- Email: [email protected]
Prospective memory and its correlates and predictors in schizophrenia: An extension of previous findings
Archives of Clinical Neuropsychology 23 (2008) 613–622
Prospective memory and its correlates and predictors in schizophrenia: An extension of previous findings Gabor S. Ungvari a , Yu-Tao Xiang a,b,∗ , Wai-Kwong Tang a , David Shum c a
Department of Psychiatry, Chinese University of Hong Kong, Hong Kong, China b Beijing Anding Hospital, Capital Medical University, Beijing, China c School of Psychology, Griffith University, Brisbane, Queensland, Australia Accepted 5 June 2008
Abstract BACKGROUND: Prospective memory (PM) is the ability to remember to do something in the future without explicit prompts. Extending the number of subjects and the scope of our previously published study, this investigation examined the relationship between PM and socio-demographic and clinical factors, activities of daily living (ADL) and frontal lobe functions in patients with chronic schizophrenia. METHODS: One hundred and ten Chinese schizophrenia patients, 60 from the previous study and 50 additional patients recruited for this study, and 110 matched healthy comparison subjects (HC) formed the study sample. Patients’ clinical condition and activity of daily living were evaluated with the Brief Psychiatric Rating Scale (BPRS) and the Functional Needs Assessment (FNA). Timeand event-based PM tasks and three tests of prefrontal lobe functions (Design Fluency Test [DFT], Tower of London [TOL], Wisconsin Card Sorting Test [WCST]) were also administered. RESULTS: Patients’ level of ADL and psychopathology were not associated with PM functions and only anticholinergic medications (ACM) showed a significant negative correlational relationship with PM tasks. Confirming the findings of the previous study, patients performed significantly more poorly on all two PM tasks than HC. Performance on time-based PM task significantly correlated with age, education level and DFT in HC and with age, DFT, TOL and WCST in patients. Patients’ performance on the event-based PM correlated with DFT and one measure of WCST. In patients, TOL and age predicted the performance on time-based PM task; DFT and WCST predicted the event-based task. CONCLUSIONS: Involving a large sample of patients with matched controls, this study confirmed that PM is impaired in chronic schizophrenia. Deficient PM functions were related to prefrontal lobe dysfunction in both HC and patients but not to the patients’ clinical condition, nor did they significantly affect ADL. ACMs determined certain aspects of PM. © 2008 National Academy of Neuropsychology. Published by Elsevier Ltd. All rights reserved. Keywords: Schizophrenia; Prospective memory; Activity of daily living; Psychotropic drugs; Prefrontal lobe
∗ Corresponding author at: Department of Psychiatry, Shatin Hospital, Shatin, N.T., Hong Kong, China. Tel.: +852 2636 7748; fax: +852 2647 5321. E-mail address: [email protected] (Y.-T. Xiang).
0887-6177/$ – see front matter © 2008 National Academy of Neuropsychology. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.acn.2008.06.005
614
G.S. Ungvari et al. / Archives of Clinical Neuropsychology 23 (2008) 613–622
1. Introduction Prospective memory (PM), defined as “memory for activities to be performed in the future” (Einstein & McDaniel, 1990), is gaining attention in research and clinical practice (Kliegel, McDaniel, & Einstein, 2007; McDaniel & Einstein, 2007; West, 2007). Of the various aspects of PM, time- and event-based subtypes of PM (Einstein & McDaniel, 1990; Kvavilashvili & Ellis, 1996) are addressed in this study. Time-based PM refers to remembering to carry out an intention at a specific moment. Event-based PM is defined as remembering to perform an intended action triggered by an external event. To date a number of studies have addressed PM and its correlates in schizophrenia (Elvevag, Maylor, & Gilbert, 2003; Henry, Rendell, Kliegel, & Altgassen, 2007; Kondel, 2002; Kumar, Nizamie, & Jahan, 2005, 2007; Meissner, Hacker, & Heilemann, 2001; Ritch, Velligan, Tucker, Dicocco, & Maples, 2003; Shum, Ungvari, Tang, & Leung, 2004; Twamley et al., 2007; Wang et al., 2007; Woods, Twamley, Dawson, Narvaez, & Jeste, 2007). Despite the widely different methods in these studies to assess PM, schizophrenia patients were found to perform significantly worse than healthy comparison subjects (HC) in PM tasks in all investigations that had a HC group. Therefore, PM deficits were proposed to be trait-dependent characteristics of the illness (Kumar et al., 2005; Wang et al., 2007). However, there were methodological limitations in the above-mentioned studies. Sample sizes were relatively small ranging from 10 (Kondel, 2002) to 72 subjects (Twamley et al., 2007). The selection of subjects (e.g., the sampling method or inclusion/exclusion criteria used) was not elaborated in the studies published so far with one exception (Shum et al., 2004). Two studies did not include HC groups (Kondel, 2002; Ritch et al., 2003). Two studies (Kumar et al., 2005, 2007) examined drug-free, acutely ill psychotic patients and nearly one-third of them could not recall the PM task. The PM impairment found in these studies, therefore, could be confounded by retrospective memory. Except for two studies (Ritch et al., 2003; Twamley et al., 2007), the impact of PM deficit on patients’ activity of daily living (ADL) was not measured. None of the studies have examined the relationship between clinical symptoms measured by standard rating scales and PM in schizophrenia, nor has there been exploration of the potential impact of psychotropic drugs on PM functions. In clinical practice, it is expected that PM has a major impact on patients’ ADL and quality of life. Due to the devastating effects of psychopathology and the social stigma associated with schizophrenia, a sizeable proportion of patients are unemployed, socially isolated and lead a disorganized life (Shen, 2002). Impairment of PM could make any rehabilitative efforts difficult as patients could not follow through steps towards any organized meaningful activity. In addition, problems related to deficient PM, such as missing appointments or take medication, could lead to early relapse and further functional impairment. To date, only two studies have addressed the relationship of PM with ADL. Ritch et al. (2003) found that secondary verbal memory and sum score of time- and event-based PM predicted more than 29% of the variance of ADL scores in 50 schizophrenia patients. Twamley et al. (2007) also found that better performance on PM tests predicted higher functional capacity measured by a standardized ADL scale in 72 schizophrenia patients. In a previous study (Shum et al., 2004) we assessed three subtypes (time-, event-, and activity-based) of PM based on the paradigm designed by Einstein, McDaniel, Richardson, Guynn and Cunfer (1995) and their association with prefrontal lobe functions in 60 schizophrenia patients and 60 matched HC. As an extension of the previous study, the primary objective of this investigation was to clarify the relationships between PM and a host of socio-demographic and clinical variables and ADL as these associations were not addressed in our preliminary study. The secondary objective was to confirm the results of our preliminary study concerning the relationship between time-and event-based PM and frontal lobe tests using a larger sample thereby increasing the power of the investigation. We focused on timeand event-based PM because these are the most frequently investigated aspects of PM, particularly in subjects with schizophrenia. In this report activity-based PM was not included because of the ceiling effect associated with the task used in our previous study and the inappropriateness of conducting correlational and regression analyses on data with limited or no variation. The first hypothesis was that psychopathology, psychotropic drugs and ADL would be significantly associated with the patients’ performance on PM tasks. This is because these factors are related to impaired cognitive functions in schizophrenia patients (Lindenmayer, Khan, Iskander, Abad, & Parker, 2007; O’Grada & Dinan, 2007). The second hypothesis was that patients with long-term schizophrenia would perform significantly worse on both types of PM compared to matched HC. The third hypothesis was that performance on prefrontal lobe functions would be significantly correlated with performance on PM tasks confirming the results of our previous study (Shum et al., 2004).
