situation such that those caregivers who appraise symptoms and behaviours more negatively will also have a more negative appraisal of the situation. 2. The relative’s negative appraisal of the situation will be correlated with EE, such that a more negative appraisal will be associated with a higher score on the Level of Expressed Emotion scale. 3. More negative appraisal of the situation will be associated with more avoidant coping. More positive appraisal of the situation will be associated with more active coping. 4. A higher score on dispositional optimism will be associated with more positive appraisal of the situation and with more active coping. 5. A longer DUP and/or DUI will be associated with higher EE and a more negative rather than a more positive appraisal of the situation. 125 6. More Avoidant Coping will be associated with higher levels of EE and more Active Coping will be associated with lower levels of EE. 7. A more internal control attribution will be positively correlated with higher levels of EE and with more negative appraisal of the situation. 10.3 Participants The percentages of the different cultural groups in this study reflected the figures found nationally: Chinese 84%, Malay 12%, Indian 6%. For Study those patients of Chinese background who participated in Study were asked to identify their main caregiver and to agree to their caregiver being approached. On obtaining the permission of the patient, the caregiver was contacted and invited to participate in the study. Of the total patients seen (n=161), 135 were Chinese. Of these, 20 refused to allow access to their caregiver and 24 caregivers refused to participate. This left a total of 91 caregivers who were interviewed (57 females 34 males), ages ranged from 26 to 72 years (M = 48 years, SD = 11.3 years) all were of Chinese background. Other demographic details are given in Table 10.1. There were no differences in patient’s EE ratings between those caregivers who participated and those who refused. 126 Table 10.1 Demographic details of caregivers Number Relationship to patient Mother 45 (50%) Father 15 (17%) Sibling 14 (15%) Spouse 10 (11%) Other (e.g. grandparent, uncle, cousin) (4%) Religion of caregiver Christianity 22 (24%) Buddhism 40 (48%) Taoism Others (not specified) 10.4 (10%) 11 (12%) Procedure Appointments were made at the caregiver’s convenience as soon as possible after the patient interview. If the caregiver had accompanied the patient on the clinic visit, the interview was conducted at that time. If this was not convenient, appointments were made at a time and place convenient to them, either at their home or at another time in the clinic. If the caregiver had not accompanied the patient on the clinic visit, the caregiver was contacted by telephone and an appointment made to visit them at their home or a location convenient to them. A bilingual interpreter was available for caregivers who did not speak English. The interpreter spoke Mandarin and one of the Chinese dialects. Mandarin versions of the questionnaires were also available. Questions were read to those who could not read. Interviews typically took between 1-2 hours to complete. 127 10.5 Instruments completed by the caregiver 10.5.1 Appraisal Appraisal was measured by the Experience of Caregiving Inventory (ECI) (Szmukler et al.1996). This is a 66 item inventory comprising ten sub-scales. Eight sub-scales comprise Negative ECI: Difficult behaviours (e.g. Unpredictable); Negative symptoms (e.g. Withdrawn); Stigma (e.g. ‘Covering up his illness’); Problems with services (e.g. ‘Dealing with psychiatrists); Effect on family (e.g. ‘How family members not understand the situation’) ; Need to back up (e.g. ‘Has difficulty looking after money’); Dependency (e.g. ‘Unable to the things you want’) and Loss (‘What sort of life he might have had’). Two sub-scales comprise Positive ECI: Positive personal experiences (e.g. ‘I have learnt more about myself’) and Good aspects of the relationship (e.g. ‘I have contributed to his well being’). Respondents are given the instruction ‘how often over the past month have you thought about…’. They are then asked to mark on a 5-point scale ranging from (never) to (nearly always) for each of the items. The authors of the scale report satisfactory internal consistency, Cronbach’s alpha ranged from 0.74 to 0.91. This scale has previously been used in Singapore (Personal communication, Ms Boon-Keng Seng, Institute of Mental Health, Singapore). For the purposes of this study, Negative ECI was divided into two parts. Two of the Negative ECI sub-scales are concerned with symptoms and behaviours: ‘Difficult Behaviours’ and ‘Negative Symptoms’, these two sub-scales are highly correlated (r(91) = .82, p < .01 two tailed), and are the only sub-scales that are concerned with behaviours or symptoms. These two sub-scales were combined to form ‘Symptom Appraisal’. The remaining sub-scales of Negative ECI were combined and were called ‘Negative appraisal’, these indicate the caregiver’s negative experience of the situation in general. Positive ECI was referred to as Positive Appraisal. 128 10.5.2. Coping Coping was measured by the COPE (Carver et al., 1989) which is a multidimensional coping inventory. It consists of 52 items incorporating 13 conceptually distinct sub-scales: Active coping (e.g. ‘I take additional action to try to get rid of the problem’), Planning (e.g. ‘ I make a plan of action’), Suppression of competing activities (e.g. ‘I put aside other activities in order to concentrate on this’), Restraint coping (e.g. ‘I force myself to wait for the right time to something’), Instrumental social support (e.g. ‘I ask people who have had similar experiences what they did’), Emotional social support (e.g. ‘I talk to someone about how I feel’), Denial (e.g. ‘I refuse to believe that it has happened’), Focusing on and venting of emotions (e.g. ‘I let my feelings out’), Behavioral disengagement (e.g. I give up the attempt to get what I want’), Mental disengagement (e.g. ‘I sleep more than usual), Acceptance (e.g. ‘I learn to live with it’), Positive reinterpretation (‘I look for something good in what is happening’) and Turning to religion (e.g. ‘I seek God’s help’). This can be used to measure dispositional or situational coping. In this case, the situational instructions were used which are: ‘During the past two or three months, how often have you thought about him/her being:… ’. They were then asked to rate their responses on a four point scale from (I usually don’t this at all) to (I usually this a lot). The original authors report internal consistencies measured by Cronbach’s Alpha of .54 to .90. Test-retest reliabilities were fairly stable over six and eight week periods ranging from .42 to .89. The COPE has been used previously in Singapore (Bishop et al., 2001; Diong & Bishop, 1999; Khoo & Bishop, 1996). 