Coronavirus term paper

Anxiety, Depression, and Anger: Core Components of Negative Affect in Medical Populations

Ephrem Fernandez and Robert D. Kerns

NEGATIVE AFFECT

Within psychology, the term ‘affect’ has evolved out of restricted usages within psy- choanalysis and clinical psychiatry into a general term that refers to any kind of subjec- tive feeling (Tomkins, 1962). Imposed with a metaphor from chemistry, affect is now regarded as either positive or negative in valence, the former implying pleasant feelings and the latter implying unpleasant feelings. Other terms used interchangeably with negative affect are ‘dysphoria’ and ‘distress’, though sometimes the words ‘stress’ and ‘suffering’ are also used to loosely suggest negative affect. The main point of consensus is that negative affect refers to any form of subjective feeling that is experienced as unpleasant in quality. Such unpleasantness can also vary quantitatively, that is, on a dimension of intensity. This common property of affect (be it positive or negative) is also labeled as activation or arousal.

Various types of negative affect have appeared in the diagnostic criteria for psychi- atric disorders (e.g. schizophrenia, post-trau- matic disorder, borderline personality disorder, obsessive-compulsive disorder) partly because ‘distress’ is regarded as one of the associated features of all mental disorders (American Psychiatric Association, 2000). Yet people with somatic complaints of med- ical disease have rarely been examined for clinically significant levels of negative affect. This is probably an outcome of the mind–body dualism that has infused the health sciences for centuries. In this chapter, we report on some of the recent findings that do point to a spectrum of negative affect in medical populations. Supported by theoreti- cal foundations and empirical data, we direct our attention to three specific types of nega- tive affect: anger, fear, and sadness, or their clinical equivalents of anger, anxiety, and depression, respectively. This, we call the core of negative affect (CONA). With refer- ence to medical populations, we focus on the

32

9781412946513-Ch32 5/12/08 3:25 PM Page 659

three highly prevalent ailments in developed as well as developing countries of the world: cardiovascular disease (CVD), cancer, and HIV/AIDS. Furthermore, we draw parallels between the CONA as manifested in these populations and CONA as already researched in one population: patients who suffer from pain.

ANXIETY, DEPRESSION, AND ANGER

Anxiety and depression have often been studied as twin features of negative affect, but more recently, anger has been introduced as a close relative to form a new triad of neg- ative affect. Barlow (1991) made a bridge between the experimental psychology of emotions and the clinical psychology of emotional disorders, by postulating how fear, sadness, and anger lie at the root of anxiety, depression, and anger disorders. Spielberger et al. (1995) grouped depression, anxiety, and anger under the label of ‘emotional vital signs’, a construct later echoed by Ghosh and Puja (2004). Examining pain patients as a ‘test population’, Fernandez et al. (1999) and Fernandez (2002) showed that there is ample empirical evidence to position anxiety, depression, and anger within the core of negative affect.

Vital signs

The idea of emotional vital signs was origi- nally spun out of Spielberger’s view that anx- iety was analogous with heart rate, anger with blood pressure, and depression with fever. The analogy may not be perfect since the term ‘vital signs’ as used in medicine refers to objective signs that the systems of the body (required to keep a person alive) are in working order or normal. When measured values for respiration, heart rate, blood pres- sure, and temperature are zero, the person is evidently dead; when they reach a certain norm for the species, the organism is essentially

alive and well. In the case of anxiety, depres- sion, and anger, zero values would point to healthy emotional functioning, while high values, though not necessarily a sign that life is threatened, do raise concerns for the well- being of oneself or others. Profound depres- sion could forebode suicidality, extreme anger could potentiate acts of destruction, and high-grade fear could be crippling or dis- abling. In that sense, if one were to select three affective types as indices of a person’s emotional health, anxiety, depression, and anger would probably be the most appropriate choices.

The core

In using the word ‘core’ to refer to the group of three negative affects, we do not imply anything that resides deep within the individ- ual. These subjective feelings are not neces- sarily hidden as part of an individual’s inner life. In fact, they are quite open to observa- tion and measurement. It is their ubiquity and functional significance that earns them mem- bership within the core of negative affect. This kind of pervasiveness and importance is also captured in the common adage that depression is the common cold of psychiatry, the notion in much of psychology that anxiety is inherent in neuroses if not in our very exis- tence as humans, and the vast and recurrent media coverage of acts of anger and rage.

Evolutionary roots

The clinical syndromes of anxiety, depres- sion, and anger are rooted in fear, sadness, and anger, respectively. These three discrete emotions play a primordial and universal role in the defense against aversive stimuli. Fear, for instance, is regarded as the most basic of all emotions because it motivates escape or avoidance from predators or other insur- mountable threats, thereby being crucial for survival. As Marks puts it ‘Fear is a vital evolutionary legacy ... Without fear, few

660 THE SAGE HANDBOOK OF PERSONALITY THEORY AND ASSESSMENT

9781412946513-Ch32 5/12/08 3:25 PM Page 660

would survive long under natural conditions’ (1987: 3). Anger is a twin emotion of fear in the defense against aversive stimuli. Thus Walter Cannon (1929) coined the term ‘fight or flight’ to refer to the twin options of fleeing out of fear or fighting out of anger during an emergency. Inasmuch as anger mobilizes the organism to retaliate in the face of provocation or assault, it promotes survival.

Surprisingly omitted from evolutionary accounts of basic emotion is sadness. This emotion may be viewed as a third option in the repertoire of responses toward threat or attack. When escape is not possible, when retaliation is not feasible, and the prospect of defeat is looming, then sadness is the emo- tion that arises in the service of the next most appropriate response of yielding or submission. A variant of it is what Seligman and colleagues term ‘learned helplessness’ (Peterson et al., 1993; Seligman, 1972). Buerki and Adler simply call it ‘giving up’ in order to conserve resources:

If a person has experienced certain situations, in which fight or flight was impossible or of no avail, he or she might react with conservation–with- drawal when exposed anew ... Conservation–with- drawal is primarily a biological reaction pattern, the counterpart of Cannon’s fight–flight reaction. Both reaction patterns are directed toward adapta- tion to stressful situations. They are aimed at self- protection and self-preservation. Fight–flight attempts to reach its goal by engaging, conserva- tion–withdrawal by disengaging and saving of energy. (2005: 5–6)

In the face of an overwhelming offensive, fighting would be a waste of resources if not an acceleration toward death. Similarly, when fighting or fleeing are not viable options in the face of overwhelming adver- sity, the emotion is likely to be sorrow and dejection which primes the individual to yield or surrender.

Physiological mechanisms

It has been portrayed that certain emotions have biochemical commonalities such as hormones and neurotransmitters (serotonin,

dopamine), and involve the same brain struc- tures. The evidence for this has been highly conflicting and no attempt will be made to review these findings here. Besides, it is not necessary to show biochemical specificity to justify the existence of different emotions or to show biochemical commonality to argue for the similarity of emotions.

What is relatively clear is that anger and fear involve the hypothalamic–pituitary axis in order to mobilize the organism toward vig- orous action of fight or flight. However, sym- pathetic reactivity is not only the result of negative affect but can be even greater during positive affect (Heponiemi et al., 2006). Also, depression is the one component of negative affect that is least likely to involve sympathetic activation, and that makes sense because the goal in depression is not one of action as much as inaction.

Recently, Ryff et al. (2006) found that anxiety and anger had more in common with regard to biological correlates. Women with an average age of 74 years old completed psychometric tests of anxiety, depression, and anger in addition to providing urine and blood samples on multiple occasions. It was found that traits of anxiety were negatively associated with systolic blood pressure (SBP) and positively associated with glyco- sylated hemoglobin. Traits of anger were inversely correlated with SBP and positively associated with glycosylated hemoglobin. Depression did not have any significant asso- ciations with the above biological correlates but was positively associated with weight.