G.S. Ungvari et al. / Archives of Clinical Neuropsychology 23 (2008) 613–622
615
2. Methods 2.1. Study settings and participants The selection and assessment of the subjects have been described in detail elsewhere (Shum et al., 2004). Briefly, schizophrenia inpatients were recruited from the long-term rehabilitation unit in Hong Kong. HC without a history of substance abuse and psychiatric, neurological or major medical disorders were selected from a vocational training center (Hong Kong College of Technology) and matched to patients according to age, sex and educational level. Patients who met the following inclusion criteria entered the study: (1) diagnosis of schizophrenia according to DSM-IV (APA, 1994); (2) Chinese ethnicity; (3) age between 18 and 50 years; (4) length of illness of at least 2 years; (5) at least primary education and ability to understand the requirements of the study; (6) proficiency in the Cantonese dialect; (7) availability of at least one informant to provide and/or corroborate demographic and clinical data; (8) current antipsychotic medication (AP) less than 800 mg/day in chlorpromazine equivalent according to standard guidelines (Taylor, Paton, & Kerwin, 2005); (9) ability and willingness to give consent to participate in the study. Exclusion criteria were: (1) electroconvulsive therapy (ECT) in the past 12 months; (2) past or current significant drug/alcohol abuse; (3) history of, or current significant medical condition including head trauma; (4) current administration of antidepressant or benzodiazepine drugs; (5) dose of benzhexol exceeding 6 mg/day. The study protocol was approved by the local Clinical Research Ethics Committee. Written consent was obtained from all patients and controls. 3. Outcome measures 3.1. Brief Psychiatric Rating Scale Psychotic symptoms were measured by two mean subscores of the Brief Psychiatric Rating Scale (BPRS): (1) positive symptoms of conceptual disorganization, suspiciousness, hallucinatory behavior, and unusual thought content; (2) negative symptoms of emotional withdrawal, motor retardation, blunted affect, and disorientation (Xiang, Weng, Leung, Tang, & Ungvari, 2008). 3.2. Functional Needs Assessment-Chinese version (FNA-C) ADL was assessed with the locally validated Chinese version of Functional Needs Assessment (FNA) (Dombrowski, Kane, Tuttle, & Kincaid, 1990; Law, 1999). The FNA is a 26-item interviewer-rated, comprehensive scale of performance-based components of ADL including self-care (e.g., personal hygiene, care of belongings, and eating) and community living skills (e.g., safety, use of transport, shopping, preparing food, laundry, budgeting, and social manners). Each component is scored between 0 and 25 points. The sum score of the FNA was entered into the statistical analysis. 3.3. PM tasks Each PM tasks lasted 25 min. Prior to testing, all subjects were asked to remove their wristwatch. Subjects were told that they would undertake an ongoing task by themselves, with the research assistant (RA) in the next room. They were also required to perform a number of PM-related actions depending on the experimental condition. The ongoing task was a three-choice general-knowledge test presented on a computer screen. Subjects were asked to read the questions which appeared on the center of the screen and to respond by pressing the corresponding answer key on the keyboard. They were given feedback for each question, and their cumulative score of correct answers was shown at the bottom of the screen. In the time-based condition, subjects were required to contact the RA every 5-min using an intercom to inform him of the cumulative correct score on the general-knowledge task. To keep track of the time, they could press the space bar on the computer bringing a digital clock on the screen for 2 s. They were allowed to press the space bar at any time for reference as often as they liked. The measure obtained in this task was the percentage of correct prospective responses at 5, 10, 15, 20 and 25 min. Responses within ±20 s of these times were regarded as correct.
616
G.S. Ungvari et al. / Archives of Clinical Neuropsychology 23 (2008) 613–622
In the event-based condition, subjects had to contact the RA via the intercom whenever they saw the word “police” written in any position of the general-knowledge questions of the ongoing task. Five questions with the target word appeared at 2, 8, 12, 16 and 22 min. These times were randomly selected to avoid familiarity with a fixed time interval. The measure obtained was the percentage of correct responses. 3.4. Frontal lobe tests For the Design Fluency Test (DFT) (Jones-Gotman & Milner, 1977) the measures were the number of novel outputs for the free and fixed response conditions. The 5-min free-response condition did not have any restriction imposed on design generation. In the 4-min fixed-response condition, subjects were required to produce designs that contain exactly four lines or components. Measures were the number of novel output scores for each condition. For the Wisconsin Card Sorting Test (WCST) (Heaton, Chelune, Talley, Kay, & Curtis 1993) the following measures were obtained: (1) the number of categories completed, (2) the number of trials to complete the first category, (3) the total number of errors, and (4) perseverative errors. This study used a version of Tower of London (TOL) (Shallice, 1982) developed by Tunstall (1999). The measure was the number of attempts to achieve a given arrangements of the disks; the score ranged from 0 to 30. 3.5. Procedures All subjects were assessed in a quiet and well-lit room, always in the morning. Neuropsychological tests were administered first followed by the PM tasks. Testing usually lasted 2–3 h and was completed within a day for controls and two consecutive days for patients. FNA-C and the BPRS were rated on the same days as the above tests. For the PM tasks, all subjects were asked to repeat the instructions before starting the task to ensure that they understood them. Instructions were repeated and explanations were provided for those participants who seemed uncertain about the exact nature of the tasks. Upon completion of the PM tasks, subjects were again asked to repeat the requirements of the tests to make sure they remembered and understood what was expected of them. 3.6. Statistical analysis Data were analyzed using SPSS for Windows (Version 13.0). Comparisons between the two groups of subjects with regard to the percentage of correctly answered general knowledge questions on the two PM tasks, mean percentage of correct responses on the two PM tasks, and mean scores on prefrontal lobe tests were performed with independent samples t-test and Mann–Whitney U-test where appropriate. Associations between PM tasks and all prefrontal lobe tests, social functions and the continuous socio-demographic and clinical variables were analyzed using Pearson’s correlation coefficients given that the data followed normal distribution; otherwise Spearman rank correlation analysis was performed. Stepwise multiple linear regression analyses were used separately to identify factors that predict performance on each PM task in patients. Categorical variables were entered into the regression equations as dummy variables. To avoid multicolinearity, tolerance was used to measure the strength of the linear relationships between the independent variables; a tolerance value of 0.6 and above was regarded as acceptable. The normality of distributions for continuous variables was checked with the one-sample Kolmogorov–Smirnov test. Two-tailed tests were used in all analyses with the significance level set at 0.01. 4. Results The patient group consisted of 110 (72 males) schizophrenia inpatients of whom 60 were from the previous study (Shum et al., 2004) while 50 additional patients were recruited for this study. All patients meeting inclusion criteria at the rehabilitation unit participated in the study. Their mean age, educational level and length of illness were 31.7 ± 7.4 (range: 20–48), 10.2 ± 2.3 (range: 4–16), and 8.4 ± 5.2 (range: 2–23) years, respectively. The control group comprised 110 (73 males) healthy persons, 60 from the previous study (Shum et al., 2004) and 50 newly recruited from the same source as before. Their mean age and educational level were 32 ± 7 (range: 18–48) and 10 ± 2 years (range: 5–16), respectively. The two groups did not differ in age (t(218) = −0.21, p = 0.82) and educational level (t(218) = 0.56, p = 0.57).
G.S. Ungvari et al. / Archives of Clinical Neuropsychology 23 (2008) 613–622
617
Table 1 Comparison of mean scores of patients and healthy comparison subjects on general knowledge questions, PM tasks, and prefrontal lobe tests Measures
General knowledge for time-based taskb General knowledge for event-based taskc Time-based taskd Event-based taskf DFTg (free response condition) DFT (fixed response condition) TOLh (total score) WCSTi (perseverative errors) WCST (categories completed) WCST (trials to complete the first category) a b c d e f g h i
HCa
Patients
Mean
S.D.
Mean
S.D.
t/z
d.f.
p
72.96 80.46 80.90 92.97 13.93 15.76 21.61 14.41 5.29 19.40
6.56 5.49 26.18 12.80 7.81 7.44 2.21 12.10 1.47 21.97
63.04 70.47 38.18 75.34 7.42 7.07 19.16 34.32 3.18 51.35
8.88 9.74 34.32 31.89 8.08 5.62 3.79 25.37 2.37 48.94
9.41 9.36 −8.42e 5.38 −6.84e −8.63e 5.85 −7.03e −7.29e −5.52e
218 218
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
218
218
Healthy comparison subjects. General knowledge/time-based = % of general knowledge questions answered correctly for the time-based task. General knowledge/event-based = % of general knowledge questions answered correctly for the event-based task. Time-based = mean % of correct responses on time-based PM task. Mann–Whitney test. Event-based = mean % of correct responses on event-based PM task. DFT = Design Fluency Test. TOL = Tower of London. WCST = Wisconsin Card Sorting Test.
