Sage Journals: Discover world-class research

Abstract

Objective: To repor t on the relationship between quality of life (QOL), psychiatric symptoms and neuropsychological functioning in a sample of young people who have experienced a first episode of psychosis 2–3 years following initial presentation.

Method: Fifty-one participants aged 15–27 years old completed the short form of the World Health Organization Quality of Life scale (WHOQOL-Brèf), a self-repor t instrument assessing physical, psychological, social and environmental aspects of QOL. A comprehensive neuropsychological batterywas administered. Measures of psychiatric symptoms including depression (as assessed by the Calgary Depression Scale), positive, negative and general psychopathology (as assessed by the Positive and Negative Syndrome Scale) were obtained.

Results: Multiple regression analyses were used to evaluate the ability of neuropsychological measures and psychiatric symptoms to predict QOL. When neuropsychological variables were considered on their own, cognitive flexibility, verbal fluency, verbal ability and sustained attention explained up to 28% of the variance in the four domains of QOL. However, in the presence of psychiatric symptoms, neuropsychological variables were no longer significant predictors for physical and psychological QOL; depression, general psychopathology and negative symptoms together explained up to 43% of the variance in QOL, with neuropsychological variables remaining significant for social and environmental QOL.

Conclusions: In young people with their first episode of psychosis, QOL is more strongly related to levels of psychopathology, particularly depression, than neuropsychological deficits. This finding replicates previous studies in chronic schizophrenia that have suggested QOL is more strongly related to levels of psychopathology than the presence of neuropsychological deficits.

Keywords

cognition depression first-episode psychosis psychopathology quality of life

Quality of life (QOL) is used as a valuable outcome measure for evaluating the impact of psychosis on patients' daily lives. Although no consensus exists regarding how to define QOL, most definitions refer to the individual's subjective satisfaction with various aspects of their lives including physical health, psychological state, social functioning and living conditions [1–3]. There is some debate regarding the capacity of people with a psychotic illness to make accurate assessments of QOL because of poor insight or the effect of their symptoms. However, validation studies of a QOL measure such as the World Health Organization Quality of Life (WHOQOL) scale [4] have shown that there is a significant correlation between results obtained from patientrated and clinician-rated QOL [5–7]. In addition, studies have shown that individuals with psychosis can make reasonable judgments about their QOL in the absence of severe psychopathology [3].

Symptomatology not only affects the ability of patients to assess QOL, but also has been assumed to play a substantial role the in the patients' QOL. In reviewing the literature regarding QOL and chronic schizophrenia, Pinikahana et al. [8] concluded that when psychopathology was considered, negative symptoms were usually found to be the most important determinant of QOL among patients with psychotic disorders. In the few published longitudinal first-episode psychosis (FEP) studies, negative symptoms were again observed to be an important determinant of QOL [9] along with depressive symptoms [10].

It is well established that individuals with psychotic disorders have impaired cognitive functioning [11], [12]. However, few studies have investigated the relationship between QOL and cognitive functioning [8], [13]. Early studies [14], [15] reported a positive relation between QOL and psychomotor speed, aspects of both verbal and visual memory and learning and verbal fluency. When the effects of cognitive dysfunction and psychopathology on QOL have been considered together, the results of crosssectional studies (focusing on mainly chronic schizophrenia) have suggested that psychopathology [16], and particularly negative symptoms [11], [17], [18] might be more strongly related to QOL than cognition.

These studies reported at least one significant relationship between QOL and cognitive functioning. Yet no individual cognitive domain was consistently associated with QOL and on balance, it would appear that both cognitive impairment and psychiatric status may be related to QOL. Clarification of these relationships has important implications for management and treatment of individuals with psychosis. Ifwe can determine specific predictors of QOL outcomes in these patients, we identify targets for remediation which might, in turn, improve their QOL.

The purpose of this study was to examine the relative contributions of neuropsychological deficits and psychiatric symptomatology to QOL. To the best of our knowledge, we are the first to report on this issue in a sample of young people who have experienced a first episode of psychosis. The aims and hypotheses of the current study were as follows:

1 To investigate the ability of neuropsychologicalvari ables alone to predict QOL. Given the variable results of previous studies, we intended this as an exploratory analysis and expected that one or more neuropsychological variables would significantly predict QOL.