AFFECTIVE FORM

As pointed out earlier, research has resound- ingly demonstrated that affect can be charac- terized in terms of valence and intensity. In other words, it can be distinguished qualita- tively as well as quantitatively. Being high in affective arousal says nothing about whether the person is elated or upset, just as being low in emotional arousal leaves open the

ANXIETY, DEPRESSION, AND ANGER: CORE COMPONENTS OF NEGATIVE AFFECT 661

9781412946513-Ch32 5/12/08 3:25 PM Page 661

possibility that the person may be gloomy or just glad. In addition to valence and intensity, affect can also be described in terms of form. By this we mean that affect (which we intro- duced as a general term) can assume differ- ent configurations depending on its patterns of occurrence.

State versus trait

One binary distinction, now popular in psy- chology, is between affect as a momentary state versus affect as an enduring trait. Thus, the anger a person experiences can be qualified in terms of whether it is a passing event or a habitual occurrence. Most of the effort in making this distinction is credited to Spielberger and colleagues who first pub- lished the state–trait anxiety scale (STAI; Spielberger et al., 1977), then the state– trait anger expression inventory (STAXI; Spielberger, 1988), and more recently, the state–trait depression scale (STDS; Krohne et al., 2002). In doing so, they have proposed that affective quality be distinguished according to whether it is a state happening ‘right now’ or a trait that is present ‘most of the time’. This mirrors the dichotomy between situational and dispositional aspects of behavior that have been the subject of much discussion by personality theorists and behaviorists.

Emotion, mood, temperament

The state–trait dichotomy was certainly an advancement upon vernacular labels for affect, and it soon caught on as a practice in psychological research to describe both state and trait when assessing anxiety, anger, or depression. However, the state–trait instru- ments are limited by some ambiguities (Fernandez, 2002). Asking subjects to report how they feel ‘right now’ still leaves unclear the distinction between emotion and mood, both of which may be present at a point in time and hence get subsumed under ‘state’.

Similarly, asking how a person ‘feels gener- ally’ may elicit answers that could pertain to either mood or trait because both mood and trait share the property of taking up more time. Clearly, the domain that is most obscured by the state–trait distinction of affect is mood.

A further improvement would be to refine the dichotomy into a trichotomy which allows for any affective quality to assume the form of an emotional episode, a mood state, or a temperamental trend (Table 32.1). The first of these three forms represents a rela- tively sharp and short-lived change in affec- tive intensity, the second represents a medium-term duration of affect, and the third represents the recurrent frequency of a par- ticular affect. These in turn correspond to the phasic, tonic, and cyclic properties of all affect. Emotion occurs as an episode and is therefore phasic, mood persists and is there- fore tonic, and temperament is the recurrence of a particular emotion and therefore has a cyclic quality.

These three different forms of any affec- tive quality are sometimes reflected by the semantic variations within many languages. In English for example, when a person becomes angry, that condition may be labeled anger or fury; when the anger persists for an extended time, the person may be said to be in a ‘crabby’ or irritable’ mood, whereas one who is habitually angry may be deemed a hostile or fractious person (Table 32.1). Language, however, turns out to be a crude instrument for labeling affect because of numerous individual differences in word

662 THE SAGE HANDBOOK OF PERSONALITY THEORY AND ASSESSMENT

Table 32.1 Emotion, mood, and temperament forms of affect

Affective form Affective quality Emotion Mood Temperament Fear Afraid Anxious Nervous Anger Angry Irritable/ Hostile

irascible Sadness Sad Depressive/ Melancholic

dysthymic

9781412946513-Ch32 5/12/08 3:25 PM Page 662

usage and the fact that any single language has its fair share of gaps and redundancies in labeling phenomena.

CORE OF NEGATIVE AFFECT (CONA) IN MEDICAL POPULATIONS

In our present review of the research on anger, fear, and sadness in medical popula- tions, it was not always possible to clearly delineate what was emotion from what was mood-related, or temperament but we do regard this tripartite form of affect as a nec- essary frame of reference for future research in this field. Another obstacle to firm conclu- sions in this endeavor was the uncertainty of the exact role or influence played by each affective type within each medical condition. As in the context of pain, affect could be a precipitant, a predisposing factor, an aggra- vator, a perpetuating factor, a consequence, or just a correlate (Fernandez, 2002). With regard to the last of these, it would also help to know if we are referring to co- occurrence, covariance, or equivalence between two variables. This is another pro- posed extension of our methodological approach to studying affect in illness, even though past literature may not lend itself to such a level of discrimination.

Surveying the last five years of published research, we set out to find studies of CVD, cancer, and HIV/AIDS in which all three core components of negative affect had been investigated. The product was a handful of studies quite divergent in terms of their design and their hypotheses. Nevertheless, these studies mark the beginnings of a new line of enquiry into the CONA and they are therefore the subject of review in the accom- panying section.

Cardiovascular disease (CVD)

In an extensive narrative review of CONA in coronary heart disease (CHD), Suls and

Bunde (2005) reported (1) evidence for depression in the development (precipitation) of CHD; (2) some evidence for depression leading to disease progression (exacerbation) in CHD; (3) evidence for anxiety in the development of CHD; (4) meager evidence for anxiety in the progression of CHD; (5) some evidence for hostility in the deve- lopment of CHD; and (6) minimal evidence of hostility in the progression of CHD. This means that anxiety, depression, and anger are primarily precipitants rather than aggravators of CHD. This is only in partial agreement with the findings on pain, where anxiety is a definitely a precipitator, depression is largely a consequence, and anger is at least a corre- late of pain. Suls and Bunde do not comment on the relative or collective effects of the triad of emotions on CHD because of insuffi- cient research on all three affective qualities within the same samples.

Mixed results in the review by Suls and Bunde may be due to methodologically diverse studies – especially the use of different meas- ures of affect across studies. Also, Suls and Bunde relied on significance levels rather than effect sizes to reach their inferences. Their interpretations that anxiety and depression (but not anger) are related to increased CHD risk in healthy samples may be re-evaluated on close inspection of their data as summarized in Table 32.2. As shown in the table, the actual percentage of studies reporting significant rela- tionships between affect and development of CHD never deviated far from chance levels nor did it differ appreciably across the three affec- tive types: 53% for depression, 42% for anxi- ety, and 48% for anger (Table 32.2). The role of depression as an aggravator of existing CHD is unclear due to what the authors identified as negative significant effects. Unfortunately, the exact number of negative significant effects was not specified. Other than that, the percentage of studies reporting significant aggravation of CHD by affect is remarkably similar: 29% for anxiety, and 27% for anger. Despite these findings, the role of anger (relative to its CONA counterparts) is seemingly understated in the etiology of CHD.

ANXIETY, DEPRESSION, AND ANGER: CORE COMPONENTS OF NEGATIVE AFFECT 663

9781412946513-Ch32 5/12/08 3:25 PM Page 663

It must also be pointed out that Suls and Bunde used measures of anger expression as predictors of CHD prognosis, when alterna- tively anger inhibition has also been impli- cated in CHD (Brosschot and Thayer, 1998; Magai et al., 2003; Smith and MacKenzie, 2006). The suppression or internalization of anger may demand greater cognitive effort and involve vagal mechanisms that increase the risk of cardiovascular deterioration. Given that many of the studies reported used the STAXI to assess anger, a distinction could have been made between internalized and externalized anger.