Table 1 shows the performances of patients and HC on general knowledge questions, PM tasks, and prefrontal lobe tests. Table 2 presents the correlation analyses of PM tasks with prefrontal lobe tests and basic socio-demographic and clinical variables. Tables 3 and 4 depict the results of stepwise multiple regression analyses exploring the predictors of performance on each PM tasks in patients. All continuous variables which showed significant bivariate correlation relationships to PM task at a = 0.05 level and categorical variables including use of antipsychotic polypharmacy (APP), traditional AP only, atypical AP only and anticholinergic medications (ACM) were entered as independent variables, and each PM task score was entered as a dependent variable. 5. Discussion To the best of our knowledge this is the first study to examine the impact of both socio-demographic and clinical variables on time- and event-based PM tasks in schizophrenia. The first hypothesis that symptomatology, use of psychotropic drugs, and patients’ ADL would be significantly associated with their performance on PM tasks was only partly confirmed. ACMs are known to influence other domains of cognitive capacity in schizophrenia (e.g., Brebion, Bressan, Amador, Malaspina, & Gorman, 2004) and they also impaired patients’ performance on event-based PM tasks. This is a new finding as hitherto no study has determined the association between the use of ACMs, and performance on PM tasks. The results underline the deleterious effect of ACM on PM tasks suggesting that ACMs should be prescribed sparingly and only for the strictest possible indication, particularly for clinically stable patients. Despite their general sedative effects, AP drugs did not influence PM functions although the doses were in the therapeutic range. Several lines of evidence suggest that atypical APs improve various aspects of cognitive functions in schizophrenia including memory, learning and executive functions (Lindenmayer et al., 2007; O’Grada & Dinan, 2007). Unexpectedly, in this study atypical APs did not have an impact on patients’ performance on PM tasks. The lack of association between PM and AP treatment warrants further investigations. In agreement with Wang et al. (2007) and Kumar et al.’s (2005, 2007) studies, and contrary to a recent report (Woods et al., 2007), no significant inverse relationship between negative or positive symptoms and PM tasks was found in this study. Different instruments measuring both symptomatology and PM tasks used in the aforementioned studies may
618
G.S. Ungvari et al. / Archives of Clinical Neuropsychology 23 (2008) 613–622
Table 2 Correlation between PM tasks and prefrontal lobe tests and basic socio-demographic and clinical variables in patients and healthy comparison subjects Measures
PM tasks Time-based (n = 110)
Event-based (n = 110)
Healthy comparison subjects DFTa (free response condition) DFT (fixed response condition) TOLb (total score) WCSTc (perseverative errors) WCST (categories completed) WCST (trials to complete the first category) Age Educational level (years)
0.22* 0.28** 0.23* −0.18 0.006 −0.06 −0.27** 0.27**
0.03 0.06 0.11 0.06 0.11 −0.13 −0.03 0.10
Patients DFT (free response condition) DFT (fixed response condition) TOL (total score) WCST (perseverative errors) WCST (categories completed) WCST (trials to complete the first category) Age Education (years) Length of illness (years) BPRSd total BPRS positive BPRS negative CPZeq (mg)e FNA-Cf On APPg On traditional APh only On atypical AP only On ACMi
0.23* 0.29** 0.30** −0.28** 0.29** −0.29** −0.25** 0.13 −0.23* −0.07 0.003 0.10 −0.17 0.16 −0.09 0.001 0.15 −0.24*
0.41** 0.34** 0.15 −0.20* 0.25** −0.23* −0.05 0.21* −0.08 −0.11 −0.18 0.04 −0.11 0.04 −0.18 −0.01 0.07 −0.25**
*p < 0.05; **p < 0.01. a DFT = Design Fluency Test. b TOL = Tower of London. c WCST = Wisconsin Card Sorting Test. d BPRS = Brief Psychotic Rating Scale. e CPZeq = Chlorpromazine equivalent (mg). f FNA-C = Functional Needs Assessment-Chinese version. g APP = Antipsychotic polypharmacy. h AP = Antipsychotic drug. i ACM = Anticholinergic medication.
account for this discrepancy. At this point it seems that psychopathology and length of illness (Kumar et al., 2005, 2007; Meissner et al., 2001) exert minimal influence of PM functions. Evidence is emerging that overall cognitive functions and social adaptation in schizophrenia are closely related (Addington & Addington, 2008). It was expected that deficits of PM in schizophrenia patients could possibly lead to disorganized ADL (Shum et al., 2004). Contrary to previous findings (Ritch et al., 2003; Twamley et al., 2007) however, in this study there was no significant negative correlation between PM and patients’ ADL. This absence of significant relationship could possibly be explained by the relative insensitivity of the FNA. The relationship between PM tasks and ADL needs to be further explored. In line with earlier results (Henry et al., 2007; Woods et al., 2007), better time-based PM performance was associated with younger age in this study. As expected, higher educational level was also significantly correlated with better time-based PM performance.
G.S. Ungvari et al. / Archives of Clinical Neuropsychology 23 (2008) 613–622
619
Table 3 Predictors of time-based PM task in the patient group by stepwise multiple regression analysis Beta Variables that entered the model (adjusted R2 = 0.17; F(3,106) = 8.93; p < 0.001) 0.25 TOLa (total score) Age −0.24 0.23 WCSTb (categories completed) Variables had failed to enter the model DFTc (free response condition) DFT (fixed response condition) WCST (perseverative errors) WCST (trials to complete the first category) Length of illness Sex On APPd On traditional APe only On atypical AP only On ACMf a b c d e f
T
p
2.88 −2.74 2.57
0.005 0.007 0.011
– –
1.01 1.69
0.31 0.09
– – – – – – – –
−0.98 0.10 0.72 0.07 −1.14 1.20 0.19 −1.82
0.32 0.92 0.47 0.94 0.25 0.23 0.84 0.07
TOL = Tower of London. WCST = Wisconsin Card Sorting Test. DFT = Design Fluency Test. APP = Antipsychotic polypharmacy. AP = Antipsychotic drug. ACM = Anticholinergic medication.
The second hypothesis that schizophrenia patients would show impairment in the two types of PM compared to HC was confirmed. The discrepancy between the two groups was not likely to be due to the patients’ inability to comprehend and remember the instructions of the PM tasks since all subjects could understand and recall the instructions and the details of the two PM tasks before and after the tasks. The general knowledge questions for the PM tasks were expected to place similar cognitive demand on patients and HC since the two groups were matched in terms of age and education. Yet, patients still showed a poorer performance on these questions. This is probably due to the combination of cognitive Table 4 Predictors of event-based PM task in the patient group by stepwise multiple regression analysis Beta Variables that entered the model (adjusted DFTa (fixed response condition) On ACMb WCSTc (perseverative errors)
R2
= 0.20; F(3,106) = 9.83; p < 0.001) 0.26 −0.22 −0.20
Variables had failed to enter the model DFT (free response condition) WCST (categories completed) WCST (trials to complete the first category) Sex On APPd On traditional APe only On atypical AP only Educational level a b c d e
DFT = Design Fluency Test. ACM = Anticholinergic medication. WCST = Wisconsin Card Sorting Test. APP = Antipsychotic polypharmacy. AP = Antipsychotic drug.
T
p
2.87 −2.58 −2.16
0.005 0.011 0.033
–
1.43
0.15
– – – – – – –
0.21 −0.22 −0.25 −1.50 0.56 −0.23 1.46
0.83 0.81 0.79 0.13 0.57 0.81 0.14
620
G.S. Ungvari et al. / Archives of Clinical Neuropsychology 23 (2008) 613–622
deficits related to the illness, the effects of psychotropic drugs, as well as the impact of negative symptoms such as decreased interest and motivation. It is not yet clear what neuropsychological mechanisms account for schizophrenia patients’ poor performance on PM tasks. Woods et al.’s study (2007) found defective cue detection and self-initiated retrieval underlying event- and time-based PM impairment. For habitual PM tasks, which are particularly relevant from a practical point of view (e.g., regularly taking medication at certain intervals), deficient internal source monitoring seems to be an important contributing factor (Elvevag et al., 2003). Meissner et al. (2001) came to a similar conclusion emphasizing the deficits in strategic thinking underlying PM deficits. Another possible explanation put forward by Kumar et al. (2005) postulates that inhibitory mechanisms that suppress irrelevant thoughts are suspended in giving way to the interruption of the relevant cognitive processes including PM memory in schizophrenia (Hasher & Zacks, 1998). Twamley et al. (2007) reported that PM was positively associated with a number of cognitive domains including attention, working memory, processing speed, learning and executive functions. Henry et al. (2007) proposed that the unique deficit of PM in schizophrenia is a general difficulty mostly independent of other cognitive and executive functions. However, without including other illness groups (e.g., mood disorders, brain injury) as comparisons, the specificity of such impairment will remain a mere conjecture. Schizophrenia is associated with impairment in prefrontal lobe functions (e.g., Tan, Callocitt, & Weinberger, 2007). Time-based tasks place more demand on the prefrontal cortex (Einstein et al., 1995) requiring intact executive functions (Ritch et al., 2003). This is one possible explanation for the larger difference in performance on time-based task compared to the event-based task between patients and HC. These results are in line with our earlier findings (Shum et al., 2004) and those of other (Kumar et al., 2007; Wang et al., 2007), but not all (Woods et al., 2007) reports. The discrepancy between these studies may be due to the diversity of assessment instruments in terms of length, ongoing task, delay interval, and response modality used to measure PM and prefrontal cortex functions. An alternative explanation for the different performance on time- and event-based tasks could be that the event-based task in this study was simply less effortful thus requiring less involvement of executive functions. However, the two tasks in this and other studies (e.g., Shum, Valentine, & Cutmore, 1999) were comparable in all aspects (e.g., ongoing task, the interval between PM task encoding and performance) except for the PM cue (external vs. self-initiated). Yet, the performance on event-based task was consistently superior to the time-based one in all studies which argues for the important, if not exclusive, role of the executive functions’ integrity. The third hypothesis that performance on frontal lobe functions would be significantly correlated with performance on PM tasks was also supported confirming our previous findings (Shum et al., 2004) in a considerably larger sample of patients. The possibility of a causal link between PM and frontal lobe dysfunctions was discussed in our previous report (Shum et al., 2004). As no large scale studies have assessed PM and other cognitive functions simultaneously in various patient groups, the connection between PM and executive functions related to intact prefrontal cortex remains a hypothesis requiring further research. 