10.5.3. Control Attributions Control attributions were measured by the Control Attributions Scale (Greenley, McKee, Stein & Griffin-Francell, 1989, cited in Greenberg, Kim & Greenley, 1997). This is a five point scale which measures the degree to which a caregiver 129 believes the patient is in control of their behaviours. Participants respond using a point scale from (strongly agree) to (strongly disagree). Example items would be, “my son/daughter can make his/her strange thoughts and feelings come and go at will” and “My son/daughter doesn’t try hard enough to get better”. The reported internal consistency of the scale is .70 (Greenberg et al., 1997). This scale had not been translated into Chinese nor used in Singapore previously. The scale was translated by a fluent English/Chinese Singaporean. It was then back translated by a separate translator; there were no discrepancies in the back translation from the original. Given the brevity of the scale and the face value of the questions it was not felt necessary to further validate the scale. 10.5.4. Dispositional Optimism Dispositional Optimism was measured by the revised Life Orientation Scale (LOT-R; Scheier et al., 1994). The LOT-R is a revised version of the original LOT (Scheier & Carver, 1985) and was designed to eliminate some of the content overlap with coping that existed in the original LOT. The LOT-R scale consists of six items measuring optimism and four filler items. Three items are negatively worded (e.g. ‘If something can go wrong for me it will) and three are positively worded (e.g., ‘In uncertain times, I usually expect the best.’). Scores on the negatively worded items are reversed and summed with scores on the positively worded items to obtain a single summary score. Participants are asked to indicate the extent to which they agree with the items on a point scale from (strongly disagree) to (strongly agree). Scheier et al., (1994) report internal consistency for the scale to be .78 with test-retest reliability at months of .68 and 12 months at .60 suggesting the scale is stable over time. The LOT and LOT-R are also highly correlated (r=.95) which ensures continuity of findings (Scheier et al., 1994). This scale has also been validated on a Hong Kong Chinese sample and was found to have good reliability and validity, Cronbach’s alpha .70 and 130 .65 after months (Lai et al., 1998). This scale has also previously been used in Singapore (Khoo & Bishop, 1996). 10.5.5 Comparison of Caregivers’ and Patients’ perspectives of the family atmosphere. In order to examine the extent to which the patient’s perception of the family atmosphere would match that of the family, a subset (n=36) of patients completed the Family Relationships Index (FRI: Holahan & Moos, 1981, 1982). The FRI comprises three subscales of the Family Environment Scale (FES: Moos & Moos, 1981, 1986) which is a 27-item index of the quality of family relationships. The three subscales form the relationship domain: ‘Cohesion’ which is the degree of commitment, help and support family members provide for one another, (e.g. ‘Families members really help and support one another’); ‘Expressiveness’ which is the extent to which family members are encouraged to act openly and to express their feelings directly, (e.g. ‘We say anything we want to around home’) and ‘Conflict’ which is the amount of openly expressed anger, aggression and conflict among family members. (e.g. ‘Family members often criticize each other’). The respondents answer True (1) or False (2) to the statements. The responses for each of the sub-scales are then summed to produce scores. It has been shown to have good internal consistency (alpha = .89), moderate association between the three scales (r= .43) and good construct validity (Holahan & Moos, 1981). 10.6 Instruments completed by the patient The measure of Expressed Emotion was provided by the patients in Part 1, DUP and DUI are also taken from Part 1. 131 10.7 Validation of Concepts The concept of caregiving and the concept of EE were both investigated. In order to this, ten patients from the EPIP programme who had been declared stable by the psychiatrist, were approached and asked to complete the LEE. All gave their permission for their caregiver to be contacted and all caregivers agreed to take part in an in-depth interview. The research question was general in nature to gain an overall understanding of the relative’s experience of living with his/her mentally ill relative. 10.7.1 Concept of Caregiving As the notion of caregiving is an important part of this study, it was felt necessary to ascertain that the caregiving experience in Singapore did not differ significantly from that of western caregivers. This would have two specific benefits: (1) the Experience of Caregiving Inventory (ECI) could be used with confidence in this study, and (2) should there be aspects of caregiving that are unique to Singapore, these could be detected, and used to better inform psychoeducation or interventions with caregivers. In an attempt to explore the data fully and in a manner not driven by any one theory, the in-depth interviews were analysed using a grounded theory approach. Generally speaking, the issues raised in the in-depth interviews were very similar to those experienced by caregivers in Western countries. The same emotions, practical issues and coping strategies emerged and it was felt that the ECI could be used with confidence. Full details and results are given in Appendix B. 10.7.2 Concept of Expressed Emotion