A subsequent study by Kubzansky et al. (2006) appeared in response to the limitations of previous research in which anxiety, depres- sion, and anger had been measured either singly or else as parts of a broader construct. The authors proposed a measure of general distress common to anxiety, depression, and anger in addition to orthogonal measures that were termed ‘iso-anxiety’, ‘iso-depression’, and ‘iso-anger’, respectively. They turned to the MMPI-2 which has 72 items that make up three content scales for measuring anxiety, depression, and anger, respectively. Responses to these 72 items were extracted from a

sample of 1,306 men who had completed the MMPI-2, and these data were subsequently analyzed using principal factor analysis with orthogonal varimax rotation. Based on this, three near-orthogonal scales were created for measuring the three corresponding affective types. Additionally, a fourth ‘general distress’ scale was constructed to include items that loaded equally strongly on more than one factor. The same sample of men was followed up for an average of 11 years at which point the MMPI-2 was re-administered. Data were analyzed in terms of multivariate-adjusted rel- ative risks of CHD for those highest versus lowest on each of the scales. Results showed a strong association between general distress and the incidence of CHD. Iso-anxiety was significantly associated with CHD outcomes, especially for myocardial infarction; iso-anger was associated primarily with angina pectoris; and iso-depression was not significantly asso- ciated with any CHD outcome. The authors concluded that their results call for an appreci- ation of the shared as well as unique contribu- tions of negative emotions in the development of CHD.

It should also be noted that independent investigations have shown that acute outbursts

664 THE SAGE HANDBOOK OF PERSONALITY THEORY AND ASSESSMENT

Table 32.2 Number of studies showing affective influences on coronary heart disease (CHD), based on Suls and Bunde (2005)

Significance of effect Marginal or select

Direction of effects Significant significant Non-significance Total Depression → CHD 10 7 2 19 Depression ↑ CHD 24 5 15 44 Anxiety → CHD 5 3 3 12 or 11 Anxiety ↑ CHD 4 1 9 14 Anger → CHD

Cynical hostility 5 2 4 11 Trait anger 1 1 1 3 Anger expression 5 2 2 9 ∑ 11 5 7 23

Anger ↑ CHD Cynical hostility 1 0 5 6 Trait anger 1 1 1 3 Anger expression 2 0 4 6 ∑ 4 1 10 15

→ Precipitating factor ↑ Exacerbating factor

9781412946513-Ch32 5/12/08 3:25 PM Page 664

of anger, fear, and sadness can trigger heart attacks (Carroll et al., 2002; Kamarck and Jennings, 1991; Lear and Kloner, 1996; Mittleman et al., 1995). However, cardiovas- cular reactivity is not only the result of nega- tive affect and can be even greater during intensely positive affect (Heponiemi et al. 2006). By implication, it is the sudden inten- sification of arousal during emotion that seems to be a precipitating factor in cardiac incidents. In the long term, anger, depres- sion, and anxiety may also encourage other unhealthy behaviors (e.g. smoking) that increase the risk of CHD (Smith and Ruiz, 2002).

The role of multiple affective qualities in cardiac incidents is also reflected in the rela- tively new construct called vital exhaustion (VE). As conceptualized by its originator, this includes irritability, demoralization, and fatigue (Appels, 1990; Appels and Mulder, 1988a, 1988b). Here, elements of anger and sadness are combined with fatigue. The anger seems to be internalized rather than externalized in people with this condition (Bages et al., 1999). VE seems to overlap partially with the type A personality which is characterized as a pattern of hostility, impa- tience, and competitiveness (Rosenman et al., 1975). It is quite possible that the fatigue and depression of VE may actually be a byproduct of (prolonged) type A-related behavior. In terms of life events, sustained job stress/conflict, unemployment, and bereavement have been known to culminate in VE (Falger and Schouten, 1992).

Whatever its bases, VE was initially regarded as a precipitator of myocardial infarction (Appels, 1990). It has also been shown to be associated with angina pectoris (Appels and Mulder, 1988a) and cardiac events following angioplasty (Kop, 1995). It is not a stretch to find the depressive and anergic elements of VE following serious cardiac incidents.

It bears mentioning that both the type A and VE constructs have had their share of mixed results in their relationship to CVD (e.g. Miller et al., 1991). This is not surprising

given the curious admixture of somatic, affective, and behavioral features within these constructs. Nonetheless, what is common to both constructs is a role of affect, even though VE emphasizes depression and type A emphasizes anger. Yet other psycho- logical investigations have revealed a part played by anxiety in CVD (e.g. Barger and Sydeman, 2005; Herrmann-Lingen and Buss, 2007). In sum, it pays to go in search of all three of these core affective qualities, keep- ing in mind that each may enter the picture through a different pathway, namely as precipitator, exacerbator, consequence, or perpetuator of CVD.

Cancer

Almost opposite to the anger-prone type A personality that is implicated in CHD, a personality prone to repressing negative emo- tions was articulated (Temoshok, 1987). Such non-expression of negative affect was sus- pected as a factor in the etiology of cancer. It came to be known as the type C personality.

Lieberman and Goldstein (2006) therefore investigated whether the ventilation of anxi- ety, depression, fear, and anger would have an impact on depression and quality of life in patients already diagnosed with breast cancer. The patients engaged in emotional expression through the medium of Internet bulletin boards for a period of about six months. The use of negative emotional words in each of the affective categories was exam- ined in relation to the dependent measures. Regression analyses revealed that anger expression was associated with improved quality of life and reduced depression, thus hinting at the psychodynamic notion of depression as anger turned inward. However, the expression of anxiety or fear was associ- ated with increased depression and reduced quality of life. The expression of sadness was not significantly related to the outcome measures. While these findings by no means show that suppression of negative affect causes cancer, they encourage the view that

ANXIETY, DEPRESSION, AND ANGER: CORE COMPONENTS OF NEGATIVE AFFECT 665

9781412946513-Ch32 5/12/08 3:25 PM Page 665

unexpressed anger is related to psychosocial impairment in breast cancer.

More extensive coverage of the research in this area was achieved in a meta-analysis by McKenna et al. (1999). Aggregating effect sizes across 46 studies, they found only a modest relationship between the presence of anxiety, depression, or anger (or their equiva- lent temperaments) and the development of breast cancer. The average effect sizes did not exceed 0.38 even when some of the depend- ent measures were combined into a broader construct of emotional denial/ repression.

A recent prospective study by Tijhuis et al. (2000) attempted to find out cancer incidence and mortality as a function of emotional con- trol of anxiety, depression, and anger. Almost a thousand men born between 1900 and 1920 and living in Zutphen, Netherlands were examined medically for cancer and also interviewed and assessed using the Courtald Emotional Control Scale (CECS) (Watson and Greer, 1983) in 1985, 1990, 1993, and 1995. Focusing on a final sample of 590 men, it was found that from 1985 to 1995, 119 of them were diagnosed with cancer and 71 died of cancer. Descriptive statistics for the sample revealed the highest level of emo- tional control for anxiety (19.2), a slightly lower level for depression (18.4) and a slightly lower level for anger (16.4) with almost equivalent degrees of variability. When Cox proportional hazards models were used to determine effects of emotional con- trol on cancer incidence and mortality, it was found that men within the highest and inter- mediate tertiles of controlled depression had a significantly increased risk of cancer mor- tality even after adjustment for other risk fac- tors such as age, marital status, and SES; this was not the case for men who suppressed anxiety or anger. Control of depression was also significantly related to cancer incidence, but anger control or anxiety control were not. This study is nevertheless informative because it shows that cancer patients are con- sumed not only by the somatic demands of their disease but also by a struggle to control anxiety, anger, and depression even though

only one of these (when controlled) seems to increase the incidence and mortality associated with cancer.

Also using the CECS, an Australian study on breast cancer failed to find any significant associations between cancer outcome and emotional suppression of any kind, before or after controlling for age effects (O’Donnell et al., 2000). Once again, the more important message for our purposes is that the cancer patients did seem to experience components of the core of negative affect, as implied by their scores on emotional control for each of these.