5.1. Strengths and limitations of the study and future directions of research The strengths of the study are its large sample size, matched HC, standardized PM assessment tools and the independent assessment of the various aspects of the investigation. The study included all available schizophrenia patients who were hospitalized during the study period avoiding major selection bias. However, the results should be interpreted with caution because of methodological limitations. First, the study examined Chinese subjects with chronic, clinically stable schizophrenia thus the results may not be applicable to patients of other ethnicity or at other stages of their illness. Second, a potential selection bias was the exclusion of illiterate and more severely ill patients receiving high-doses of psychotropic drugs who could not cooperate. Third, retrospective memory and other aspects of cognitive functions were not assessed. Subjects’ cognitive capacity was judged only by clinical impression and their ability to recall the PM tasks before and after the testing. Fourth, the study adopted a cross-sectional design; the causality of relationships is, therefore, tentative. Fifth, participants were required to remove their wristwatch during PM tasks, which might have limited the ecological validity of the time-based test. Sixth, to avoid or minimize any possible retroactive interference in PM tasks, other types of assessments should have been arranged between the two PM tasks. In future studies more detailed and sensitive instruments measuring psychopathology and social functions should be employed. In addition to HC, control groups comprising patients with other psychiatric and neurological disorders
G.S. Ungvari et al. / Archives of Clinical Neuropsychology 23 (2008) 613–622
621
should be included. Future studies should also include comprehensive cognitive assessment combining with neuroimaging investigations. This approach coupled with a longitudinal design would allow researchers to determine if there is a characteristic signature of PM deficit in schizophrenia and may shed light on the dynamics of its development. Acknowledgements This study was sponsored by a Hong Kong Research Grants Council Earmarked Research Grant. The authors are grateful to the assistance of Mr. Jacky Yuen in data collection and to the late Dr. Jin Pang Leung for his contribution in setting up the study. References Addington, J., & Addington, D. (2008). Social and cognitive functioning in psychosis. Schizophrenia Research, 99, 176–181. APA. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington: American Psychiatric Association. Brebion, G., Bressan, R. A., Amador, X., Malaspina, D., & Gorman, J. M. (2004). Medications and verbal memory impairment in schizophrenia: The role of anticholinergic drugs. Psychological Medicine, 34, 369–374. Dombrowski, S. B., Kane, M., Tuttle, N. B., & Kincaid, W. (1990). Functional needs assessment program for chronic psychiatric patients. Tucson, Arizona: Communication Skill Builders. Einstein, G. O., & McDaniel, M. A. (1990). Normal aging and prospective memory. Journal of Experimental Psychology-Learning Memory and Cognition, 16, 717–726. Einstein, G. O., McDaniel, M. A., Richardson, S. L., Guynn, M. J., & Cunfer, A. R. (1995). Aging and prospective memory: Examining the influences of self-initiated retrieval processes. Journal of Experimental Psychology-Learning Memory and Cognition, 21, 996–1007. Elvevag, B., Maylor, E. A., & Gilbert, A. L. (2003). Habitual prospective memory in schizophrenia. BMC Psychiatry, 3, 9. Hasher, L., & Zacks, R. T. (1998). Working memory dysfunction and its relevance to schizophrenia. In G. Bower (Ed.), The psychology of learning and motivation (pp. 193–225). New York: Academic Press. Heaton, R. K., Chelune, G. J., Talley, J. L., Kay, G. G., & Curtis, G. (1993). Wisconsin Card Sorting Test (WCST) Manual: Revised and expanded. Henry, J. D., Rendell, P. G., Kliegel, M., & Altgassen, M. (2007). Prospective memory in schizophrenia: Primary or secondary impairment? Schizophrenia Research, 95, 179–185. Jones-Gotman, M., & Milner, B. (1977). Design fluency: The invention of nonsense drawings after focal cortical lesions. Neuropsychologia, 15, 653–674. Kliegel, M., McDaniel, M. A., & Einstein, G. O. (2007). Prospective memory: Cognitive neuroscience, developmental and applied perspectives. Mahwah, NJ: Lawrence Erlbaum. Kondel, T. K. (2002). Prospective memory and executive function in schizophrenia. Brain and Cognition, 48, 405–410. Kumar, D., Nizamie, H. S., & Jahan, M. (2005). Event-based prospective memory in schizophrenia. Journal of Clinical and Experimental Neuropsychology, 27, 867–872. Kumar, D., Nizamie, S. H., & Jahan, M. (2007). Activity-based prospective memory in schizophrenia. Clinical Neuropsychology, 1–10. Kvavilashvili, L., & Ellis, J. (1996). Varieties of intention: Some distinctions and classification. In M. Brandimonte, G. O. Einstein, & M. A. McDaniel (Eds.), Prospective memory: Theory and applications (pp. 23–52). Mahwah, NJ: Lawrence Erlbaum. Law, K. M. (1999). Validation of functional needs assessment (Chinese Version) for people with chronic schizophrenia. Unpublished MPhil Thesis. Hong Kong, China: Hong Kong Polytechnic University. Lindenmayer, J. P., Khan, A., Iskander, A., Abad, M. T., & Parker, B. (2007). A randomized controlled trial of olanzapine versus haloperidol in the treatment of primary negative symptoms and neurocognitive deficits in schizophrenia. Journal of Clinical Psychiatry, 68, 368–379. McDaniel, M. A., & Einstein, G. O. (2007). Prospective memory. An overview and synthesis of an emerging field. Thousand Oaks, CA: Sage Publications. Meissner, F., Hacker, W., & Heilemann, H. (2001). Memory performance and instruction effects in schizophrenia. A comparative study of chronic schizophrenic patients and healthy probands (in German). Psychiatrische Praxis, 28, 180–188. O’Grada, C., & Dinan, T. (2007). Executive function in schizophrenia: What impact do antipsychotics have? Human Psychopharmacology, 22, 397–406. Ritch, J. L., Velligan, D. I., Tucker, D., Dicocco, M., & Maples, N. J. (2003). Prospective memory in schizophrenia. Schizophrenia Research, 60, 180. Shallice, T. (1982). Specific impairments of planning. Philosophical Transactions of the Royal Society of London Series B, Biological Science, 298, 199–209. Shen, Y. C. (2002). Psychiatry. Beijing: People’s Medical Publishing House., pp. 734–735. Shum, D., Ungvari, G. S., Tang, W. K., & Leung, J. P. (2004). Performance of schizophrenia patients on time-, event-, and activity-based prospective memory tasks. Schizophrenia Bulletin, 30, 693–701. Shum, D., Valentine, M., & Cutmore, T. (1999). Performance of individuals with severe long-term traumatic brain injury on time-, event-, and activity-based prospective memory tasks. Journal of Clinical Neuropsychology, 21, 49–58. Tan, H. Y., Callocitt, J. H., & Weinberger, D. R. (2007). Dysfunctional and compensatory prefrontal cortical systems, genes and the pathogenesis of schizophrenia. Cerebral Cortex, 17(Suppl 1), 171–181. Taylor, D., Paton, C., & Kerwin, R. (2005). The Maudsley Prescribing Guidelines (2005–2006). Abingdon: Taylor & Francis.
622
G.S. Ungvari et al. / Archives of Clinical Neuropsychology 23 (2008) 613–622
Tunstall, J. (1999). Improving the utility of the Tower of London: A neuropsychological test of planning. Unpublished MPhil thesis. Brisbane, Australia: Griffith University. Twamley, E. W., Woods, S. P., Zurhellen, C. H., Vertinski, M., Narvaez, J. M., Mausbach, B. T., et al. (2007). Neuropsychological substrates and everyday functioning implications of prospective memory impairment in schizophrenia. Schizophrenia Research, doi:10.1016/j.schres.2007.10.030 Wang, Y., Chan, R. C., Xin, Y., Shi, C., Cui, J., & Deng, Y. (2007). Prospective memory deficits in subjects with schizophrenia spectrum disorders: A comparison study with schizophrenic subjects, psychometrically defined schizotypal subjects, and healthy controls. Schizophrenia Research, doi:10.1016/j.schres.2007.07020 West, R. (2007). Towards a cognitive neuroscience of prospective memory. In M. Kliegel, M. A. McDaniel, & G. O. Einstein (Eds.), Prospective memory: Cognitive neuroscience developmental and applied perspectives (pp. 125–145). Mahwah, NJ: Lawrence Erlbaum. Woods, S. P., Twamley, E. W., Dawson, M. S., Narvaez, J. M., & Jeste, D. V. (2007). Deficits in cue detection and intention retrieval underlie prospective memory impairment in schizophrenia. Schizophrenia Research, 90, 344–350. Xiang, Y. T., Weng, Y. Z., Leung, C. M., Tang, W. K., & Ungvari, G. S. (2008). Subjective quality of life in outpatients with schizophrenia in Hong Kong and Beijing: Relationship to socio-demographic and clinical factors. Quality of Life Research, 17, 27–36.