2 To investigate whether neuropsychological deficits continued to predict QOL when psychiatric status was also considered. It was expected that: (i) neuropsychological deficits would not predict QOL; and (ii) psychiatric status would predict QOL.

Method

Participants

Participants were drawn from the Western Sydney First-Episode Psychosis Project, which is a longitudinal study [19]. The current paper presents only the cross-sectional results of 51 of these participants who consented to take part in the 2–3-year follow-up phase of the study. Independent sample t-tests revealed no significant differences (at α= 0.05) between the demographic characteristics of those participants who returned for 2–3-year follow-up assessment and those who did not.

Psychiatric status

According to DSM-IV [20] criteria, participants had diagnoses of schizophrenia (n = 29), schizophreniform disorder (n = 2), schizoaffective disorder (n = 6), drug-induced psychosis (n = 3), bipolar disorder (n = 8), brief psychotic disorder (n = 1) and psychotic disorder not otherwise specified (n = 2). Diagnosis was made by the consensus of three psychiatrists on the basis of a longitudinal evaluation of all available data. When assessed, 33 individuals were prescribed atypical antipsychotics with a mean chlorpromazine-equivalent dose of 390 mg day⁻¹ (SD = 248 mg day⁻¹) [21] and 18 were not prescribed any antipsychotic medication.

Psychiatric status was assessed using the Positive and Negative Syndrome Scale (PANSS; [22]). Ratings were performed by psychiatrists trained in the use of this scale and who had reached a satisfactory level of interrater reliability (κ<0.80). The PANSS subscales of Positive (7 items), Negative (7 items) and General Psychopathology (16 items) were used as independent variables. Depression was assessed with the Calgary Depression Scale (CDS; [23]) which is a clinicianrated measure using information provided by the patient and which has been widely used in psychosis research [6], [7]. Independent sample t-tests were performed comparing the clinical characteristics of those participants who returned for 2–3-year follow-up with those who did not. No significant differences were observed at an alpha level of 0.05.

Neuropsychological functioning

Participants completed a comprehensive battery ofneuropsychological tests at 2–3-year follow-up (see 24]; visual memory with the delayed recall score on the Rey Complex Figure Test (RCFT; [25]); verbal recall with the delayed recall score of the Rey Auditory Verbal Learning Test (RAVLT; [25]); attention with number of hits on the Continuous Performance Test (CPT; [26]); verbal fluency (BHR) from the Delis–Kaplan Executive Functioning System (DKEFS; [27]); and cognitive flexibility was assessed according to the number of repeated sorts on the Sorting Test of the DKEFS [27]. Mean z-scores (according to the published normative data indicated above) for each of these measures appear in Table 1.

Table 1.

Mean z-scores and standard deviations for first-episode psychosis group relative to normative data

Neuropsychological test	Mean	SD
WASI vocabulary	−0.03	1.10
RCFT delayed recall	−1.02	1.13
RAVLT delayed recall	−0.90	1.75
CPT hits	−0.45	1.01
DKEFS verbal fluency (BHR)	−0.43	0.92
DKEFS repeated sorts	0.13	0.95

CPT, Conner's Continuous Performance Test [26]; DKEFS, Delis–Kaplan Executive Function System [27]; RAVLT, Rey Auditory Verbal Learning Test [25]; RCFT, Rey Complex Figure Test [25]; WASI, Wechsler Abbreviated Scale of Intelligence [24].

Quality of life

The short form of the WHOQOL (WHOQOL-Brèf) [28] is a 26-item self-report questionnaire assessing QOL. It includes four domain scores (physical, psychological, social and environment) and two individually scored items measuring the individual's overall perception of their QOL and satisfaction with their physical health. All items are rated on a fivepoint Likert scale. Domain scores range from 0 to 100, whereas the individual items have a score range of 1–5. In all cases, higher scores reflect better QOL. The internal consistencyof the domain scores is generally good. For example, the Cronbach's coefficient alphas for physical (r = 0.87), psychological (r = 0.81) and environmental (r = 0.81) QOL are acceptable, although somewhat lower for the domain of social QOL (r = 0.68). This latter finding is partly because of scale length (the social domain includes only three items), but may also be attributable to the poor correlation between items [28]. Test–retest reliabilities for the domains are all excellent (r = 0.83–0.86). The four domains of QOL were used in analyses because of a lack of variability in the distributions of the two global measures (overall QOL and satisfaction with physical health).