A qualitative illustration of core compo- nents of negative affect in cancer patients is visible in some of the nursing literature. For example, Bowers et al. (2002) mention that even though many women with cervical cancer were depleted of physical energy, they would utter statements such as:

That was once in my mind I was angry. I wanted to get in and get over with as soon as possible and not wait a month. Then I was too weak and tired to display it very much ... One day I came home and went to bed. The longer I laid in bed the madder I got. I was screaming to myself. I thought I would call a friend, but I did not want to dump on her so I said I can’t take it anymore and I came downstairs and banged and slammed and got supper. (2002: 144–145)

A further quote by Bowers et al. captures the almost existential anxiety of the cancer patient: ‘There is a reason for everything. I don’t know it is. I don’t know why. I began to think. Get a grip on yourself and find a pur- pose’ (2002: 139). A final quote by Bowers et al. captures the despair/depression of the cancer patient. ‘I guess my life was interest- ing, with so many things, and now it is not. Life is destroyed. It was so good before’ (2002: 145). These anecdotes help remind us of the cognitive appraisals that underlie the statistical data on anxiety, depression, and anger in medical populations.

HIV/AIDS

One of the few recent empirical studies of the CONA in the context of HIV/AIDS was

666 THE SAGE HANDBOOK OF PERSONALITY THEORY AND ASSESSMENT

9781412946513-Ch32 5/12/08 3:25 PM Page 666

conducted by Atwine et al. (2005). In a rural district of southwestern Uganda, 123 chil- dren aged 11–15 years old whose parents had reportedly died from AIDS were compared with a normative sample of 110 children of similar age and gender living in intact house- holds. They were all administered an appro- priately translated version of the Beck Youth Inventories (Beck et al., 2001) which had been designed as a diagnostic aid for anxiety, depression, and anger, and for self-concept problems and disruptive behavior in youth. Results revealed significantly higher levels of disruptiveness as well as all three compo- nents of CONA in the orphaned group as compared to the non-orphaned group.

Another group of researchers (Teva et al., 2005) studied 100 HIV/AIDS patients between 18 and 70 years old who were recruited from various hospitals in Andalusia, Spain. They were administered a battery of tests suitable for assessing the CONA: the BDI, the STAI, and the STAXI. It was found that most of the 63 men and 37 women in the group reported low levels of state anger, with about one-third not expressing anger. This may be related to the additional finding that most participants were low in trait anger to begin with. Anger was higher during the symptomatic stage as opposed to the pre- symptomatic stage. Similarly, anxiety was greater during lypodystrophy than before it. Anxiety was far more prevalent in men than women. Most men also showed some depres- sion but most women did not. The authors explain these differences with reference to cognitive appraisals that differ according to gender and stage of infection.

To the extent that pain is often a symptom in HIV/AIDS, the kinds of affective distress observed in chronic pain patients are also likely to manifest in HIV/AIDS patients (Marcus et al., 2000). Morever, HIV/AIDS patients, like cancer patients, often go through stages of adjustment to this (presumably) ter- minal illness. In the traditional model of Elizebeth Kubler-Ross (1974, 1997), this begins with shock and anxiety, proceeds to anger, and ends in depression. The core

components of negative affect may thus unfold in sequence rather than appear concurrently.

Pain

Over the last half of the twentieth century, considerable research accumulated on anxi- ety, depression, and (to a lesser extent) anger in pain patients. This has already been cri- tiqued and synthesized (e.g. Banks and Kerns, 1996; Fernandez, 2002). We now turn to a few recent empirical articles on the core of negative affect in pain, followed by a dis- cussion of how this parallels the experience of negative affect associated with cancer, HIV/AIDS, or CVD.

Feeney (2004) evaluated 100 post-surgical orthopedic patients above the age of 65. These individuals were administered the geriatric depression scale (stripped of its somatic items because these do not discrimi- nate between depressed elderly and non- depressed elderly). Patients were also administered the STAI and the STAXI to generate state and trait measures of anxiety and anger, respectively. The McGill Pain Questionnaire (MPQ) (Melzack, 1975) was used to derive a total pain score by summing the rank values of pain descriptors endorsed by patients. The authors found that pain was significantly correlated with state anxiety and depression but not with any of the meas- ures of anger. Standard multiple linear regression analysis revealed that about 31% of the variance in total pain was explained by the five affective variables but only state anx- iety had a significant standardized weight, accounting for about 18% of the variance in pain. The remaining four variables did not contribute significantly to the prediction of pain over and above that accounted for by state anxiety. This is quite likely due to the particular pain measure that was chosen. In using the rank values of pain descriptors from the MPQ, the authors were opting for a crude index of pain in comparison to the scales of the multidimensional pain inventory (Kerns et al., 1985) or even other measures

ANXIETY, DEPRESSION, AND ANGER: CORE COMPONENTS OF NEGATIVE AFFECT 667

9781412946513-Ch32 5/12/08 3:25 PM Page 667

offered by the MPQ itself such as the present pain intensity. The MPQ also allows meas- urement of affective pain as a separate factor, and examining this variable would probably have led to more significant results beyond those witnessed for anxiety.

Ghosh and Puja (2004) administered the BDI, STAI, and STAXI to 50 female outpa- tients with migraine headache and an equally sized group of age-matched females with no headaches. T-tests showed significantly higher scores for the patient group on six measures (trait anxiety, trait anger, anger-in, anger-out, anger control, and depression). The significant differences on trait rather than state anger and anxiety raise the likeli- hood that patients’ headaches were not pre- cipitated by affective episodes but were predisposed by affective temperaments. This conclusion is consistent with the findings of several studies cited by the authors.

In a more specific investigation of anger expression styles as they relate to pain, Kerns et al. (1994) found that anger-in and anger expression are correlated with chronic low back pain severity, though the former is associ- ated with poorer adjustment. Similarly, Bruehl et al. (2002) found that both anger-in and anger-out affected pain sensitivity, but only the latter seemed to be mediated by impairment in antinociceptive effects of endogenous opioids.

In a broader investigation of the inhibition and expression of multiple emotions, Burns et al. (2003) randomly assigned students to three conditions: anger, sadness, and joy, respectively. In each condition, subjects recalled and described a recent event that evoked the relevant emotion. This was accompanied by a cold pressor pain test. Pain response was assessed by temporal measures of threshold and tolerance as well as by verbal descriptors on the MPQ. Unlike other findings by the same authors, a significant positive association was found between anger-out and pain threshold (but not pain tol- erance or MPQ scores); this effect was paral- leled by decreases in systolic blood pressure but not diastolic blood pressure. In contrast, induced sadness led to the largest increases

in MPQ scores of pain severity. It would be interesting to extend this line of enquiry by investigating any changes in pain sensitivity that might occur when fear is evoked using the same recall procedure as used for anger and sadness.

Our understanding of the emotions experi- enced by those in pain can be further deepened by an exploration of how exactly their pain is interpreted. As pointed out earlier, beneath the statistical data on negative affect are undercur- rents of cognitive appraisals about the medical condition. Thus, pain patients are less likely to be angry at the pain itself and more likely to be angry at the ramifications of their painful con- dition (Fernandez and Turk, 1995). Similarly, anger is to be expected in any disease or disor- der that is diagnostically ambiguous, refractory to treatment, (mis)attributed to psychological mechanisms, financially burdensome, and legally fractious. Consider the emotional reac- tions that arise in cancer patients. As Bowers, Tamlyn, and Butler mention,

Most women experienced anger not at the cancer itself, but rather in relation to the communication and contact with others as they lived with ovarian cancer. In general, the causes of women’s anger were related to misdiagnosis, late diagnosis, multi- ple testing, physicians discounting their symptoms and/or waiting for treatment, or inaccessibility to prompt treatment. (2002: 144)

Just as chronic pain can generate life inter- ference which culminates in depression (Rudy et al., 1988), so can cancer, CVD or HIV/AIDS become depressing via their lim- iting effects on day-to-day functioning. The process of functional decline can be met with considerable apprehension, worry, and out- right dread, especially if death is imminent. It is therefore not far-fetched to also consider the existential anxieties that are probably added to the other objects of anxiety, depres- sion, and anger in these patients.