Before attempting to measure EE in a Singaporean population, it was necessary to firstly investigate the existence of this concept in Singapore. To this, the data from the in-depth interviews was analysed looking at four attitudes reported by 132 Leff and Vaughn (1985) that they found identified those caregivers high in EE from those in low in EE. The aim of this part of the study was to investigate the phenomenon of EE in Singapore and to see if there were any differences in its manifestation from Western counterparts. The data from the in-depth interviews gave good initial evidence that similar attitudes are displayed by both caregivers in the West and in Singapore, and, most importantly, that these attitudes could discriminate between high and low EE families in this population (as measured by the LEE). Full details and results of this procedure are given in Appendix B. 133 desires. This can often cause tension because what one wants for oneself may not include one’s duty to another. This tension or stress then results in expression of anger, etc, against the patient. On the other hand in a collectivist society, the caregiver faced with the same situation does not have the dilemma of choosing between what is best for the other, and what is best for the self, because, they are essentially the same thing. So, perhaps in this situation the caregivers are not faced with the dilemma of choosing their own desires over their obligations, so the resulting tensions and subsequent negative behaviors not arise. In comparing the rates of Negative Appraisal for this study with those from Western cultures, there are certain areas of negative appraisal that seem to be somewhat lower than Western studies: ‘Problems with services’, ‘Effects on the family’, and ‘Loss’. For the first item it could be that the EPIP programme in which these individuals were involved was providing a very satisfactory service. With regards to ‘Effects on the family’, cultural factors may help to explain this. There are certain items within this scale which are not relevant in this population, for instance ‘Others leaving home because of the effect of the illness’ and ‘The illness causing a family break-up’. As mentioned earlier, it is not customary in Chinese societies for adult children to leave home until they are married, this is also the case in Singapore (Hsu, 1995) so even if the illness was affecting other family members, leaving home would not be an option. Similarly, the traditional view of the importance of the family in Chinese culture (discussed earlier) would mean that the family remaining intact would be the priority. Regarding ‘Loss’, there are reports (Birchwood, 1992) that first-episode families have not yet fully experienced loss as the illness is in its early stages, and it is not until sometime later in the course of the illness that relatives fully experience loss. This may be what is happening here. 160 12.1.4 Positive Appraisal It is only fairly recently that positive aspects of the caregiving role have been considered alongside negative aspects. In this study, a medium effect size was seen for positive appraisal, with Low EE groups having a more positive appraisal than High EE groups. Positive Appraisal had a significant negative direct effect on EE in the multivariate analysis. Other studies which have looked at Positive Appraisal and EE have not found a difference between high and low EE caregivers. Looking at the rates of Positive Appraisal in other cultures (e.g. Addington et al., 2003; Szmukler et al., 1996), there is very little difference in the actual scores so cultural factors must be explored to explain the difference. Greenberg, Greenley and Benedict (1994) suggest that the caregiving task in collectivist cultures may be seen as a source of reward instead of burden. This could be related to the concept of Yuan just discussed. In addition, if caregiving is seen as not only positive, but rewarding, then this may engender positive affect in the caregiver which may result in more emotional support. This may explain the very strong negative correlation between Positive Appraisal and LES; caregivers were expressing more warmth (either in physical, practical or emotional ways) towards the patient. Cultural beliefs about illness may be also be underlying the relationship between positive appraisal and EE. In the in-depth interviews, many caregivers believed that the illness was caused by the patient ‘thinking too much’. This may be reflecting the concept of ‘xiao xin yan’ which mainland Chinese caregivers have also given as a cause of the illness (Yang, 2003). Yang (2003, p.22) describes this concept as ‘someone who is narrow minded, overly sensitive, selfish and picky; petty and not able to let go of thoughts or feelings’. It is believed that after exposure to a stressor, the patient’s inability to let go of their thoughts increases the mental strain which results in the illness (Yang 2003). Use of this concept as an explanation has several advantages for the caregivers. First, they can avoid admitting that that the relative has a mental 161 illness and perceive them as more ‘normal’. Second, the Chinese believe that narrow minded people can be persuaded to adopt a different way of thinking and this may make them more helpful towards the ill relative. Third, it prevents the stigmatizing label of schizophrenia or psychotic and may prevent subsequent feelings of shame and its associated consequences (discussed later). It is clear how each of these factors could make the situation seem better for the caregiver. If they perceive the person as normal they will be less affected by the situation, and be less likely to display high EE attitudes and behaviours. In fact, as mentioned above, they may be more helpful and supportive in an effort to change the person from their narrow minded ways. 12.1.5 Language difference in Positive Appraisal A difference in positive appraisal was seen between the Chinese respondents and the English respondents with the latter having higher positive appraisal; the difference actually lay in the sub-scale ‘Positive personal experiences’. There are two potential reasons for this difference. The first is