Statistical analyses

Tests of normality revealed the dependent variables (i.e. QOL domains) to be normally distributed. In order to examine the ability of neuropsychological variables alone to predict scores on each of the four QOL domains (physical, psychological, social and environmental), separate multiple regression analyses using the backward stepwise elimination method were performed. Subsequently, the ability of psychiatric status and neuropsychological variables together to predict QOL was examined with a further four multiple regression analyses. All analyses were conducted using the Statistical Package for Social Sciences (SPSS)V 11.0 [29].

Results

At the time of testing, subjects had a mean age of 21.5 (SD = 2.9) and education of 12.1 years (SD = 1.9). Sixty-seven per cent were male. All participants were outpatients, 38 of whom lived with their family of origin. The clinical characteristics of our study sample are as follows: mean PANSS Positive subscale score 13.41 (SD = 6.03); mean PANSS Negative score 14.04 (SD = 6.48); mean PANSS General Psychopathology score 29.84 (SD = 10.08); and mean CDS score 3.06 (SD = 2.90).

The results of multiple regression analyses using the backwardstepwise elimination method to predict the four QOL domain scores from the six neuropsychological variables are presented in Table 2. In each case QOL was entered as the dependent variable and the six neuropsychological variables were entered as independent variables. The criterion for entry into an equation was set at 0.03 to provide reasonable protection against Type 1 error. Excluded variables are not reported. All domains of QOL were significantly predicted by one or more neuropsychological variables. Specifically, physical QOL was significantly predicted by performance on a test of sustained attention (F_(1,48) = 7.63, p= 0.008), which accounted for 13.7% of the variance in this outcome measure. Environmental QOL was significantly predicted by performance on tests of verbal ability, verbal fluency and cognitive flexibility (F_(3,46) = 5.91, p= 0.002), which accounted for 27.8% of the variance in this outcome measure. Social QOL was significantly predicted by performance on tests of verbal fluencyand cognitive flexibility (F_(2,47) = 6.93, p= 0.002), which accounted for 22.8% of the variance. Psychological QOL was significantly predicted by performance on tests of verbal fluency and cognitive flexibility (F_(2,47) = 3.88, p= 0.027), accounting for 14.2% of the variance.

Table 2.

Results of multiple linear regression of neuropsychological variables alone on eachofthe four quality of life (QOL) domains

Outcome variable	Predictors	Beta	t-value	Significance of t
Physical QOL	Sustained attention	0.37	2.76	0.008
Environmental QOL	Verbal ability	0.37	2.73	0.0009
	Verbal fluency	0.36	2.33	0.0243
	Cognitive flexibility	0.40	2.60	0.013
Psychological QOL	Verbal fluency	0.39	2.37	0.022
	Cognitive flexibility	0.42	2.55	0.014
Social QOL	Verbal fluency	0.38	2.45	0.018
	Cognitive flexibility	0.57	3.69	0.001

The results of further multiple regression analyses when psychiatric status variables were also entered are presented in Table 3. In these analyses, QOL was entered as the dependent variable, whereas the four psychiatric status variables and six neuropsychological variables were entered as two families of independent variables. Psychiatric variables were entered into the model first because previous research has uniformly supported the role of these variables in QOL, and neuropsychological variables were entered second as there is relatively less support for the unique contribution of these variables to QOL. The criterion for entry into an equation remained at 0.03. Psychiatric status variables were significant predictors of all four domains of QOL. The environmental and social QOL were also significantly predicted by neuropsychological variables. Physical QOL was significantly predicted by General Psychopathology (F_(1,48) = 22.80, p< 0.001), which accounted for 32.2% of the variance. Environmental QOL was significantly predicted by General Psychopathology, Negative Symptoms and performance on a measure of verbal ability (F_(3,46) = 11.34, p< 0.001), accounting for 42.5% of the variance. Psychological QOL was significantly predicted by the Depression score (F_(1,48) = 31.09, p< 0.001), which accounted for 39.3% of the variance. Social QOL was significantly predicted by Depression and performance on tests of sustained attention, verbal fluency and cognitive flexibility (F_(4,45) = 8.47, p< 0.001), which accounted for 43.0% of the variance in this outcome measure.