CONA COMORBIDITY

The preceding literature review shows that anxiety, depression, and anger do exist

668 THE SAGE HANDBOOK OF PERSONALITY THEORY AND ASSESSMENT

9781412946513-Ch32 5/12/08 3:25 PM Page 668

(sometimes in isolation, sometimes in com- bination) to a clinically significant degree in patients with medical ailments. The next issue concerns the extent to which the com- ponents of CONA are comorbid with one another, and what corollaries arise thereby. Across studies of medical patients, there have been repeated observations of a close relationship between anxiety and depression. The association between anger and each of its two counterparts in the CONA has been less researched. Given that anger and fear are twin emotions that mobilize the individual to fight or flight, it is likely that in any set of nomothetic data from individuals facing provocation or danger, there would be traces of both anger and fear. Up to this time, however, the emphasis has been on the so- called comorbidity between anxiety and depression.

Comorbidity statistics

Comorbidity, at its simplest, refers to the co- occurrence of two disorders in the same indi- vidual. However, it makes a difference whether the individual is evaluated for episode comorbidity or lifetime comorbidity. The former refers to multidiagnostic co- occurrence at one point in time. This is likely to be exceeded by the latter which means multiple diagnoses occurring at any point in the individual’s lifetime. Based on an exten- sive epidemiological study, Robins et al. (1991) reported a 60% lifetime comorbidity of psychiatric disorders. About one-third of patients diagnosed with anxiety disorder were also diagnosed with a depressive disorder (Sanderson et al. 1990).

Going beyond co-occurrence to correla- tion, the picture remains similar. Anxiety and depression have been repeatedly shown to co-vary in a positive direction. Dobson’s (1985) review of the relevant literature showed that the correlation between scores on anxiety and scores on depression ranged from +0.27 to +0.94, with an average corre- lation of +0.61. This average correlation was

only a little less than the average correlation of +0.66 between anxiety scales.

Principally, there are five main explana- tions for comorbidity of the core components of negative affect: definitional overlap, instrument overlap, response set, misinter- pretation of data, and phenomenological bases. It is necessary to evaluate the tenabil- ity of each of these explanations as they have implications for the theoretical and applied potential of CONA as a construct.

Definitional overlap

The definitional overlap pertains to a similar- ity of conceptualization, in this case, between multiple diagnostic labels. Specifically, if there are similarities in the way anxiety and depression are operationally defined, it would not be surprising that when one is identified, so is the other. For the definition of clinically significant depression and anxi- ety, we turn to the Diagnostic and Statistical Manual of Mental Disorders, version IV-Text Revision (DSM-IV-TR). In this nosological system, the dysthymic variant of depressive disorder comprises at least two years of at least two of the following symptoms: poor appetite, sleep disturbance, fatigue, low self- esteem, poor concentration, and perceived hopelessness. By comparison, generalized anxiety disorder (GAD) comprises more than six months of worry/anxiety with at least three of the following symptoms: restless- ness, fatigue, difficulty concentrating, irritability, muscle tension, and sleep distur- bance. As immediately apparent, 50% of the symptoms of dysthymia are found in GAD and vice versa. This degree of overlap may account for the high comorbidty between these two disorders, as reported in the National Comorbidity Survey (NCS) of 8,000 respondents across the US (Kessler et al., 1994). In this study, the six-month comorbid- ity of GAD and dysthmia was quantified by an odds ratio of 21.5, odds ratio being the ratio of frequency of two disorders being simultaneously present or absent to the

ANXIETY, DEPRESSION, AND ANGER: CORE COMPONENTS OF NEGATIVE AFFECT 669

9781412946513-Ch32 5/12/08 3:25 PM Page 669

frequency of each one being present on its own – in other words, the ratio of a joint occurrence to a singular occurrence. The GAD–dysthymia odds ratio was among the highest for any pair of psychiatric disorders. Similarly high odds ratios have been reported for other pairs of anxiety and depressive dis- orders, in particular panic disorder and major depression, with an odds ratio of 21.3 in the Epidemiological Catchment Area (ECA) study of 20,000 respondents in five US com- munities (Robins et al., 1991). In fact, the average pairwise associations between affec- tive disorders (inclusive of mania) and anxi- ety disorders have been higher than that between anxiety disorders (Kessler, 1995). In short, the overlap of DSM diagnostic criteria may account for some of the comorbidity between clinical anxiety and depression.

Instrument overlap

The idea that high comorbidity between anx- iety and depression could be due to instru- ment overlap occurred to various scholars who noticed that many psychological tests discriminated poorly between the two affec- tive types (Clark et al., 1990). The same applies to the comorbidity of anger, depres- sion, and anxiety. An inspection of the items in the Beck Depression Inventory (BDI) (Beck and Steer, 1993a) and the Beck Anxiety Inventory (BAI) (Beck and Steer, 1993b) reveals similarity of content as does a comparison of the STAI and STAXI. Anxiety and anger are likely to share common physi- ological reactivity by virtue of their common roots in sympathetic activation. However, if psychological tests rely on the subjective feelings as defining features of these syn- dromes, then they are less likely to generate overlapping profiles.

Interpretation of data

Suls and Bunde (2005) adopt Watson’s view that the frequent correlation in self-report

data for anxiety and depression must be partly rooted in a common latent factor or shared underlying dimension called negative affect. Reacting to this multicollinearity, Ketterer (1996) and others have suggested that we replace the measurement of anxiety, depression, and anger as separate entities with a global measure of negative affect.

Certainly, multicollinearity between vari- ables (especially if it exceeds 0.80) suggests redundancy. But just because entities are cor- related does not mean that they are con- nected. It simply means they co-vary. Thus, the strong collinearity between height and weight is not grounds for collapsing the two into one construct. Culture is closely associ- ated with race, but it still makes much sense to tease the two apart. Verbal and quantitative IQ tend to be highly correlated, yet they are often viewed as distinct areas of ability.

A useful lesson in the interpretation of multicollinearity can be found in the context of measuring pain components. Turk et al. (1985) discovered that in a multiple-group confirmatory factor analysis, the sensory, affective, and evaluative subscales of the MPQ turned out to be highly intercorrelated: r = 0.81 between sensory and affective, r = 0.67 between affective and evaluative, and r = 0.64 between evaluative and sensory, thus yielding an average correlation of 0.71 among the three constructs. Moreover, the cross-construct correlations exceeded the within-construct correlations. The authors took this as a sign of lack of distinctiveness of the subscales and therefore recommended using the total factor score rather than individual scores on the three subscales. However, in a rebuttal, Melzack adduced several bits of evidence from perceptual psy- chophysics to show that a high correlation among variables is not a sign of redundancy and does not necessitate collapsing the vari- ables into one. Specifically, increases in light intensity are associated with enhanced dis- criminability of color, contours, texture, and distance, yet we do not suggest conflating color and texture into one variable. Similarly, in audition, increased volume enhances

670 THE SAGE HANDBOOK OF PERSONALITY THEORY AND ASSESSMENT

9781412946513-Ch32 5/12/08 3:25 PM Page 670

discrimination of pitch, timbre, and spatial location, but this is not grounds for abandon- ing separate measures of timbre and pitch (Melzack and Katz, 1992).

Response set

In an extensive and seminal review paper, Russell and Carroll (1999) strongly disputed the idea proposed by Watson et al. (1988) that positive and negative affect are inde- pendent unipolar dimensions. In the process, Russell and Carroll also offered empirical data and reasoned arguments that now allow us to seriously doubt the value of collapsing anxiety and depression (and anger for that matter) into an undifferentiated phenomenon called ‘negative affect’. Citing the classic work of Bentler (1969), it was pointed out that spurious correlations can emerge from self-report tests when there is an acquiescent response style in test-taking. Russell and Carroll then went on to cite about a dozen other studies containing empirical evidence of how this acquiescent response set has in fact influenced measures of affect. This may well account for the frequently observed cor- relations between self-report measures of anxiety and depression as well as of anger.