that it was due to a poor translation of this scale, however the ECI had been used before in Singapore and validated for use in this population, and it would seem odd that one scale out of eight suffered from translation problems. The second reason may be that it was due to the difference in the backgrounds of the Chinese respondents to the English respondents. It has been noted that Chinese speakers (first language) tend to adhere more to traditional Chinese practices than English speakers in Singapore (Chang et al., 2003). The English group showed a significantly higher score on this subscale and examination of the items in the scale may shed some light on this. The sub-scale consists of items which all start with the pronoun ‘I’ e.g. I have learnt more about myself. The only other sub-scale which contains items starting with ‘I’ is the other sub-scale which makes up Positive ECI, ‘good aspects of the relationship’ which has items which start with ‘I’. 162 To focus on the self, and to take personal credit for something reflects a very individualistic oriented focus, by contrast, in the collectivist tradition which is found in many Asian countries, the focus is on the group and group harmony; someone from a collectivist culture would not readily take credit in this way. Although there was no overall language difference for the other positive ECI subscale, further examination of the items within the sub-scale ‘good aspects of the relationship’ showed that the language groups also differed significantly on two of the three items which started with ‘I’. This suggests that collectivism may be a factor here. Including a measure of individualism and collectivism could help clarify these issues in future studies and avoid biases in imported scales such as the one found here. 12.1.6 Summary – Appraisal The caregiver’s perception of the patient’s symptoms and behaviours is positively associated with how he or she perceives the situation generally. The lack of relationship between Negative Appraisal and EE is possibly explained by the Chinese cultural concept of Yuan and the concept of Collectivism. A further Chinese concept is suggested to explain the association between Positive Appraisal and EE, that of xiao xin yan. Collectivism may also be contributing to differences in scores on some of the sub-scales of the ECI between Chinese and English speaking respondents. The importance of the family is also stressed. Cultural factors have a large role to play in Appraisal and evidence of several of the cultural factors outlined by Jenkins and Karno (1992) outlined in Chapter can also be seen here. For instance, the ‘cultural meaning of kin relations’ can be seen in the concept of Yuan and ‘cultural beliefs about illness’,in the form of xiao xin yan. No evidence was seen for the violation of cultural rules; this is something that may need to be explored in a future study through qualitative research. Regarding vocabularies of emotion, this factor is seen in this study as relating more to attributions and is discussed later. 163 12.2 Coping 12.2.1 Coping and EE The literature on EE and coping strongly suggests that EE is associated with Avoidant Coping, the hypothesis therefore reflected this. However, this hypothesis was not supported in this study. Contrary to the quite substantial support for the relationship between Avoidant Coping and EE in the literature (e.g. Hinrichsen & Lieberman, 1999) there was no relationship between these variables for this population. For instance, Raune, Kuipers and Bebbington (2004) in their study of first episode caregivers found that of the sub-scales they used for Avoidant Coping (Behavioural disengagement, Mental disengagement, Alcohol/drug use and Denial) all except Denial were significantly associated with high EE. In this study the LEE(S) or its sub scales were only correlated (negatively and significantly) with Active Coping sub scales. This correlation however, failed to reach significance when Positive Appraisal was controlled for. So neither Active nor Avoidant Coping in this study was associated with EE. Further, Active Coping was not a mediator between Positive Appraisal and EE. This does not lend support to the literature (e.g. Birchwood & Smith, 1987) which suggests that coping may explain the difference between high and low EE. Nor does it lend support for those researchers who adopt a stress-coping-appraisal approach to studying EE. The fact that Active Coping was not directly associated with EE does not mean that it has no part to play in the model. Its role in influencing appraisal, rather than vice versa, is now discussed. 12.2.2 Coping leads to appraisal? The path model which posited appraisal leading to coping (Model 1) was chosen over the alternative model (coping leading to appraisal, Model 3) on the basis of the AIC scores. However, it cannot be ignored that this model also provided a very good fit to the data and provides a plausible explanation of what is occurring in this 164 situation. Research into positive affect may help to explain what is occurring in this situation. Folkman and Moskowitz’s (2000) research into AIDS caregivers produced the finding that three types of coping were related to the occurrence and maintenance of positive affect: Positive Reappraisal, Problem Focused coping (such as planning, task oriented, active coping) and infusing ordinary events with positive meaning. The first two clearly fall within the realm of what in this study was termed Active Coping. So, it could be hypothesized that in this study, the caregivers’ active coping led them to initiate, or increase, positive affect which would lead to more positive appraisal of the situation. Folkman and Moskowitz’s (2000) claim that progress in understanding exactly how coping works has been limited because it has relied mainly on research involving negative outcomes. While it is widely accepted that negative affect is associated with stress, research has shown that positive affect can also occur at times of great stress (e.g. Viney, 1986). Folkman and Moskowitz’s claim that positive and negative affect can occur during the same stressful period has direct relevance for this study. Caregivers