Factor structure of the Hypomanic Attitudes and Positive Predictions Inventory and associations with analogue bipolar symptoms in a student sample
Introduction
Research into the cognitive processes underlying psychological disorders has facilitated the development of a number of theory-driven measures, which can be used to aid the formulation of psychosocial interventions. The current study describes a factor analysis of a model-driven measure of cognitive style proposed to make people vulnerable to mood swings and bipolar disorder (BD), the Hypomanic Attitudes and Positive Predictions Inventory (HAPPI; Mansell, 2006), which is a potential therapeutic tool.
In parallel with psychological approaches to BD, a number of measures have been used to examine the processes underlying the development and maintenance of BD, including measures of reward sensitivity (Behavioural Inhibition and Behavioural Activation Scales; Carver & White, 1994), and highly ambitious goal-attainment beliefs (Willingly Approached Set of Statistically Unlikely Pursuits, WASSUP; Johnson and Carver, 2006, Johnson et al., 2009). There are also findings that suggest BD is associated with a negative cognitive style similar to that of individuals with unipolar depression (UPD) (Mansell & Scott, 2006).
In a further line of work, internal attributions of circadian rhythm disruption have been implicated in vulnerability to BD (Healy and Williams, 1989, Jones, 2001). The Hypomanic Interpretations Questionnaire (HIQ; Jones, Mansell, & Waller, 2006) was developed to measure positive self-dispositional interpretations of hypomania-relevant experiences (e.g., racing thoughts, impulsivity), and distinguished individuals with BD from non-clinical controls. The Interpretations of Depression Questionnaire (IDQ; Jones & Day, 2008) measures negative self-dispositional explanations of dysphoric experiences, such as pessimistic ideation. The HIQ and IDQ have been associated with bipolar risk in analogue samples (Jones and Day, 2008, Jones et al., 2006).
The unique contribution of both positive and negative cognitions to BD has highlighted the need for a multifaceted approach to understanding the psychological mechanisms underlying BD, to complement and integrate the behavioural and cognitive components of existing approaches.
The HAPPI was developed in conjunction with an integrative cognitive model (Mansell et al., 2007), which proposed that individuals prone to mood swings and BD appraise perceived changes to internal state (related to cognition, behaviour, physiology, or mood) in an extreme manner, incorporating positive and negative personal and interpersonal beliefs. The appraisal occupying awareness at a given time prompts behavioural responses to enhance or control internal state. A self-activating belief may trigger ascent behaviours, such as goal-directed activity, thereby increasing activation. Conversely, a catastrophic appraisal may decrease activation by triggering descent behaviours, such as withdrawing from other people. Further internal state changes, provoked by these behaviours, are again appraised in an extreme, personalised manner, culminating in further internal state changes. At the extreme, these mood swings would manifest as clinical episodes of mania and depression.
The HAPPI (Mansell, 2006) was formulated to measure these appraisals. Items represented five categories of belief about internal states: Self Activation, Response Style, Other Positive, Other Negative, and Catastrophic beliefs. The HAPPI, and its subscales, distinguished between those with a diagnosis of BD and non-clinical controls (Mansell, 2006, Mansell and Jones, 2006), and between individuals with BD and UPD (Alatiq et al., 2010, Seal et al., 2008).
In a student sample (Mansell, Rigby, Tai, & Lowe, 2008), principal components analysis (PCA) of the original HAPPI was conducted, identifying five latent variables; Catastrophic, Reduced Social Regulation, Activating Response Style, Success Activation and Triumph Over Fear, and Loss of Control When Activated. The HAPPI was associated with a history of hypomanic symptoms. Further, the independent dimensions had unique associations with activation and depression over a 24 h period, as measured by the Internal State Scale (ISS; Bauer et al., 1991).
In a subsequent analogue study (Dodd et al., 2010), items were added to the HAPPI to reflect further types of belief implicated in research, clinical experience and theory. Goal-attainment appraisals supplemented the original Self Activation items, reflecting findings of extreme goal-attainment and ambitious beliefs in high-risk and bipolar populations (Johnson and Carver, 2006, Lam et al., 2004). Items reflecting self-criticism and perceived criticism from others were added, as self-criticism was discussed in Mansell et al.’s (2007) model as a means of assigning personal significance to internal state changes. Finally, items were included to reflect clinical observations that individuals with BD often describe overwhelming confusion when high.