Phenomenological bases

Of course, anxiety, depression, and anger (or their corresponding emotions of fear, sad- ness, and anger) often co-occur, but this is not sufficient grounds for resorting or revert- ing to a general concept of ‘negative affect’. Some of the association is phenomenologi- cally based. First, at any point in time, each of the three emotions may be rooted in quite different events: a patient may be angry because of conflict on the job, depressed because of illness, and anxious about the welfare of family members. Second, the same things that make people depressed can also make them anxious and angry. Failure in a task/test often leaves one feeling sorry or

sad for oneself, angry at the person evaluat- ing one’s performance, and worried about the consequences for one’s goal attainment. Killing of an admired leader often leads to sorrow for the leader’s suffering or depriva- tion of rights, anger toward the killers, and apprehension about how to cope without the leader. Popularly called ‘mixed emotions’, these co-occur because of different appraisals of the same event. So, co-occurring emotions can be due to (1) different reactions to differ- ent event or (2) different reactions to the same event. It would not make sense to com- bine such multiple emotions into one amor- phous ‘negative affect’ because these emotions originate from quite different circumstances or else are differentiated by separate appraisals of the same event.

MEASURING CORE COMPONENTS OF NEGATIVE AFFECT

The current componential representation of negative affect is consistent with a major per- spective in affect science called differential emotions theory or the theory of discrete emotions. Accordingly, the tests used to assess the core components of negative affect should be selected to allow the differentiation of negative affect into its core components of fear, sadness, or anger, or their respective clinical equivalents of anxiety, depression, and anger. The options for assessing these types of affect would therefore exclude the positive and negative affect scales (PANAS) (Watson, et al., 1988) which are predicated on a view of undifferentiated negative affect. Moreover, the single word descriptors that make up the PANAS (distressed, upset, hos- tile, irritable, scared, afraid, ashamed, guilty, nervous, and jittery) are gross labels that are unsuited for accessing the underlying appraisals of each emotion. If anything, this is what may obscure some of the fine differ- ences among anxiety, depression, and anger or their emotional equivalents of fear, sad- ness, and anger. It should also be noted that

ANXIETY, DEPRESSION, AND ANGER: CORE COMPONENTS OF NEGATIVE AFFECT 671

9781412946513-Ch32 5/12/08 3:25 PM Page 671

the concept of negative affect as proposed by Watson et al. does not pertain to negative emotions as much as ‘subjective distress and unpleasurable engagement that subsumes a variety of aversive mood states, including anger, contempt, disgust, guilt, fear, and nervousness, with low NA being a state of calmness and serenity’ (1988: 1063).

The use of single-word adjectives for assessing affect dates back to the multiple affect adjective checklist (MAACL) (Zuckerman and Lubin, 1965) – revised as the MAACL-R (Zuckerman and Lubin, 1985). This instrument does generate scores for anxiety, depression, and hostility but its factor structure is still an unsettled matter (e.g. Gotlib and Meyer, 1986). Another instrument of the same genre is the profile of mood states (POMS) (McNair et al., 1981) which lists 65 adjectives of affect to be rated on a four-point scale of amount/frequency. Subscale scores are generated for all three components of the CONA in addition to three other subscales pertaining to energy levels and cognitive function. Psychometrically, it has received some support though ques- tions remain about how to interpret its results (e.g. Boyle, 1987).

The differential emotions scale (DES-IV; Izard et al., 1974) takes affect assessment a step deeper by replacing single word adjec- tives with actual statements that better reflect the experience of emotion. Subscale scores are generated for 11 types of affect, among them anger, fear, and sadness. There has been limited psychometric evaluation of the DES-IV although some of the empirical outcomes are encouraging (Boyle, 1986).

Apart from the above instruments directed specifically at affect, there are more general tests such as the SCL-90-R and the MMPI-2. Both of these are commonly used in health psychology to cast a wide net for detecting psychopathology. In the process, they allow the identification of clinically significant levels of the CONA. One special advantage of these tests is that their psychometric valid- ity and reliability have been the subject of extensive research and are now fairly well

established. However, they are broad in scope and therefore bring in more data than is needed for our current goals of assessing negative affect.

CONCLUSION

It is the thesis of this chapter that there are three key components to negative affect: fear, sadness, and anger, which can take the form of emotions, moods, or temperaments. Previous research has studied them mainly as discrete emotions or else as the clinical syn- dromes of anxiety, depression, and anger.

The three core components of negative affect have an evolutionary history that has earned them special roles in survival. In par- ticular, they are part of the individual’s reper- toire of defenses against threat, attack, or adversity in general. Thus, anxiety, depres- sion, and anger are prevalent in medical pop- ulations such as those afflicted with CVD, cancer, or HIV/AIDS. Research has pointed to the comorbidity of these affective types. The frequent co-occurrence or covariation of these affective types does not mean that they should be collapsed into one broad category called negative affect. Close scrutiny has revealed that the comorbidity is in part due to overlap in nosological criteria for anxiety and depression and in part due to overlapping items across psychological tests. The comor- bidity may also be an artifact of response sets. Most important, anger, fear, and sadness are linked by unique threads of cognitive appraisals in response to the same situation or else by multiple appraisals in response to multiple stimuli.

Future research may benefit greatly from the assessment of the three core components of negative affect in medical populations. This is not strictly tied to any premise that anxiety, depression, and anger co-occur, co- vary, or are equivalent. Rather, the prime reason is that there is a high probability of one or more of these affective types in anyone who faces adversity. Perhaps,

672 THE SAGE HANDBOOK OF PERSONALITY THEORY AND ASSESSMENT

9781412946513-Ch32 5/12/08 3:25 PM Page 672

by including three subscales on one and the same test of affect, scores for anxiety, depres- sion, and anger can be output on the same metric, yet on three separate but parallel con- tinua. This would enable the charting of a profile of the individual’s core components of negative affect. Protracted over time, such a chart might also reveal patterns that allow us to differentiate the emotional, mood- related, or temperamental aspects of anger, fear, and sadness. In this way, the landscape of a person’s affective function can be better mapped to identify, with greater specificity, the areas in need of clinical attention.

REFERENCES

American Psychiatric Association (2000) Diagnostic and Statistical Manual of Mental Disorders Fourth Edition Text Revision (DSM- IV-TR). Washington, DC: American Psychiatric Publishing.

Appels, A. (1990) ‘Mental precursors of myocardial infarction’, British Journal of Psychiatry, 156(2): 465–71.

Appels, A. and Mulder, P. (1988a) ‘Excess fatigue as a precursor of myocardial infarc- tion’, European Heart Journal, 9(7): 758–64.

Appels, A. and Mulder, P. (1988b) ‘A question- naire to assess premonitory symptoms of myocardial infarction’, International Journal of Cardiology, 17(1): 15–24.

Atwine, B., Cantor-Graae, E. and Bajunirwe, F. (2005) ‘Psychological distress among AIDS orphans in rural Uganda’, Social Science and Medicine, 61(3): 551–64.

Bages, N., Appels, A. and Falger, P.R.J. (1999) ‘Vital exhaustion as a risk factor of myocar- dial infarction: A case control study in Venezuela’, International Journal of Behavioral Medicine, 6(3): 279–90.

Banks, S.M. and Kerns, R.D. (1996) ‘Explaining high rates of depression in chronic pain: A diathesis-stress framework’, Psychological Bulletin, 119(1): 95–110.

Barger, S.D. and Sydeman, S.J. (2005) ‘Does generalized anxiety disorder predict coronary heart disease risk factors independently of major depressive disorder?’, Journal of Affective Disorders, 88(1): 87–91.