had both negative and positive experiences of caregiving. The two types of coping identified above, Positive reappraisal and Problem focused coping, have direct relevance for this study, but how they bring about positive affect? Looking first at Positive reappraisal, this involves reframing a situation to see it in a positive light. Research has shown that positive reappraisal can involve the activation of deeply held values, for instance, caregivers have reported that their caregiving duties have been a way of actively showing their love; the often burdensome duties of caregiving were reappraised as worthwhile and were seen as having value (Folkman, Chesney & Christopher-Richards, 1994). This may also be related to the practical way that Chinese families (particularly mothers) have of showing love. Having a child with an illness may actually give the caregiver more opportunities to show love by practical actions, taking to the hospital, giving 165 medication, etc. So the caregivers here may have reappraised the situation and seen it as an opportunity to care practically for an adult child, in much the same way they would have done when the child was young. This clearly has implications for mothers in this study but more work needs to be done to identify the implications for spouses and other relatives. Turning now to Problem Focused coping, the underlying idea here is that people try to gain some control over a usually uncontrollable situation. Generally, caregivers attempt to identify situation specific goals and work towards them. This focused approach gives the caregiver a sense of accomplishment, mastery and self efficacy, all of which contribute to the feeling of positive affect. This presents opportunities for interventions, where caregivers could be helped to clarify these situation specific goals. Examples of situation specific goals could be getting the patient to comply with his or her medication, keeping hospital appointments, and so on. 12.2.3 Coping in this population In this study, the types of coping measured by the COPE were found to be applicable for this population. The in-depth interviews did not suggest the use of any other coping strategies not covered by the COPE. Regarding coping in general, the sub-scales Positive reinterpretation and growth, Acceptance, Turning to religion and Active coping were used most in this study. Behavioural Disengagement, Denial and mental disengagement were used least. This follows very much the findings of Bishop et al. (2001). Although Bishop et al. used the dispositional version of The Cope, moderate correlations have been found on most of the sub-scales between the dispositional and situational versions (Carver et al., 1989). This finding lends support for the idea that the types of coping strategies used are the same across different populations. Looking in detail at the individual sub-scales of the COPE and how they relate to appraisal, it seems that Focus on and venting of emotions was positively 166 related to Negative and Symptom Appraisal. Also of note is that both the social support sub-scales were strongly positively related to Positive Appraisal. These sub-scales just mentioned were not examined in this study as they did not come under Active or Avoidant Coping, however their utility should be investigated in a future study. 12.2.4 Differences between men and women on Avoidant Coping A difference was noted between men and women on certain sub-scales within Avoidant Coping: ‘Turning to religion’ and ‘Denial’. A sex difference in using religion as a form of coping is consistent with findings on gender differences reported in the literature. In a meta analysis, Tamres, Janicki and Helgeson (2002) found that in response to coping with others’ health, women used Religion significantly more than men. Looking at caregivers coping with a relative with schizophrenia specifically, Magliano et al. (1998) found that women used religion as a coping strategy more often than men. With regards to the use of Denial, Tamres et al. (2002) found no difference between the sexes on this coping strategy. Chan, Tang and Chan (1999) suggest that women from collectivist cultures tend to use less assertive and more indirect coping strategies, such as denial, to avoid being labeled as troublemakers and to avoid disrupting the harmony of the group. So, sex differences in Denial as a coping strategy may be explained by the concept of Collectivism, but clearly this needs to be explored further in a future study. 12.2.5 Summary – Coping Based on theory, the model which describes coping leading to appraisal to EE is preferred to one that hypothesizes coping as a mediator between appraisal and EE. The coping strategies of the COPE seem to be relevant for this population with the possible exception of the sub-scale Denial. 167 12.3 Optimism 12.3.1 Optimism and Positive Appraisal Regarding the role of optimism, the hypothesis was that those caregivers who were more optimistic would also have a more positive appraisal of the situation and use more active coping strategies. This hypothesis was partially supported in this study. The path model clearly showed that optimism had a direct effect on appraisal. This supports Khoo and Bishop’s (1996) suggestion that optimists may appraise the situation more positively. There are several theories that go some way to explaining the relationship between optimism and positive appraisal. One of these, the reformulated learned helplessness theory (RLHT; Abramson, Seligman & Teasdale, 1978) suggests an attributional approach to explaining optimism. This theory suggests that optimistic explanations for negative events are seen as being external, unstable and specific. In this way, caregivers with an optimistic disposition would be likely to see the illness as being caused by external factors, of a changeable nature and affecting only specific areas of their life. The theory also states that the unstable and specific nature of the optimists’ explanation allows them to feel in control, they are seen to be able to remain motivated when faced with negative situations and ultimately cope more effectively (Gillham, Shatte, Reirich & Seligman 2001). So this theory would explain why caregivers high in optimism, would have a more positive outlook. 