Table 3.

Results of multiple linear regression of psychiatric and neuropsychological variables on each of the four quality of life (QOL) domains

Outcome variable	Predictor	Beta	t-value	p-value
Physical QOL	General psychopathology	−0.57	−4.78	<0.001
Environmental QOL	Negative symptoms	0.46	2.40	0.021
	General psychopathology	−0.58	−4.92	<0.001
	Verbal ability	0.36	2.49	0.016
Psychological QOL	Depression	−0.63	−5.58	<0.001
Social QOL	Depression	−0.46	−3.74	0.001
	Sustained attention	−0.33	−2.69	0.010
	Verbal fluency	0.32	2.30	0.026
	Cognitive flexibility	0.50	3.62	0.001

Discussion

In general, the hypothesis that neuropsychological deficits would predict diminished QOL when considered alone was confirmed. Neuropsychological variables produced small-effect sizes (variance ranging from 14% to 28%), which is broadly consistent with previous reports of significant but fairly modest associations between QOL and cognitive functioning [11],[16–18]. Verbal fluency and cognitive flexibility were found to be significantly associated with poorer ‘subjective’ QOL in three out of the four domains. This is particularly interesting in light of the common perception that executive deficits are associated with poorer insight [30].

The finding that neuropsychological functioning was unrelated to physical and psychological QOL in the presence of psychiatric status variables supported our second hypothesis and was consistent with previous reports suggesting that the relationship between neuropsychological functioning and these aspects of QOL is largely mediated by the degree of patients' psychiatric distress [11], [18].

Contrary to our expectations, however, social QOL was significantly predicted by performance on neuropsychological tests of sustained attention, verbal fluency and cognitive flexibility, and environmental QOL being predicted by verbal ability. This represents the first evidence that cognitive functioning provides a sizeable additional explanation of any aspect of QOL over and above the contribution of psychiatric symptoms. Despite this, it is important not to overinterpret the specific relationships in the absence of replication. However, an interaction of verbal fluency and cognitive flexibility with social QOL is appealing, given the need for adequate communication abilities for reasonable social interaction.

Interestingly, higher negative symptoms were also associated with higher environmental QOL, which goes against the published works that report a negative correlation between QOL and overall negative symptomatology [8]. Again this finding requires replication.

In line with our third hypothesis, psychiatric status predicted QOL (when entered with the neuropsychological variables). Lower General Psychopathology scores predicted higher physical and environmental QOL, whereas lower depression predicted higher psychological and social QOL. It is important to remember that the PANSS General Psychopathology scale includes items that rate depression, so the agreement of both measures of psychopathology may well represent the overwhelming importance of depression to QOL.

The relationship between social and psychological QOL was specifically with depression as measured by the CDS, supporting the primary role of mood over positive and negative symptoms. This finding is certainly consistent with several previous reports in chronic schizophrenia [16], [31], [32]. Moreover, Priebe et al. [10] found changes in depression and anxiety over time to correlate with changes in QOL reports in their sample of FEP patients. However, given that the bulk of prior research has suggested a primary role for negative symptoms in QOL, it may have been expected that they would emerge as a significant predictor in the current study. This was not observed.

Our results inevitably reflect our choice of instrument. Unlike the Quality of Life Scale [33], which was developed assuming that primary enduring negative symptoms of schizophrenia would impact upon patients' life satisfaction [34], [35], the WHOQOL-Brèf was not. This may make our findings more difficult to compare with other studies, and until a ‘gold-standard’ measure of QOL is established, will be a continued source of confusion in published works.