Barlow, D.H. (1991) ‘Disorders of emotion’, Psychological Inquiry, 2(1): 58–71.

Beck, A.T. and Steer, R.A. (1993a) Manual for the Beck Anxiety Inventory. San Antonio, TX: Psychological Corporation.

Beck, A.T. and Steer, R.A. (1993b) Manual for the Beck Depression Inventory. San Antonio, TX: Psychological Corporation.

Beck, J., Beck, A. and Jolly, J. (2001) Beck Youth Inventories of Emotional and Social Impairment Manual. San Antonio, TX: Psychological Corporation.

Bentler, P.M. (1969) ‘Semantic space is approx- imately bipolar’, Journal of Psychology, 71(1): 33–40.

Bowers, D.E., Tamlyn, D. and Butler, L.J. (2002) ‘Women living with ovarian cancer: Dealing with an early death’, Health Care for Women International, 23(2): 135–48.

Boyle, G.J. (1986) ‘Higher-order factors in the Differential Emotions Scale (DES-III)’, Personality and Individual Differences, 7(3): 305–10.

Boyle, G.J. (1987) ‘A cross-validation of the factor structure of the Profile of Mood States: Were the factors correctly identified in the first instance?’, Psychological Reports, 60(2): 343–54.

Brosschot, J.F. and Thayer, J.F. (1998) ‘Anger inhibition, cardiovascular recovery, and vagal function: A model of the link between hos- tility and cardiovascular disease’, Annals of Behavioral Medicine, 20(4): 326–32.

Bruehl, S., Burns, J., Chung, O.Y., Ward, P. and Johnson, B. (2002) ‘Anger and pain sensitiv- ity in chronic low back pain patients and pain-free controls: The role of endogenous opioids’, Pain, 99(1–2): 223–33.

Buerki, M.D. and Alder, M.D. (2005) ‘Negative affect states and cardiovascular disorders: a review and the proposal of a unifying biopsy- chosocial concept’, General Hospital Psychiatry, 27(3): 180–8.

Burns, J.W., Kubilus, A. and Bruehl, S. (2003) ‘Emotion induction moderates effects of anger management style on acute pain sen- sitivity’, Pain, 106(1–2): 109–18.

Cannon, W.B. (1929) Bodily Changes in Pain, Hunger, Fear, and Rage (2nd edn). New York: Appleton.

Carroll, D., Ebrahim, S., Tilling, K., Macleod, J. and Smith, G.D. (2002) ‘Admissions for myocardial infarction and World Cup

ANXIETY, DEPRESSION, AND ANGER: CORE COMPONENTS OF NEGATIVE AFFECT 673

9781412946513-Ch32 5/12/08 3:25 PM Page 673

football: database survey’, British Medical Journal, 325(7378): 1439–42.

Clark, D.A., Beck, A.T. and Stewart, B. (1990) ‘Cognitive specificity and positive-negative affectivity: Complementary or contradictory views on anxiety and depression?’, Journal of Abnormal Psychology, 99(2): 148–55.

Dobson, K.S. (1985) ‘The relationship between anxiety and depression’, Clinical Psychology Review, 5(4): 307–24.

Falger, P.R.J. and Schouten, E.G.W. (1992) ‘Exhaustion, psychological stressors in the work environment, and acute myocardial infarction in men’, Journal of Psychosomatic Research, 36(8): 777–86.

Feeney, S.L. (2004) ‘The relationship between pain and negative affect in older adults: anx- iety as a predictor of pain’, Journal of Anxiety Disorders, 18(6): 733–44.

Fernandez, E. (2002) Anxiety, Depression, and Anger in Pain: Research Findings and Clinical Options. Dallas, TX: Advanced Psychological Resources.

Fernandez, E., Clark, T.S. and Rudick-Davis, D. (1999) ‘A framework for conceptualization and assessment of affective disturbance in pain’, in A.R. Block, E.F. Kremer and E. Fernandez (eds), Handbook of Pain Syndromes: Biopsychosocial Perspectives. Mahwah, NJ: Erlbaum, pp. 123–47.

Fernandez, E. and Turk, D.C. (1995) ‘The scope and significance of anger in the experience of chronic pain’, Pain, 61(2): 165–75.

Ghosh, S.N. and Puja (2004) ‘The emotional vital signs in migraine’, Psychological Studies, 49(4): 272–7.

Gotlib, I.H. and Meyer, J.P. (1986) ‘Factor analysis of the Multiple Affect Adjective Check List: A separation of positive and negative affect’, Journal of Personality and Social Psychology, 50(6): 1161–5.

Heponiemi, T., Ravaja, N., Elovainio, M., Naatanen, P. and Jarvinen, L.K. (2006) ‘Experiencing positive affect and negative affect during stress: Relationships to cardiac reactivity and to facial expressions’, Scandinavian Journal of Psychology, 47(5): 327–37.

Herrmann-Lingen, C. and Buss, U. (2007) ‘Anxiety and depression in patients with coronary heart disease’, in J. Jordan, B. Bardé and A.M. Zeiher (eds), Contributions Toward Evidence-based Psychocardiology: A Systematic Review of the

Literature. Washington, DC: American Psychological Association, pp. 125–57.

Izard, C.E., Dougherty, F.E., Bloxom, B.M. and Kotsch, W.E. (1974) ‘The differential emotions scale: A method of measuring the subjective experience of discrete emotions’, Unpublished manuscript, Vanderbilt University.

Kamarck, T. and Jennings, J.R. (1991) ‘Biobehavioural factors in sudden cardiac death’, Psychological Bulletin, 109(1): 42–75.

Kerns, R.D., Rosenberg, R. and Jacob, M.C. (1994) ‘Anger expression and chronic pain’, Journal of Behavioral Medicine, 17(1): 57–67.

Kerns, R.D., Turk, D.C. and Rudy, T.E. (1985) ‘The West Haven-Yale Multidimensional Pain Inventory (WHYMPI)’, Pain, 23(4): 345–56.

Kessler, R.C., Mcgonagle, K.A., Zhao, S., Nelson, C.B., Hughes, M., Eshleman, S., Wittchen, H-U. and Kendler, K.S. (1994) ‘Lifetime and 12-month prevalence of DSM- III-R psychiatric disorders among persons aged 15–54 in the United States: Results from the National Comorbidity Survey’, Archives of General Psychiatry, 51(1): 8–19.

Kessler, R.C. (1995) ‘Epidemiology of psychi- atric comorbidity’, in M.T. Tsuang, M. Tohen and G.E.P. Zahner (eds), Textbook in Psychiatric Epidemiology. New York: Wiley, pp. 179–97.

Ketterer, M.W. (1996) ‘Anger and myocardial infarction [Letter]’, Circulation, 94(7): 1788–9.

Kop, W.J. (1995) ‘Vital exhaustion, cardiac events and the relationship to depression’, European Psychopharmacology, 5(3): 220–1.

Krohne, H.W., Schmukle, S.C., Spaderna, H. and Spielberger, C.D. (2002) ‘The State-Trait Depression Scales: An international compari- son’, Anxiety, Stress and Coping: An International Journal, 15(2): 105–22.

Kubler-Ross, E. (1974) Questions and Answers on Death and Dying. New York: Macmillan.

Kubler-Ross, E. (1997) AIDS: The Ultimate Challenge. New York: Touchstone.

Kubzansky, L.D., Cole, S.R., Kawachi, I., Vokonas, P. and Sparrow, D. (2006) ‘Shared and unique contributions of anger, anxiety, and depression to coronary heart disease: A prospective study in the normative aging study’, Annals of Behavioral Medicine, 31(1): 21–9.

Lear, J. and Kloner, R.A. (1996) ‘The Northridge earthquake as a trigger for acute myocardial infarction’, American Journal of Cardiology, 77(14): 1230–2.