Prospective memory and its correlates and predictors in schizophrenia: An extension of previous findings Gabor S. Ungvari a , Yu-Tao Xiang a,b,∗ , Wai-Kwong Tang a , David Shum c a
Department of Psychiatry, Chinese University of Hong Kong, Hong Kong, China b Beijing Anding Hospital, Capital Medical University, Beijing, China c School of Psychology, Griffith University, Brisbane, Queensland, Australia Accepted 5 June 2008
Abstract BACKGROUND: Prospective memory (PM) is the ability to remember to do something in the future without explicit prompts. Extending the number of subjects and the scope of our previously published study, this investigation examined the relationship between PM and socio-demographic and clinical factors, activities of daily living (ADL) and frontal lobe functions in patients with chronic schizophrenia. METHODS: One hundred and ten Chinese schizophrenia patients, 60 from the previous study and 50 additional patients recruited for this study, and 110 matched healthy comparison subjects (HC) formed the study sample. Patients’ clinical condition and activity of daily living were evaluated with the Brief Psychiatric Rating Scale (BPRS) and the Functional Needs Assessment (FNA). Timeand event-based PM tasks and three tests of prefrontal lobe functions (Design Fluency Test [DFT], Tower of London [TOL], Wisconsin Card Sorting Test [WCST]) were also administered. RESULTS: Patients’ level of ADL and psychopathology were not associated with PM functions and only anticholinergic medications (ACM) showed a significant negative correlational relationship with PM tasks. Confirming the findings of the previous study, patients performed significantly more poorly on all two PM tasks than HC. Performance on time-based PM task significantly correlated with age, education level and DFT in HC and with age, DFT, TOL and WCST in patients. Patients’ performance on the event-based PM correlated with DFT and one measure of WCST. In patients, TOL and age predicted the performance on time-based PM task; DFT and WCST predicted the event-based task. CONCLUSIONS: Involving a large sample of patients with matched controls, this study confirmed that PM is impaired in chronic schizophrenia. Deficient PM functions were related to prefrontal lobe dysfunction in both HC and patients but not to the patients’ clinical condition, nor did they significantly affect ADL. ACMs determined certain aspects of PM. © 2008 National Academy of Neuropsychology. Published by Elsevier Ltd. All rights reserved. Keywords: Schizophrenia; Prospective memory; Activity of daily living; Psychotropic drugs; Prefrontal lobe
∗ Corresponding author at: Department of Psychiatry, Shatin Hospital, Shatin, N.T., Hong Kong, China. Tel.: +852 2636 7748; fax: +852 2647 5321. E-mail address: [email protected] (Y.-T. Xiang).
0887-6177/$ – see front matter © 2008 National Academy of Neuropsychology. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.acn.2008.06.005
614
G.S. Ungvari et al. / Archives of Clinical Neuropsychology 23 (2008) 613–622
1. Introduction Prospective memory (PM), defined as “memory for activities to be performed in the future” (Einstein & McDaniel, 1990), is gaining attention in research and clinical practice (Kliegel, McDaniel, & Einstein, 2007; McDaniel & Einstein, 2007; West, 2007). Of the various aspects of PM, time- and event-based subtypes of PM (Einstein & McDaniel, 1990; Kvavilashvili & Ellis, 1996) are addressed in this study. Time-based PM refers to remembering to carry out an intention at a specific moment. Event-based PM is defined as remembering to perform an intended action triggered by an external event. To date a number of studies have addressed PM and its correlates in schizophrenia (Elvevag, Maylor, & Gilbert, 2003; Henry, Rendell, Kliegel, & Altgassen, 2007; Kondel, 2002; Kumar, Nizamie, & Jahan, 2005, 2007; Meissner, Hacker, & Heilemann, 2001; Ritch, Velligan, Tucker, Dicocco, & Maples, 2003; Shum, Ungvari, Tang, & Leung, 2004; Twamley et al., 2007; Wang et al., 2007; Woods, Twamley, Dawson, Narvaez, & Jeste, 2007). Despite the widely different methods in these studies to assess PM, schizophrenia patients were found to perform significantly worse than healthy comparison subjects (HC) in PM tasks in all investigations that had a HC group. Therefore, PM deficits were proposed to be trait-dependent characteristics of the illness (Kumar et al., 2005; Wang et al., 2007). However, there were methodological limitations in the above-mentioned studies. Sample sizes were relatively small ranging from 10 (Kondel, 2002) to 72 subjects (Twamley et al., 2007). The selection of subjects (e.g., the sampling method or inclusion/exclusion criteria used) was not elaborated in the studies published so far with one exception (Shum et al., 2004). Two studies did not include HC groups (Kondel, 2002; Ritch et al., 2003). Two studies (Kumar et al., 2005, 2007) examined drug-free, acutely ill psychotic patients and nearly one-third of them could not recall the PM task. The PM impairment found in these studies, therefore, could be confounded by retrospective memory. Except for two studies (Ritch et al., 2003; Twamley et al., 2007), the impact of PM deficit on patients’ activity of daily living (ADL) was not measured. None of the studies have examined the relationship between clinical symptoms measured by standard rating scales and PM in schizophrenia, nor has there been exploration of the potential impact of psychotropic drugs on PM functions. In clinical practice, it is expected that PM has a major impact on patients’ ADL and quality of life. Due to the devastating effects of psychopathology and the social stigma associated with schizophrenia, a sizeable proportion of patients are unemployed, socially isolated and lead a disorganized life (Shen, 2002). Impairment of PM could make any rehabilitative efforts difficult as patients could not follow through steps towards any organized meaningful activity. In addition, problems related to deficient PM, such as missing appointments or take medication, could lead to early relapse and further functional impairment. To date, only two studies have addressed the relationship of PM with ADL. Ritch et al. (2003) found that secondary verbal memory and sum score of time- and event-based PM predicted more than 29% of the variance of ADL scores in 50 schizophrenia patients. Twamley et al. (2007) also found that better performance on PM tests predicted higher functional capacity measured by a standardized ADL scale in 72 schizophrenia patients. In a previous study (Shum et al., 2004) we assessed three subtypes (time-, event-, and activity-based) of PM based on the paradigm designed by Einstein, McDaniel, Richardson, Guynn and Cunfer (1995) and their association with prefrontal lobe functions in 60 schizophrenia patients and 60 matched HC. As an extension of the previous study, the primary objective of this investigation was to clarify the relationships between PM and a host of socio-demographic and clinical variables and ADL as these associations were not addressed in our preliminary study. The secondary objective was to confirm the results of our preliminary study concerning the relationship between time-and event-based PM and frontal lobe tests using a larger sample thereby increasing the power of the investigation. We focused on timeand event-based PM because these are the most frequently investigated aspects of PM, particularly in subjects with schizophrenia. In this report activity-based PM was not included because of the ceiling effect associated with the task used in our previous study and the inappropriateness of conducting correlational and regression analyses on data with limited or no variation. The first hypothesis was that psychopathology, psychotropic drugs and ADL would be significantly associated with the patients’ performance on PM tasks. This is because these factors are related to impaired cognitive functions in schizophrenia patients (Lindenmayer, Khan, Iskander, Abad, & Parker, 2007; O’Grada & Dinan, 2007). The second hypothesis was that patients with long-term schizophrenia would perform significantly worse on both types of PM compared to matched HC. The third hypothesis was that performance on prefrontal lobe functions would be significantly correlated with performance on PM tasks confirming the results of our previous study (Shum et al., 2004).