12.3.2 Optimism and Active Coping The findings of this study not support the large body of research which suggests that optimism works through coping. Although Optimism and Active Coping were not significantly correlated, the path analysis showed an indirect effect (.19, p = .01) of Optimism on Active Coping. This is possibly due to the significant relationship of Optimism with the sub-scale, Positive reinterpretation and growth. Optimism was 168 also significantly correlated with the coping sub-scale Social support for emotional reasons; it may therefore be useful in a future study to look more closely at the sub-scales of the COPE rather than relying on composites such as Active Coping where information may be lost. The indirect effect of optimism on Active Coping may be explained by Aspinwall and Richter (1999). These authors believe that caregivers high in Dispositional Optimism may be more likely to engage in active coping because optimists are able to recognize that some tasks are unsolvable, and suspend further coping efforts on these tasks. This enables them to put their time and resources to better use focusing instead on solvable tasks. So, these caregivers may have been able to recognize that some aspects of the situation were unsolvable, such as getting back their son/daughter to pre-illness state, and were able instead to concentrate on controllable aspects, such as providing support, ensuring compliance with medication, and so on. 12.3.3 Alternative personality variable – Shame Proneness In this study DO was examined as a personality variable of the caregiver that may affect levels of EE. Apart from its relationship with positive appraisal, DO did not influence other aspects of the situation. Drawing on some ideas from the vocabularies of emotion outlined by Jenkins & Karno (1992), another personality variable which presents itself as a possible explanation of high EE in this population is that of shame-proneness (the tendency to experience shame across a variety of situations). One aspect that has been examined by researchers as potentially contributing to high EE in caregivers is guilt-proneness (Bentsen et al., 1998). Whilst these authors concluded that guilt-proneness is one determinant of higher EE, it seems that given the prominence of the emotion of shame in the Chinese culture, that this might be a more relevant emotion to consider. Further, the consequences of shame and shame-proneness seem to relate more to the nature of high EE. 169 As mentioned earlier Shame is a central feature in Chinese culture (Li, Wang & Fischer, 2004). Although shame can be found in both collectivist and individualistic cultures, it is seen more extensively in collectivist cultures (Fung & Chen, 2001) where it is seen as a group rather than an individual concern (Wilson, 1981). As with any collectivist culture the group is of paramount importance and something like shame would be reflected onto the group to which the individual belongs so when the individual experiences shame, the group experiences shame. Although these notions are steeped in traditional Chinese thinking, the concept appears to still have relevance in today’s society; in a recent study of US and Chinese College students, Chinese students were seen to extend their emotions of shame and guilt to their close relatives significantly more than the Americans (Stipek, 1998). According to Tangney (1991), personality characteristics such as shame and guilt-proneness influence the interpretation of interpersonal events, and affect subsequent behaviour. As discussed earlier, mental illness still carries with it a great deal of stigma, resulting in feelings of shame and embarrassment (Teschinsky, 2000). Compared to guilt-prone individuals, shame-prone individuals are more likely to respond to a negative situation such as a relative with a mental illness, by projecting blame, becoming more angry, showing less sympathy and less empathy, and avoiding the situation (Tangney, 1995). It is clear to see how shame-proneness could lead to high EE behaviours: avoidance of the situation (withdrawing from the patient; lack of emotional support), anger (criticism and hostility), lack of empathy (lack of emotional support). A further related finding is that highly shame prone people deal very poorly with interpersonal conflict, in addition, their responses to such conflict tend to make the situation worse rather than better. (Covert, Tangney, Maddux &Heleno, 2003). This is an area which needs further exploration. Tangney, Wagner & Gramzow (1989) have formulated the Test of Self-Conscious Affect (TOSCA) which assesses shame- and guilt-proneness. A future study could examine this personality variable in this 170 population, to clarify the role of shame- and guilt-proneness as predictors of EE. 12.3.4 Summary – Optimism Optimism is seen to be related to appraisal, rather than coping in this study. The internal consistency on LOT-R in this study was also very low, indicating that it is not, in this case, a reliable measure. Perhaps the LOT-R is not accessing all aspects of optimism completely in this group of caregivers. To clarify this situation, a future study could perhaps use the Chinese Affect Scale (CAS: Hamid & Cheng, 1995) in conjunction with the LOT. The CAS is a measure of positive and negative affect based on affective terms indigenous to the Hong Kong Chinese. Lai (1997) found correlations between higher scores on the LOT and higher scores on positive affect and lower scores of negative affect as measured by the CAS. Clearly, the affective terms would need to be validated for a Singaporean population but this would give a better indication of how exactly the Singaporeans express their optimism. Finally, the personality variable shame-proneness is suggested as a contributor to EE, to be investigated in a future study. 12.4 Control Attributions In this study, it was hypothesized that caregivers who believed their relatives to be more in control of their illness, i.e. who held more internal control attributions, would also be higher in EE, and have more negative appraisals. 