674 THE SAGE HANDBOOK OF PERSONALITY THEORY AND ASSESSMENT

9781412946513-Ch32 5/12/08 3:25 PM Page 674

Lieberman, M.A. and Goldstein, B.A. (2006) ‘Not all negative emotions are equal: The role of emotional expression in online support groups for women with breast cancer’, Psycho-Oncology, 15(2): 160–8.

Magai, C., Kerns, M.D., Gillespie, M. and Huang, B. (2003) ‘Anger experience and anger inhibition in sub-populations of African American and European American older adults and relation to circulatory disease’, Journal of Health Psychology, 8(4): 413–32.

Marcus, K.S., Kerns, R.D., Rosenfeld, B. and Breitbart, W. (2000) ‘HIV/AIDS-related pain as a chronic pain condition: Implications of a biopsychosocial model for comprehensive assessment and effective management’, Pain Medicine, 1(3): 260–73.

Marks, I. (1987) Fears, Phobias, and Rituals: Panic, Anxiety, and their Disorders. New York: Oxford University Press.

McKenna, M.C., Zevon, M.A., Corn, B. and Rounds, J. (1999) ‘Psychosocial factors and the development of breast cancer: A meta- analysis’, Health Psychology, 18(5): 520–31.

McNair, D.M., Lorr, M. and Droppleman, L.F. (1981) Profile of Mood States. San Diego: Educational and Industrial Testing Service.

Melzack, R. (1975) ‘The McGill Pain Questionnaire: Major properties and scoring methods’, Pain, 1(3): 277–259.

Melzack, R. and Katz, J. (1992) ‘The McGill Pain Questionnaire: Appraisal and current status’, in D.C. Turk and R. Melzack (eds), Handbook of Pain Assessment. New York: Guilford Press, pp. 152–68.

Miller, T.Q., Turner, C.W., Tindale, R.S., Posavac, E.J. and Dugoni, B.L. (1991) ‘Reasons for the trend toward null findings in research on Type A behavior’, Psychological Bulletin, 110(3): 469–85.

Mittleman, M.A., Maclure, M., Sherwood, J.B., Mulry, R.P., Tofler, G.H., Jacobs, S.C., Friedman, R., Benson, H. and Muller, J.E. (1995) ‘Triggering of acute myocardial infarction onset by episodes of anger’, Circulation, 92(7): 1720–5.

O’Donnell, M.C., Fisher, R., Irvine, K., Rickard, M. and McConaghy, N. (2000) ‘Emotional suppression: Can it predict cancer outcome in women with suspicious screening’, Psychological Medicine, 30(5): 1079–88.

Peterson, C., Maier, S.F. and Seligman, M.E.P. (1993) Learned Helplessness: A Theory for

the Age of Personal Control. New York: Oxford University Press.

Robins, L.N., Locke, B.Z. and Reiger, D.A. (1991) ‘An overview of psychiatric disorders in America’, in L.N. Robins and D.A. Reiger (eds), Psychiatric Disorders in America: The Epidemiological Catchment Area Study. New York: The Free Press, pp. 328–66.

Rosenman, R.H., Brand, R.J., Jenkins, C.D., Friedman, M., Straus, R. and Wurm, M. (1975) ‘Coronary heart disease in the Western Collaborative Group Study: Final follow-up experience of 8 1/2 years’, Journal of the American Medical Association, 233(8): 872–7.

Rudy, T.E., Kerns, R.D. and Turk, D.C. (1988) ‘Chronic pain and depression: Toward a cog- nitive-behavioral mediation model’, Pain, 35(2): 129–40.

Russell, J.A. and Carroll, J.M. (1999) ‘On the bipolarity of positive and negative affect’, Psychological Bulletin, 125(1): 3–30.

Ryff, C.D., Love, G.D., Urry, H.L., Muller, D., Rosenkranz, M.A., Friedman, E.M., Davidson, R.J. and Singer, B. (2006) ‘Psychological well-being and ill-being: Do they have distinct or mirrored biological correlates?’, Psychotherapy and Psychosomatics, 75(2): 85–95.

Sanderson, W.C., DiNardo, Rapee, R.M. and Barlow, D.H. (1990) ‘Syndrome comorbidity in patiens diagnosed with a DSM-III-R anxiety disorder’, Journal of Abnormal Psychology, 99(3): 308–12.

Seligman, M.E. (1972) ‘Learned helplessness’, Annual Review of Medicine, 23(5): 407–12.

Smith, T.W. and MacKenzie, J. (2006) ‘Personality and risk of physical illness’, Annual Review of Clinical Psychology, 2: 435–67.

Smith, T.W. and Ruiz, J. (2002) ‘Psychosocial influences on the development and course of coronary heart disease: Current status and implications for research and practice’, Journal of Consulting and Clinical Psychology, 70(3): 548–68.

Spielberger, C.D. (1988) State-Trait Anger Expression Inventory Professional Manual. Odessa, FL: Psychological Assessment Resources.

Spielberger, C.D., Gorsuch, R.L., Lushene, R.E., Vagg, P.R. and Jacobs, G.A. (1977) ‘The State–Trait Anxiety Inventory: Test Manual for Form Y’, Palo Alto: Consulting Psychologists Press.

ANXIETY, DEPRESSION, AND ANGER: CORE COMPONENTS OF NEGATIVE AFFECT 675

9781412946513-Ch32 5/12/08 3:25 PM Page 675

Spielberger, C.D., Ritterband, L.M., Sydeman, S.J., Reheiser, E.C. and Unger, K.K. (1995) ‘Assesessment of emotional states and personality traits, measuring psychological vital signs’, in J.N. Butcher (ed.), Clinical Personality Assessment: Practical Approaches. New York: Oxford University Press, pp. 42–58.

Suls, J. and Bunde, J. (2005) ‘Anger, anxiety, and depression as risk factors for cardiovascular disease: The problems and implications of overlapping affective dispositions’, Psychological Bulletin, 131(2): 260–300.

Teva, I., Bermúdez, M., Hernández-Quero, J. and Buela-Casal, G. (2005) ‘Evaluación de la depresión, ansiedad e ira en pacientes con VIH/SIDA [Assessment of depression, anxiety, and anger in HIV/AIDS patients]. Salud Mental, 28(5): 40–9.

Temoshok, L. (1987) ‘Personality coping style, emotion and cancer: Towards an integrative model’, Cancer Surveys, 6(3): 545–67.

Tijhuis, M.A.R., Elshout, J.R.A.F. Feskens, E.J.M., Janssen, M., and Kromhout, D. (2000) ‘Prospective investigation of emo- tional control and cancer risk in men (the

Zutphen Elderly Study) (The Netherlands)’, Cancer Causes and Control, 11(7): 589–95.

Tomkins, S.S. (1962) Affect Imagery Consciousness (Vol I). The positive affects. New York: Springer.

Turk, D.C., Rudy, T.E. and Salovey, P. (1985) ‘The McGill Pain Questionnaire reconsidered: Confirming the factor structure and examin- ing appropriate uses’, Pain, 21(4): 386–97.

Watson, D., Clark, L.A. and Tellegen, A. (1988) ‘Development and validation of brief meas- ures of positive and negative affect: The PANAS scales’, Journal of Personality and Social Psychology, 54(6): 1063–70.

Watson, M. and Greer, S. (1983) ‘Development of a questionnaire measure of emotional control. Journal of Psychosomatic Research, 27(4): 299–305.

Zuckerman, M. and Lubin, B. (1965) Manual for the Multiple Affect Adjective Check List. San Diego: Educational and Industrial Testing Service.

Zuckerman, M. and Lubin, B. (1985) Manual for the MAACL-R: The Multiple Affect Adjective Check List Revised. San Diego: Educational and Industrial Testing Service.

676 THE SAGE HANDBOOK OF PERSONALITY THEORY AND ASSESSMENT

9781412946513-Ch32 5/12/08 3:25 PM Page 676