G.S. Ungvari et al. / Archives of Clinical Neuropsychology 23 (2008) 613–622
615
2. Methods 2.1. Study settings and participants The selection and assessment of the subjects have been described in detail elsewhere (Shum et al., 2004). Briefly, schizophrenia inpatients were recruited from the long-term rehabilitation unit in Hong Kong. HC without a history of substance abuse and psychiatric, neurological or major medical disorders were selected from a vocational training center (Hong Kong College of Technology) and matched to patients according to age, sex and educational level. Patients who met the following inclusion criteria entered the study: (1) diagnosis of schizophrenia according to DSM-IV (APA, 1994); (2) Chinese ethnicity; (3) age between 18 and 50 years; (4) length of illness of at least 2 years; (5) at least primary education and ability to understand the requirements of the study; (6) proficiency in the Cantonese dialect; (7) availability of at least one informant to provide and/or corroborate demographic and clinical data; (8) current antipsychotic medication (AP) less than 800 mg/day in chlorpromazine equivalent according to standard guidelines (Taylor, Paton, & Kerwin, 2005); (9) ability and willingness to give consent to participate in the study. Exclusion criteria were: (1) electroconvulsive therapy (ECT) in the past 12 months; (2) past or current significant drug/alcohol abuse; (3) history of, or current significant medical condition including head trauma; (4) current administration of antidepressant or benzodiazepine drugs; (5) dose of benzhexol exceeding 6 mg/day. The study protocol was approved by the local Clinical Research Ethics Committee. Written consent was obtained from all patients and controls. 3. Outcome measures 3.1. Brief Psychiatric Rating Scale Psychotic symptoms were measured by two mean subscores of the Brief Psychiatric Rating Scale (BPRS): (1) positive symptoms of conceptual disorganization, suspiciousness, hallucinatory behavior, and unusual thought content; (2) negative symptoms of emotional withdrawal, motor retardation, blunted affect, and disorientation (Xiang, Weng, Leung, Tang, & Ungvari, 2008). 3.2. Functional Needs Assessment-Chinese version (FNA-C) ADL was assessed with the locally validated Chinese version of Functional Needs Assessment (FNA) (Dombrowski, Kane, Tuttle, & Kincaid, 1990; Law, 1999). The FNA is a 26-item interviewer-rated, comprehensive scale of performance-based components of ADL including self-care (e.g., personal hygiene, care of belongings, and eating) and community living skills (e.g., safety, use of transport, shopping, preparing food, laundry, budgeting, and social manners). Each component is scored between 0 and 25 points. The sum score of the FNA was entered into the statistical analysis. 3.3. PM tasks Each PM tasks lasted 25 min. Prior to testing, all subjects were asked to remove their wristwatch. Subjects were told that they would undertake an ongoing task by themselves, with the research assistant (RA) in the next room. They were also required to perform a number of PM-related actions depending on the experimental condition. The ongoing task was a three-choice general-knowledge test presented on a computer screen. Subjects were asked to read the questions which appeared on the center of the screen and to respond by pressing the corresponding answer key on the keyboard. They were given feedback for each question, and their cumulative score of correct answers was shown at the bottom of the screen. In the time-based condition, subjects were required to contact the RA every 5-min using an intercom to inform him of the cumulative correct score on the general-knowledge task. To keep track of the time, they could press the space bar on the computer bringing a digital clock on the screen for 2 s. They were allowed to press the space bar at any time for reference as often as they liked. The measure obtained in this task was the percentage of correct prospective responses at 5, 10, 15, 20 and 25 min. Responses within ±20 s of these times were regarded as correct.
616
G.S. Ungvari et al. / Archives of Clinical Neuropsychology 23 (2008) 613–622
In the event-based condition, subjects had to contact the RA via the intercom whenever they saw the word “police” written in any position of the general-knowledge questions of the ongoing task. Five questions with the target word appeared at 2, 8, 12, 16 and 22 min. These times were randomly selected to avoid familiarity with a fixed time interval. The measure obtained was the percentage of correct responses. 3.4. Frontal lobe tests For the Design Fluency Test (DFT) (Jones-Gotman & Milner, 1977) the measures were the number of novel outputs for the free and fixed response conditions. The 5-min free-response condition did not have any restriction imposed on design generation. In the 4-min fixed-response condition, subjects were required to produce designs that contain exactly four lines or components. Measures were the number of novel output scores for each condition. For the Wisconsin Card Sorting Test (WCST) (Heaton, Chelune, Talley, Kay, & Curtis 1993) the following measures were obtained: (1) the number of categories completed, (2) the number of trials to complete the first category, (3) the total number of errors, and (4) perseverative errors. This study used a version of Tower of London (TOL) (Shallice, 1982) developed by Tunstall (1999). The measure was the number of attempts to achieve a given arrangements of the disks; the score ranged from 0 to 30. 3.5. Procedures All subjects were assessed in a quiet and well-lit room, always in the morning. Neuropsychological tests were administered first followed by the PM tasks. Testing usually lasted 2–3 h and was completed within a day for controls and two consecutive days for patients. FNA-C and the BPRS were rated on the same days as the above tests. For the PM tasks, all subjects were asked to repeat the instructions before starting the task to ensure that they understood them. Instructions were repeated and explanations were provided for those participants who seemed uncertain about the exact nature of the tasks. Upon completion of the PM tasks, subjects were again asked to repeat the requirements of the tests to make sure they remembered and understood what was expected of them. 3.6. Statistical analysis Data were analyzed using SPSS for Windows (Version 13.0). Comparisons between the two groups of subjects with regard to the percentage of correctly answered general knowledge questions on the two PM tasks, mean percentage of correct responses on the two PM tasks, and mean scores on prefrontal lobe tests were performed with independent samples t-test and Mann–Whitney U-test where appropriate. Associations between PM tasks and all prefrontal lobe tests, social functions and the continuous socio-demographic and clinical variables were analyzed using Pearson’s correlation coefficients given that the data followed normal distribution; otherwise Spearman rank correlation analysis was performed. Stepwise multiple linear regression analyses were used separately to identify factors that predict performance on each PM task in patients. Categorical variables were entered into the regression equations as dummy variables. To avoid multicolinearity, tolerance was used to measure the strength of the linear relationships between the independent variables; a tolerance value of 0.6 and above was regarded as acceptable. The normality of distributions for continuous variables was checked with the one-sample Kolmogorov–Smirnov test. Two-tailed tests were used in all analyses with the significance level set at 0.01. 4. Results The patient group consisted of 110 (72 males) schizophrenia inpatients of whom 60 were from the previous study (Shum et al., 2004) while 50 additional patients were recruited for this study. All patients meeting inclusion criteria at the rehabilitation unit participated in the study. Their mean age, educational level and length of illness were 31.7 ± 7.4 (range: 20–48), 10.2 ± 2.3 (range: 4–16), and 8.4 ± 5.2 (range: 2–23) years, respectively. The control group comprised 110 (73 males) healthy persons, 60 from the previous study (Shum et al., 2004) and 50 newly recruited from the same source as before. Their mean age and educational level were 32 ± 7 (range: 18–48) and 10 ± 2 years (range: 5–16), respectively. The two groups did not differ in age (t(218) = −0.21, p = 0.82) and educational level (t(218) = 0.56, p = 0.57).
G.S. Ungvari et al. / Archives of Clinical Neuropsychology 23 (2008) 613–622
617
Table 1 Comparison of mean scores of patients and healthy comparison subjects on general knowledge questions, PM tasks, and prefrontal lobe tests Measures
General knowledge for time-based taskb General knowledge for event-based taskc Time-based taskd Event-based taskf DFTg (free response condition) DFT (fixed response condition) TOLh (total score) WCSTi (perseverative errors) WCST (categories completed) WCST (trials to complete the first category) a b c d e f g h i
HCa
Patients
Mean
S.D.
Mean
S.D.
t/z
d.f.
p
72.96 80.46 80.90 92.97 13.93 15.76 21.61 14.41 5.29 19.40
6.56 5.49 26.18 12.80 7.81 7.44 2.21 12.10 1.47 21.97
63.04 70.47 38.18 75.34 7.42 7.07 19.16 34.32 3.18 51.35
8.88 9.74 34.32 31.89 8.08 5.62 3.79 25.37 2.37 48.94
9.41 9.36 −8.42e 5.38 −6.84e −8.63e 5.85 −7.03e −7.29e −5.52e
218 218
<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
218
218
Healthy comparison subjects. General knowledge/time-based = % of general knowledge questions answered correctly for the time-based task. General knowledge/event-based = % of general knowledge questions answered correctly for the event-based task. Time-based = mean % of correct responses on time-based PM task. Mann–Whitney test. Event-based = mean % of correct responses on event-based PM task. DFT = Design Fluency Test. TOL = Tower of London. WCST = Wisconsin Card Sorting Test.