12.4.1 Control Attribution and Negative Appraisal With regard to attributions and appraisal the hypothesis was supported. Higher internal control attributions were positively correlated with Negative Appraisal and Symptom Appraisal. This supports findings from the literature (e.g. Greenberg et al., 1997) where attributions have been linked to higher burden. In addition this study 171 showed how more internal control attributions are linked to the caregivers’ perception of illness symptoms and behaviours as being more troublesome. 12.4.2 Control Attributions and EE With respect to Attributions and EE the hypothesis is also supported with more internal control attributions associated with higher EE. This supports the large body of research (e.g. Barrowclough et al., 1994) on this subject. The strong association of Control Attributions with EE and with Negative and Symptom Appraisal also adds to the validity to the control attributions scale. This is important as previously many studies have rated attributions from caregivers’ speech samples or interviews. If a short, easy to administer scale like the Control Attributions Scale can be seen to be a valid way of detecting these caregivers’ attributions then this is advantageous in a clinical setting. Of the sub-scales of the LEE, more internal control attributions were associated most strongly with Lack of Emotional Support; this is interesting as previous studies have found the association to be strongest with Criticism. Once again, this can be seen to be supporting findings from Lopez et al.’s (1991) study on Mexican Americans. Similarities between this cultural group and the Singaporean group were noted earlier in Part 1. Lopez et al. found that lower internal (i.e. more external) attributions were related to more warmth, and the LES sub-scale is perceived as the inverse of warmth. Looking at this in relation to Weiner’s theory, the caregiver perceives the illness is controllable by patient and therefore the patient is seen as responsible. This results in an emotional reaction – Weiner suggests that a variety of emotions may be produced, anger being the most obvious however other emotions can be grief, shame, guilt etc. The emotion is hypothesized to result in a behavioural reaction. In other studies conducted in Western populations, anger is seen to be the emotion which is believed to result in criticism (the behavioural reaction) against the patient. However, Weiner 172 also posits other behavioural reactions one of which is ‘neglect’. In this population, the behavioural reaction of neglect could be seen to be lack of emotional support. If the caregiver feels an emotion (e.g. anger), then this may result in him or her avoiding the patient, which could be construed by the patient as being lack of emotional support. Of course, based on earlier discussions, shame rather than anger, may be the emotion aroused in this population which leads to a behavioural reaction. In Chinese culture it is thought that overt expression of an emotion such as anger may be damaging to the organs and therefore it may not be displayed as much as in western societies (Hsu, 1995) Although the relationship with Attributions and EE (in particular LES) was significant it was not a strong relationship. There is another explanation that may be relevant for explaining the importance of LES in this study. The Chicago Consortium for Stigma research (CCSR: 2005) suggests that along with responsibility attributions, an equally prominently held belief in both Western populations (Link, Phelan, Bresnahan, Stueve & Pescosolido,1999) and Chinese populations (Tsang, Tam, Chan & Cheung, 2003) is that people with mental illnesses are a danger to society. As part of a General Social Survey in America, Link et al (1999) found that 61% of the population believed that a person with schizophrenia was very likely to be violent. These authors further state that the dangerousness stereotype associated with mental illness was increasing over time rather than decreasing. The more that the general public comes into contact with mental illnesses such as schizophrenia, the less likely they are to believe that people with schizophrenia or depression could be dangerous (Angermeyer, Matschinger & Corrigan, 2004). However, for many of these caregivers this is their first experience of mental illness and it is possible that the dangerousness stereotype was still salient for them. It is thought that if a caregiver believes that a patient is, or could be dangerous, this results in fear (Corrigan, 2000; Wolff, Pathare, Craig & Leff, 1996). 173 There are two hypothesized routes which the fear can then take. The first follows Weiner’s control attribution theory: dangerousness control attribution emotion behavioural response The second follows a direct route (Corrigan et al., 2002; Angermeyer & Matschinger, 1997): dangerousness fear avoidant behaviour So, it could be that caregivers in this population perceive dangerousness in the patient, which produces fear and which causes them to adopt avoidant behaviour such as withdrawing their support from the patient, i.e high LES. Regarding cultural influences on attributions, research has shown that collectivist cultures such as the Chinese tend to make more external, rather than internal attributions (Morris & Peng, 1994). In this study there was no significant inclination to make external attributions so perhaps Singapore is following a more individualist approach in this regard. Future studies could explore this further. 12.4.3 Summary – Control Attributions Control attributions are a very important part of this model being associated with EE and Appraisal. In light of Weiner’s theory of attribution, it is suggested that the behavioural reaction raised is that of neglect, leading the caregiver to withdraw support from the patient. As mentioned earlier, lack of emotional support underpins the 174 concept of EE in this population. It is further suggested that the caregiver may perceive the patient as dangerous, resulting in fear and avoidant behaviour. This could also result in lack of emotional support. 