It is possible that the failure to identify a consistent relationship between negative symptoms and QOL may have resulted from the youth and lack of chronicity in our sample. Subjects had relatively low levels of negative symptoms and had mostly recovered from their initial psychotic episode. The dependence on chronic and institutionalized samples in other studies may at least partly account for their findings of an association between negative symptomatology and QOL. However, the average degree of depressive symptomatology in our sample was also quite low, but still emerged as being highly predictive of two aspects of QOL (psychological and social). In addition to low levels of depression, our sample showed low overall levels of positive symptomatology and cognitive impairment. Thus, by Katschnig's [3] reasoning, it would follow that the assessment of QOL in our sample was valid.

The issue of cross-sectional versus longitudinal studies in this area is important. The present study, although cross-sectional, has identified variables that will be examined over time. The next step will be to track the longitudinal relationships of symptoms and cognitive function to QOL.

This study underlines important treatment issues. If the aim of treatment is to improve our patient's overall wellbeing and not just to reduce psychotic phenomena in this cohort of young people, more attention is required in the management of their depressive symptoms. Further treatment targeting specific cognitive disabilities is also likely to materially improve their QOL. The inclusion of a QOL measure would appear to be a valuable additional assessment in the evaluation of treatment outcome.

References

Sartorius

. A WHO method for the assessment of health-related quality of life (WHOQOL). Quality of life assessment: key issues in the 1990s, Walker

Rosser

. Kluwer Academic, Dordrecht 1990; 201–207.

Gupta

Kulhara

Verma

S K

. Quality of life in schizophrenia and dysthymia. Acta Psychiatrica Scandinavica 1998; 97: 290–296.

Katschnig

. Schizophrenia and quality of life. Acta Psychiatrica Scandinavica 2000; 102(Suppl)33–37.

WHOQOL Group . The World Health Organisation WHOQOL-BREF quality of life assessment: position paper from the World Health Organisation. Social Science and Medicine 1998; 46: 1569–1585.

Dogan

Tel

Coker

Polatoz

Dogan

F B

. Psychosocial approaches in outpatients with schizophrenia. Psychiatric Rehabilitation Journal 2004; 27: 279–282.

Herrman

Hawthorne

Thomas

. Quality of life assessment in people living with psychosis. Social Psychiatry and Psychiatric Epidemiology 2002; 37: 510–518.

Becchi

Rucci

Placentino

Neri

de Girolamo

. Quality of life in patients with schizophrenia. Social Psychiatry and Psychiatric Epidemiology 2004; 39: 397–401.

Pinikahana

Happell

Hope

Keks

N A

. Quality of life in schizophrenia:a review ofthe literature from 1995–2000. International Journal of Mental Health Nursing 2002; 11: 103–111.

B C

Nopoulos

Flaum

Arudt

Andreasen

N C

. Two year outcome in first episode schizophrenia: predictive value of symptoms for quality of life. American Journal of Psychiatry 1998; 155: 1196–1201.

10.

Priebe

Roeder-Wanner

U U

Kaiser

. Quality of life in first-admitted schizophrenia patients: a follow-up study. Psychological Medicine 2000; 30: 225–230.

11.

Addington

. Neurocognitive and social functioning in schizophrenia. Schizophrenia Bulletin 1999; 25: 173–182.

12.

Fitzgerald

Lucas

Redoblado

M A

. Cognitive functioning in young people with first episode psychosis: relationship to diagnosis and clinical characteristics. Australian and New Zealand Journal of Psychiatry 2004; 38: 501–510.

13.

Green

M F

Kern

R S

Braff

D L

Mintz

. Neurocognitive deficits and functional outcome in schizophrenia: are we measuring the “right stuff”?. Schizophrenia Bulletin 2000; 26: 119–136.

14.

Meltzer

H Y

Thompson

P A

Lee

M A

Ranjan

. Neuropsychologic deficits in schizophrenia: relation to social functionand effect of antipsychotic drug treatment. Neuropsychopharmacology 1996; 14(Suppl)27–33.

15.