Table 1 shows the performances of patients and HC on general knowledge questions, PM tasks, and prefrontal lobe tests. Table 2 presents the correlation analyses of PM tasks with prefrontal lobe tests and basic socio-demographic and clinical variables. Tables 3 and 4 depict the results of stepwise multiple regression analyses exploring the predictors of performance on each PM tasks in patients. All continuous variables which showed significant bivariate correlation relationships to PM task at a = 0.05 level and categorical variables including use of antipsychotic polypharmacy (APP), traditional AP only, atypical AP only and anticholinergic medications (ACM) were entered as independent variables, and each PM task score was entered as a dependent variable. 5. Discussion To the best of our knowledge this is the first study to examine the impact of both socio-demographic and clinical variables on time- and event-based PM tasks in schizophrenia. The first hypothesis that symptomatology, use of psychotropic drugs, and patients’ ADL would be significantly associated with their performance on PM tasks was only partly confirmed. ACMs are known to influence other domains of cognitive capacity in schizophrenia (e.g., Brebion, Bressan, Amador, Malaspina, & Gorman, 2004) and they also impaired patients’ performance on event-based PM tasks. This is a new finding as hitherto no study has determined the association between the use of ACMs, and performance on PM tasks. The results underline the deleterious effect of ACM on PM tasks suggesting that ACMs should be prescribed sparingly and only for the strictest possible indication, particularly for clinically stable patients. Despite their general sedative effects, AP drugs did not influence PM functions although the doses were in the therapeutic range. Several lines of evidence suggest that atypical APs improve various aspects of cognitive functions in schizophrenia including memory, learning and executive functions (Lindenmayer et al., 2007; O’Grada & Dinan, 2007). Unexpectedly, in this study atypical APs did not have an impact on patients’ performance on PM tasks. The lack of association between PM and AP treatment warrants further investigations. In agreement with Wang et al. (2007) and Kumar et al.’s (2005, 2007) studies, and contrary to a recent report (Woods et al., 2007), no significant inverse relationship between negative or positive symptoms and PM tasks was found in this study. Different instruments measuring both symptomatology and PM tasks used in the aforementioned studies may
618
G.S. Ungvari et al. / Archives of Clinical Neuropsychology 23 (2008) 613–622
Table 2 Correlation between PM tasks and prefrontal lobe tests and basic socio-demographic and clinical variables in patients and healthy comparison subjects Measures
PM tasks Time-based (n = 110)
Event-based (n = 110)
Healthy comparison subjects DFTa (free response condition) DFT (fixed response condition) TOLb (total score) WCSTc (perseverative errors) WCST (categories completed) WCST (trials to complete the first category) Age Educational level (years)
0.22* 0.28** 0.23* −0.18 0.006 −0.06 −0.27** 0.27**
0.03 0.06 0.11 0.06 0.11 −0.13 −0.03 0.10
Patients DFT (free response condition) DFT (fixed response condition) TOL (total score) WCST (perseverative errors) WCST (categories completed) WCST (trials to complete the first category) Age Education (years) Length of illness (years) BPRSd total BPRS positive BPRS negative CPZeq (mg)e FNA-Cf On APPg On traditional APh only On atypical AP only On ACMi
0.23* 0.29** 0.30** −0.28** 0.29** −0.29** −0.25** 0.13 −0.23* −0.07 0.003 0.10 −0.17 0.16 −0.09 0.001 0.15 −0.24*
0.41** 0.34** 0.15 −0.20* 0.25** −0.23* −0.05 0.21* −0.08 −0.11 −0.18 0.04 −0.11 0.04 −0.18 −0.01 0.07 −0.25**
*p < 0.05; **p < 0.01. a DFT = Design Fluency Test. b TOL = Tower of London. c WCST = Wisconsin Card Sorting Test. d BPRS = Brief Psychotic Rating Scale. e CPZeq = Chlorpromazine equivalent (mg). f FNA-C = Functional Needs Assessment-Chinese version. g APP = Antipsychotic polypharmacy. h AP = Antipsychotic drug. i ACM = Anticholinergic medication.
account for this discrepancy. At this point it seems that psychopathology and length of illness (Kumar et al., 2005, 2007; Meissner et al., 2001) exert minimal influence of PM functions. Evidence is emerging that overall cognitive functions and social adaptation in schizophrenia are closely related (Addington & Addington, 2008). It was expected that deficits of PM in schizophrenia patients could possibly lead to disorganized ADL (Shum et al., 2004). Contrary to previous findings (Ritch et al., 2003; Twamley et al., 2007) however, in this study there was no significant negative correlation between PM and patients’ ADL. This absence of significant relationship could possibly be explained by the relative insensitivity of the FNA. The relationship between PM tasks and ADL needs to be further explored. In line with earlier results (Henry et al., 2007; Woods et al., 2007), better time-based PM performance was associated with younger age in this study. As expected, higher educational level was also significantly correlated with better time-based PM performance.
G.S. Ungvari et al. / Archives of Clinical Neuropsychology 23 (2008) 613–622
619
Table 3 Predictors of time-based PM task in the patient group by stepwise multiple regression analysis Beta Variables that entered the model (adjusted R2 = 0.17; F(3,106) = 8.93; p < 0.001) 0.25 TOLa (total score) Age −0.24 0.23 WCSTb (categories completed) Variables had failed to enter the model DFTc (free response condition) DFT (fixed response condition) WCST (perseverative errors) WCST (trials to complete the first category) Length of illness Sex On APPd On traditional APe only On atypical AP only On ACMf a b c d e f
T
p
2.88 −2.74 2.57
0.005 0.007 0.011
– –
1.01 1.69
0.31 0.09
– – – – – – – –
−0.98 0.10 0.72 0.07 −1.14 1.20 0.19 −1.82
0.32 0.92 0.47 0.94 0.25 0.23 0.84 0.07
TOL = Tower of London. WCST = Wisconsin Card Sorting Test. DFT = Design Fluency Test. APP = Antipsychotic polypharmacy. AP = Antipsychotic drug. ACM = Anticholinergic medication.
The second hypothesis that schizophrenia patients would show impairment in the two types of PM compared to HC was confirmed. The discrepancy between the two groups was not likely to be due to the patients’ inability to comprehend and remember the instructions of the PM tasks since all subjects could understand and recall the instructions and the details of the two PM tasks before and after the tasks. The general knowledge questions for the PM tasks were expected to place similar cognitive demand on patients and HC since the two groups were matched in terms of age and education. Yet, patients still showed a poorer performance on these questions. This is probably due to the combination of cognitive Table 4 Predictors of event-based PM task in the patient group by stepwise multiple regression analysis Beta Variables that entered the model (adjusted DFTa (fixed response condition) On ACMb WCSTc (perseverative errors)
R2
= 0.20; F(3,106) = 9.83; p < 0.001) 0.26 −0.22 −0.20
Variables had failed to enter the model DFT (free response condition) WCST (categories completed) WCST (trials to complete the first category) Sex On APPd On traditional APe only On atypical AP only Educational level a b c d e
DFT = Design Fluency Test. ACM = Anticholinergic medication. WCST = Wisconsin Card Sorting Test. APP = Antipsychotic polypharmacy. AP = Antipsychotic drug.
T
p
2.87 −2.58 −2.16
0.005 0.011 0.033
–
1.43
0.15
– – – – – – –
0.21 −0.22 −0.25 −1.50 0.56 −0.23 1.46
0.83 0.81 0.79 0.13 0.57 0.81 0.14
620
G.S. Ungvari et al. / Archives of Clinical Neuropsychology 23 (2008) 613–622
deficits related to the illness, the effects of psychotropic drugs, as well as the impact of negative symptoms such as decreased interest and motivation. It is not yet clear what neuropsychological mechanisms account for schizophrenia patients’ poor performance on PM tasks. Woods et al.’s study (2007) found defective cue detection and self-initiated retrieval underlying event- and time-based PM impairment. For habitual PM tasks, which are particularly relevant from a practical point of view (e.g., regularly taking medication at certain intervals), deficient internal source monitoring seems to be an important contributing factor (Elvevag et al., 2003). Meissner et al. (2001) came to a similar conclusion emphasizing the deficits in strategic thinking underlying PM deficits. Another possible explanation put forward by Kumar et al. (2005) postulates that inhibitory mechanisms that suppress irrelevant thoughts are suspended in giving way to the interruption of the relevant cognitive processes including PM memory in schizophrenia (Hasher & Zacks, 1998). Twamley et al. (2007) reported that PM was positively associated with a number of cognitive domains including attention, working memory, processing speed, learning and executive functions. Henry et al. (2007) proposed that the unique deficit of PM in schizophrenia is a general difficulty mostly independent of other cognitive and executive functions. However, without including other illness groups (e.g., mood disorders, brain injury) as comparisons, the specificity of such impairment will remain a mere conjecture. Schizophrenia is associated with impairment in prefrontal lobe functions (e.g., Tan, Callocitt, & Weinberger, 2007). Time-based tasks place more demand on the prefrontal cortex (Einstein et al., 1995) requiring intact executive functions (Ritch et al., 2003). This is one possible explanation for the larger difference in performance on time-based task compared to the event-based task between patients and HC. These results are in line with our earlier findings (Shum et al., 2004) and those of other (Kumar et al., 2007; Wang et al., 2007), but not all (Woods et al., 2007) reports. The discrepancy between these studies may be due to the diversity of assessment instruments in terms of length, ongoing task, delay interval, and response modality used to measure PM and prefrontal cortex functions. An alternative explanation for the different performance on time- and event-based tasks could be that the event-based task in this study was simply less effortful thus requiring less involvement of executive functions. However, the two tasks in this and other studies (e.g., Shum, Valentine, & Cutmore, 1999) were comparable in all aspects (e.g., ongoing task, the interval between PM task encoding and performance) except for the PM cue (external vs. self-initiated). Yet, the performance on event-based task was consistently superior to the time-based one in all studies which argues for the important, if not exclusive, role of the executive functions’ integrity. The third hypothesis that performance on frontal lobe functions would be significantly correlated with performance on PM tasks was also supported confirming our previous findings (Shum et al., 2004) in a considerably larger sample of patients. The possibility of a causal link between PM and frontal lobe dysfunctions was discussed in our previous report (Shum et al., 2004). As no large scale studies have assessed PM and other cognitive functions simultaneously in various patient groups, the connection between PM and executive functions related to intact prefrontal cortex remains a hypothesis requiring further research. 