12.5 DUP and DUI In this part of the study, the relationship between the length of DUP and/or DUI on the caregivers’ appraisal was investigated. From the literature it is noted that those caregivers living with a patient with a longer DUP and/or DUI had a more negative appraisal of symptoms (Harvey et al., 2001) however, no relationship was seen between DUP and symptom appraisal. Caregivers living with patients with longer DUI on the other hand, appraised the symptoms more negatively. This supports Harvey et al.’s (2001) finding that longer illness is associated with more negative perception of the situation. Once again it is DUI rather than DUP that is important in this population. This is perhaps related to the fact that caregivers find it easier to accept positive symptoms as being features of an illness. Prodromal features on the other hand are associated with negative symptoms, such as lack of hygiene, withdrawal, etc and it has been shown that these symptoms are more difficult for the caregiver to cope with primarily because they are less obviously illness related than the positive symptoms and in many cases have been present for over a year. A further factor is that the caregiver will have been dealing with these symptoms prior to meeting with the psychiatrist so will have been unaware that these were attributable to an illness and not to laziness or the part of the patient. It is easy to understand how the caregiver could be worn down by trying to cope with and understand these symptoms. 175 [...]... reinterpretation and growth 11.78 (2.71) 66 11.97 (2.58) Acceptance 11.47 (2.45) 53 11 .35 (2.48) Turning to Religion 11. 13 (3. 95) 89 9.84 (4.91) Active coping 11.04 (2.75) 67 10.61 (2. 43) Planning 10.86 (2.74) 69 11.44 (2.72) Suppression of competing activities 9.55 (3. 01) 68 10.01 (2 .30 ) Social support for emotional reasons 9 .35 (3. 24) 83 9.48 (3. 18) Social support for instrumental reasons 9 .34 (3. 22)... (3. 22) 77 11 .34 (2. 93) Focus on and venting of emotions 8.78 (2.74) 68 7.18 (2.92) Restraint coping 8.76 (2.61) 48 9.64 (2.24) Mental disengagement 7.27 (2.58) 67 7.28 (2.79) Denial 6 .36 (2. 43) 66 5. 13 (2.98) Behavioural Disengagement 6.25 (2. 43) 74 5 .33 (3. 05) Active Coping 63. 46 (11.64) 86 * Avoidant Coping 31 .02 (7. 53) 80 * Sub-scale totals 1 Bishop et al (2001), study on police officers (n=2 43) * not... (5.86) 13. 3 (7 .3) Effects on family 7. 63 (4.47) 65 9.4 (5.86) 10.4 (6.6) Need for backup 10.75 (4 .39 ) 50 10 .3 (5.45) 8.9 (5.7) Dependency 9.17 (3. 95) 64 10.5 (4.45) 10.1 (4.7) Loss 9.62 (4.56) 66 12.5 (5.56) 12.5 (6.9) 73. 54 (26.62) 92 87.6 (33 .65) 82.5 (*) Positive experiences 17.19 (6. 93) 83 15.8 (6 .35 ) 16 .3( 7.1) Good aspects 14. 53 (4.12) 69 14.0 (4 .32 ) 12.8(4.4) Total positive ECI 31 .47 (9 .31 ) 83 29.85... Sub-scale Suppression of competing activities 25* 28* 19 05 Acceptance 25* 13 36* -.08 Restraint coping 09 23 22 -. 03 Positive reinterpretation and growth 19 60* 09 -. 23 Planning 14 50* 14 12 Active coping 03 31* 14 10 Active Coping total 22 48* 26* 11 Behavioural Disengagement 10 08 08 -.21 Turning to Religion 08 -.04 07 13 Denial 19 06 10 -. 13 Mental disengagement 25* 26* 09 -.10 Avoidant Coping Total 22... be seen in Table 11 .3 Internal consistency was low on Restraint Coping and Acceptance sub-scales, all other internal consistency scores were average to good Missing data accounted for less than 5% and was again replaced using SPSS missing values (linear interpolation) Following Diong & Bishop (1999), the sub-scales Active coping, Planning, Restraint coping, Acceptance, Suppression of competing activities... Females (n = 34 ) t (df=89) p (n = 57) 28.05(5.94) 32 .78(7.86) 3. 03 0 03 68 Turning to religion 9.61 (3. 94) 12.04 (3. 4) 2.95 004 66 Denial 5.66 (1. 83) 6.79 (2.65) 2 .39 02 50 English Chinese (n = 45) (n = 46) 33 . 83( 7.49) 29.17(10 .38 ) 2.46 02 51 Avoidant Coping Positive Appraisal Positive personal experiences 19.40(5.48) 16.01(5.46) 11.2 008 62 Descriptive statistics for variables 11.2.1 2.70 Appraisal of situation... Focus on and venting of emotions 26* 10 30 * -.11 Social support for emotional reasons 16 45* 11 23 Social support for instrumental reasons 03 39* 08 15 Avoidant Coping Sub-scale Remaining COPE sub-scales *p < 007 (one-tailed) 1 43 11.6 Testing of Hypothesis 4 A higher score on Dispositional Optimism will be associated with more Positive Appraisal of the situation and with more active coping Optimism was... non-significant original paths and additional paths This revised model presented an excellent fit to the data (see Table 11.10) and explained a substantial amount of variance in the model, 13% of EE, 44% of Negative Appraisal, 32 % of Active Coping, 15% of Positive Appraisal and 8% of Appraisal of Symptoms However, the addition of the path from Positive Appraisal to EE resulted in the path from Active Coping to... SPSS Missing Values using linear interpolation was used to replace them 135 Table 11.2 Comparison of Experience of Caregiving Inventory (ECI) sub-scales with Western Populations (n = 91) Present Study Comparison Studies Mean (SD) α Mean (SD)1 Mean (SD)2 Difficult behaviours 11.20 (7. 13) 89 13. 9 (7.91) 11.2 (6.4) Negative symptoms 9.64 (5.78) 86 13. 4 (6.20) 10.6 (5.6) Stigma 6.47 (3. 92) 65 6 .3 (4.75)... Attributions 31 ** 22* 28** 44† Negative Appraisal 22* † 08 50*** Expressed Emotion Avoidant Coping Symptom Appraisal 13 † 05 † 15 Positive Appraisal -.28** 38 *** Optimism Figure 11.2: Model 2: Predictors of EE with Avoidant Coping as the mediating variable † *** ** * Indicates significant path in original model Indicates path added as suggested by modification indices Indicates non-significant path in original . 9.55 (3. 01) .68 10.01 (2 .30 ) Social support for emotional reasons 9 .35 (3. 24) . 83 9.48 (3. 18) Social support for instrumental reasons 9 .34 (3. 22) .77 11 .34 (2. 93) Focus on and venting of emotions. relative. 10.7.1 Concept of Caregiving As the notion of caregiving is an important part of this study, it was felt necessary to ascertain that the caregiving experience in Singapore did not differ. 133 Leff and Vaughn (1985) that they found identified those caregivers high in EE from those in low in EE. The aim of this part of the study was to investigate the phenomenon of EE in Singapore