A PCA was conducted on this extended version (Dodd et al., 2010). A six-factor solution emerged, largely overlapping with themes identified by Mansell and Jones, 2006, Mansell et al., 2008; Self Activation, Self-and-Other Critical, Catastrophic, Extreme Social Approval, Extreme Appraisals of Agitation, and Loss of Control. The HAPPI predicted activation after 3 months (Dodd et al., 2010), which has been argued to be a core feature of hypomania (Bauer et al., 1991, Benazzi, 2007).
Taken together, results suggest that the HAPPI represents several categories of extreme positive and negative appraisals of internal states. However, both studies employed relatively small samples. Further, while PCA is the most commonly used method of extraction, it has been argued that it is not a ‘true’ method of factor analysis (Costello and Osborne, 2005, Ford et al., 1986, Henson and Roberts, 2006). The purpose of the current study was to replicate the HAPPI themes using a ‘true’ factor analytic procedure in a larger sample, in line with recommendations (Costello & Osborne, 2005).
Section snippets
Participants
Students (N = 302) were recruited for three separate studies involving the HAPPI. Individuals with missing data were excluded (N = 293). The mean age was 20.67 years (SD = 3.6), and 243 were female.
HAPPI (Mansell, 2006)
Participants rated 61 statements (e.g., ‘When I feel good about myself, I realise that all my previous anxieties and fears are unfounded’) on a visual analogue scale from 0 = ‘I don’t believe this at all’ to 100 = ‘I believe this completely’.
ISS (Bauer et al., 1991)
This is a self-report measure of bipolar mood states. The present
Factor analysis of the HAPPI
Data was analysed using SPSS for Windows (Version 16.0). Maximum likelihood (ML) extraction and Varimax rotation were applied. For the overall scale, internal consistency was good (Cronbach’s α = 0.97). Normality assumptions were tenable, so it was considered appropriate to continue using ML.
A parallel analysis was conducted to identify the number of factors to retain, using SPSS syntax provided online (O’Connor, 2000a, O’Connor, 2000b). 100 random datasets were produced, with 293 cases and 61
Discussion
ML extraction using Varimax rotation produced a six-factor structure; Social Self-Criticism, Increasing Activation to Avoid Failure, Success Activation and Triumph Over Fear, Loss of Control, Grandiose Appraisals of Ideation, and Regaining Autonomy, overlapping considerably with components from previous studies (Dodd et al., 2010, Mansell et al., 2008). Social Self-Criticism, Increasing Activation to Avoid Failure and Regaining Autonomy were all uniquely associated with Conflict and Depression,
Conclusions
The model (Mansell et al., 2007) would predict that individuals who endorse strong conviction in multiple beliefs (both positive and negative) would be at increased vulnerability to mood swings, and this was partially supported by emerging relationships between HAPPI factors and high and low internal states.
The appraisals measured by the HAPPI could also be construed as reflecting hypomanic symptoms; for example, Grandiose Appraisals of Ideation. However, each of the items has been purposely
Acknowledgements
The work was carried out at the University of Manchester. This research was funded by the School of Psychological Sciences at the University of Manchester, and formed part of the first author’s PhD thesis, awarded by the University of Manchester in December 2009. Dr. Warren Mansell is supported by the ESRC Emotion Regulation of Others and Self (EROS) research grant.
References (37)
- et al.
Dysfunctional beliefs in bipolar disorder: Hypomanic vs. depressive attitudes
Journal of Affective Disorders
(2010) A cognitive approach to panic
Behaviour Research and Therapy
(1986)Circadian rhythms, multilevel models of emotion and bipolar disorder – An initial step towards integration?
Clinical Psychology Review
(2001)- et al.
Self appraisal and behavioural activation in the prediction of hypomanic personality and depressive symptoms
Personality and Individual Differences
(2008) - et al.
Appraisal of hypomania-relevant experiences: Development of a questionnaire to assess positive self-dispositional appraisals in bipolar and behavioural high risk samples
Journal of Affective Disorders
(2006) - et al.
Dysfunctional assumptions in bipolar disorder
Journal of Affective Disorders
(2004) - et al.
The Brief-HAPPI: A questionnaire to assess cognitions that distinguish between individuals with a diagnosis of bipolar disorder and non-clinical controls
Journal of Affective Disorders
(2006) - et al.
“I Won’t Do What You Tell Me!”: Elevated mood and the assessment of advice-taking in euthymic bipolar I disorder
Behaviour Therapy and Research
(2006) Notes on the psychoanalytic investigation and treatment of manic-depressive insanity and allied conditions
- et al.
Outliers in statistical data
(1994)