5.1. Strengths and limitations of the study and future directions of research The strengths of the study are its large sample size, matched HC, standardized PM assessment tools and the independent assessment of the various aspects of the investigation. The study included all available schizophrenia patients who were hospitalized during the study period avoiding major selection bias. However, the results should be interpreted with caution because of methodological limitations. First, the study examined Chinese subjects with chronic, clinically stable schizophrenia thus the results may not be applicable to patients of other ethnicity or at other stages of their illness. Second, a potential selection bias was the exclusion of illiterate and more severely ill patients receiving high-doses of psychotropic drugs who could not cooperate. Third, retrospective memory and other aspects of cognitive functions were not assessed. Subjects’ cognitive capacity was judged only by clinical impression and their ability to recall the PM tasks before and after the testing. Fourth, the study adopted a cross-sectional design; the causality of relationships is, therefore, tentative. Fifth, participants were required to remove their wristwatch during PM tasks, which might have limited the ecological validity of the time-based test. Sixth, to avoid or minimize any possible retroactive interference in PM tasks, other types of assessments should have been arranged between the two PM tasks. In future studies more detailed and sensitive instruments measuring psychopathology and social functions should be employed. In addition to HC, control groups comprising patients with other psychiatric and neurological disorders
G.S. Ungvari et al. / Archives of Clinical Neuropsychology 23 (2008) 613–622
621
should be included. Future studies should also include comprehensive cognitive assessment combining with neuroimaging investigations. This approach coupled with a longitudinal design would allow researchers to determine if there is a characteristic signature of PM deficit in schizophrenia and may shed light on the dynamics of its development. Acknowledgements This study was sponsored by a Hong Kong Research Grants Council Earmarked Research Grant. The authors are grateful to the assistance of Mr. Jacky Yuen in data collection and to the late Dr. Jin Pang Leung for his contribution in setting up the study. References Addington, J., & Addington, D. (2008). Social and cognitive functioning in psychosis. Schizophrenia Research, 99, 176–181. APA. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington: American Psychiatric Association. Brebion, G., Bressan, R. A., Amador, X., Malaspina, D., & Gorman, J. M. (2004). Medications and verbal memory impairment in schizophrenia: The role of anticholinergic drugs. Psychological Medicine, 34, 369–374. Dombrowski, S. B., Kane, M., Tuttle, N. B., & Kincaid, W. (1990). Functional needs assessment program for chronic psychiatric patients. Tucson, Arizona: Communication Skill Builders. Einstein, G. O., & McDaniel, M. A. (1990). Normal aging and prospective memory. Journal of Experimental Psychology-Learning Memory and Cognition, 16, 717–726. Einstein, G. O., McDaniel, M. A., Richardson, S. L., Guynn, M. J., & Cunfer, A. R. (1995). Aging and prospective memory: Examining the influences of self-initiated retrieval processes. Journal of Experimental Psychology-Learning Memory and Cognition, 21, 996–1007. Elvevag, B., Maylor, E. A., & Gilbert, A. L. (2003). Habitual prospective memory in schizophrenia. BMC Psychiatry, 3, 9. Hasher, L., & Zacks, R. T. (1998). Working memory dysfunction and its relevance to schizophrenia. In G. Bower (Ed.), The psychology of learning and motivation (pp. 193–225). New York: Academic Press. Heaton, R. K., Chelune, G. J., Talley, J. L., Kay, G. G., & Curtis, G. (1993). Wisconsin Card Sorting Test (WCST) Manual: Revised and expanded. Henry, J. D., Rendell, P. G., Kliegel, M., & Altgassen, M. (2007). Prospective memory in schizophrenia: Primary or secondary impairment? Schizophrenia Research, 95, 179–185. Jones-Gotman, M., & Milner, B. (1977). Design fluency: The invention of nonsense drawings after focal cortical lesions. Neuropsychologia, 15, 653–674. Kliegel, M., McDaniel, M. A., & Einstein, G. O. (2007). Prospective memory: Cognitive neuroscience, developmental and applied perspectives. Mahwah, NJ: Lawrence Erlbaum. Kondel, T. K. (2002). Prospective memory and executive function in schizophrenia. Brain and Cognition, 48, 405–410. Kumar, D., Nizamie, H. S., & Jahan, M. (2005). Event-based prospective memory in schizophrenia. Journal of Clinical and Experimental Neuropsychology, 27, 867–872. Kumar, D., Nizamie, S. H., & Jahan, M. (2007). Activity-based prospective memory in schizophrenia. Clinical Neuropsychology, 1–10. Kvavilashvili, L., & Ellis, J. (1996). Varieties of intention: Some distinctions and classification. In M. Brandimonte, G. O. Einstein, & M. A. McDaniel (Eds.), Prospective memory: Theory and applications (pp. 23–52). Mahwah, NJ: Lawrence Erlbaum. Law, K. M. (1999). Validation of functional needs assessment (Chinese Version) for people with chronic schizophrenia. Unpublished MPhil Thesis. Hong Kong, China: Hong Kong Polytechnic University. Lindenmayer, J. P., Khan, A., Iskander, A., Abad, M. T., & Parker, B. (2007). A randomized controlled trial of olanzapine versus haloperidol in the treatment of primary negative symptoms and neurocognitive deficits in schizophrenia. Journal of Clinical Psychiatry, 68, 368–379. McDaniel, M. A., & Einstein, G. O. (2007). Prospective memory. An overview and synthesis of an emerging field. Thousand Oaks, CA: Sage Publications. Meissner, F., Hacker, W., & Heilemann, H. (2001). Memory performance and instruction effects in schizophrenia. A comparative study of chronic schizophrenic patients and healthy probands (in German). Psychiatrische Praxis, 28, 180–188. O’Grada, C., & Dinan, T. (2007). Executive function in schizophrenia: What impact do antipsychotics have? Human Psychopharmacology, 22, 397–406. Ritch, J. L., Velligan, D. I., Tucker, D., Dicocco, M., & Maples, N. J. (2003). Prospective memory in schizophrenia. Schizophrenia Research, 60, 180. Shallice, T. (1982). Specific impairments of planning. Philosophical Transactions of the Royal Society of London Series B, Biological Science, 298, 199–209. Shen, Y. C. (2002). Psychiatry. Beijing: People’s Medical Publishing House., pp. 734–735. Shum, D., Ungvari, G. S., Tang, W. K., & Leung, J. P. (2004). Performance of schizophrenia patients on time-, event-, and activity-based prospective memory tasks. Schizophrenia Bulletin, 30, 693–701. Shum, D., Valentine, M., & Cutmore, T. (1999). Performance of individuals with severe long-term traumatic brain injury on time-, event-, and activity-based prospective memory tasks. Journal of Clinical Neuropsychology, 21, 49–58. Tan, H. Y., Callocitt, J. H., & Weinberger, D. R. (2007). Dysfunctional and compensatory prefrontal cortical systems, genes and the pathogenesis of schizophrenia. Cerebral Cortex, 17(Suppl 1), 171–181. Taylor, D., Paton, C., & Kerwin, R. (2005). The Maudsley Prescribing Guidelines (2005–2006). Abingdon: Taylor & Francis.
622
G.S. Ungvari et al. / Archives of Clinical Neuropsychology 23 (2008) 613–622
Tunstall, J. (1999). Improving the utility of the Tower of London: A neuropsychological test of planning. Unpublished MPhil thesis. Brisbane, Australia: Griffith University. Twamley, E. W., Woods, S. P., Zurhellen, C. H., Vertinski, M., Narvaez, J. M., Mausbach, B. T., et al. (2007). Neuropsychological substrates and everyday functioning implications of prospective memory impairment in schizophrenia. Schizophrenia Research, doi:10.1016/j.schres.2007.10.030 Wang, Y., Chan, R. C., Xin, Y., Shi, C., Cui, J., & Deng, Y. (2007). Prospective memory deficits in subjects with schizophrenia spectrum disorders: A comparison study with schizophrenic subjects, psychometrically defined schizotypal subjects, and healthy controls. Schizophrenia Research, doi:10.1016/j.schres.2007.07020 West, R. (2007). Towards a cognitive neuroscience of prospective memory. In M. Kliegel, M. A. McDaniel, & G. O. Einstein (Eds.), Prospective memory: Cognitive neuroscience developmental and applied perspectives (pp. 125–145). Mahwah, NJ: Lawrence Erlbaum. Woods, S. P., Twamley, E. W., Dawson, M. S., Narvaez, J. M., & Jeste, D. V. (2007). Deficits in cue detection and intention retrieval underlie prospective memory impairment in schizophrenia. Schizophrenia Research, 90, 344–350. Xiang, Y. T., Weng, Y. Z., Leung, C. M., Tang, W. K., & Ungvari, G. S. (2008). Subjective quality of life in outpatients with schizophrenia in Hong Kong and Beijing: Relationship to socio-demographic and clinical factors. Quality of Life Research, 17, 27–36.