Buchanan

R W

Holstein

Brier

. The comparative efficacy and long-term effect of clozapine treatment on neuropsychological test performance. Society of Biological Psychiatry 1994; 36: 717–725.

16.

Heslegrave

R J

Awad

A G

Voruganti

L NP

. The influence of neurocognitive deficits and symptoms on quality of life in schizophrenia. Journal of Psychiatry and Neuroscience 1997; 22: 235–242.

17.

Aksaray

Oflu

Kaptanoglu

Bal

. Neurocognitive deficits and quality of life in outpatients with schizophrenia. Progress in Neuro-psychopharmacology and Biological Psychiatry 2002; 26: 1217–1219.

18.

Addington

. Neurocognitive and social functioning in schizophrenia: a 2.5 year follow-up study. Schizophrenia Research 2000; 44: 47–56.

19.

Harris

Brennan

Anderson

. The clinical profile of first episode psychosis in Western Sydney –scope and general findings of the Western Sydney First Episode Psychosis Project. Australian and New Zealand Journal of Psychiatry 2005; 39: 36–43.

20.

American Psychiatric Association . Diagnostic and statistical manual of mental disorders 4th edn. American Psychiatric Press, Washington, DC 1994.

21.

Lambert

. Switching to risperdal. University of Melbourne, The Multimedia in Psychiatry Unit, Melbourne 1999.

22.

Kay

S R

Fizbein

Opler

L A

. The Positive and Negative Syndrome Scale (PANSS) for schizophrenia. Schizophrenia Bulletin 1987; 13: 261–276.

23.

Addington

. Cognitive functioning in first episode schizophrenia. Journal of Psychiatry and Neuroscience 2002; 27: 188–192.

24.

Wechsler

. Wechsler Abbreviated Scale of Intelligence. The Psychological Corporation, San Antonio, TX 1999.

25.

Spreen

Strauss

. A compendium of neuropsychological tests: administration, norms and commentary 2nd edn. Oxford University Press, New York, NY 1998.

26.

Conners

C K

. Multi-health Systems Staff. Conners' Continuous Performance Test. MHS, Toronto 1995.

27.

Delis

D C

Kaplan

Kramer

J H

. Delis–Kaplan Executive Functioning System. The Psychological Corporation, San Antonio, TX 2001.

28.

Hawthorne

Richardson

Day

McNeil

. Validation of the WHOQOL-BREF. Centre for Health Program Evaluation, Melbourne 2000.

29.

Statistical Package for the Social Sciences (SPSS) . SPSS for Windows (version 11.0). SPSS Inc, Chicago 2001.

30.

Laroi

Fannemel

Ronnebberg

. Unawareness of illness in chronic schizophrenia and its relationship to structural brain measures and neuropsychological tests. Psychiatry Research 2000; 100: 49–58.

31.

Tollefson

G D

Andersen

S W

. Should we consider mood disturbances in schizophrenia as an important determinant of quality of life. Journal of Clinical Psychiatry 1999; 60: 23–29.

32.

Huppert

J D

Weiss

K A

Lim

Pratt

Smith

T E

. Quality of life in schizophrenia: contributions of anxiety and depression. Schizophrenia Research 2001; 51: 171–180.

33.

Heinrichs

D W

Hanlon

T E

Carpenter

W T

. The Quality of Life Scale: an instrument for rating the schizophrenic deficit syndrome. Schizophrenia Bulletin 1984; 10: 388–398.

34.

Bow-Thomas

C C

Velligan

D I

Miller

A L

Olsen

. Predicting quality of life from symptomatology at exacerbation and stabilisation. Psychiatry Research 1999; 86: 131–142.

35.

Norman

R MG

Malla

A K

McLean

. The relationship of symptoms and level of functioning in schizophrenia to general wellbeing and the quality of life scale. Acta Psychiatrica Scandinavica 2000; 102: 303–309.

Relative Contributions of Psychiatric Symptoms and Neuropsychological Functioning to Quality of Life in First-Episode Psychosis

Abstract

Keywords

Method

Participants

Psychiatric status

Neuropsychological functioning

Quality of life

Statistical analyses

Results

Discussion

References