Hostname: page-component-cd9895bd7-8ctnn Total loading time: 0 Render date: 2024-12-25T00:59:27.706Z Has data issue: false hasContentIssue false
  • English
  • Français

Predictors of within-individual variability in cognitive performance in schizophrenia in a South African case–control study

Published online by Cambridge University Press:  21 June 2023

Olivia Wootton Open the ORCID record for Olivia Wootton [Opens in a new window] ,
Shareefa Dalvie ,
Rae MacGinty ,
Linda Ngqengelele ,
Ezra S. Susser ,
Ruben C. Gur  and
Dan J. Stein
Olivia Wootton*
Affiliation:
Department of Psychiatry and Neuroscience Institute, University of Cape Town, Cape Town, South Africa
Shareefa Dalvie
Affiliation:
UCT MRC Genomic and Precision Medicine Research Unit, Division of Human Genetics, Institute of Infectious Diseases and Molecular Medicine, University of Cape Town, Cape Town, South Africa
Rae MacGinty
Affiliation:
Department of Paediatrics and Child Health, University of Cape Town, Cape Town, South Africa
Linda Ngqengelele
Affiliation:
Department of Psychiatry and Neuroscience Institute, University of Cape Town, Cape Town, South Africa
Ezra S. Susser
Affiliation:
Department of Epidemiology, Mailman School of Public Health, Columbia University, New York, NY, USA New York State Psychiatric Institute, New York, NY, USA
Ruben C. Gur
Affiliation:
Brain Behavior Laboratories, Department of Psychiatry, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA, USA
Dan J. Stein
Affiliation:
Department of Psychiatry and Neuroscience Institute, University of Cape Town, Cape Town, South Africa SAMRC Unit on Risk & Resilience in Mental Disorders, Cape Town, South Africa
*
Corresponding author: Olivia Wootton; Email: [email protected]
Rights & Permissions [Opens in a new window]

Abstract

Introduction:

Cognitive dysfunction in schizophrenia may be assessed by measuring within-individual variability (WIV) in performance across a range of cognitive tests. Previous studies have found increased WIV in people with schizophrenia, but no studies have been conducted in low- to middle-income countries where the different sociocultural context may affect WIV. We sought to address this gap by exploring the relationship between WIV and a range of clinical and demographic variables in a large study of people with schizophrenia and matched controls in South Africa.

Methods:

544 people with schizophrenia and 861 matched controls completed an adapted version of The University of Pennsylvania Computerized Neurocognitive Battery (PennCNB). Demographic and clinical information was collected using the Structured Clinical Interview for DSM-IV Diagnoses. Across-task WIV for performance speed and accuracy on the PennCNB was calculated. Multivariate linear regression was used to assess the relationship between WIV and a diagnosis of schizophrenia in the whole sample, and WIV and selected demographic and clinical variables in people with schizophrenia.

Results:

Increased WIV of performance speed across cognitive tests was significantly associated with a diagnosis of schizophrenia. In people with schizophrenia, increased speed WIV was associated with older age, a lower level of education and a lower score on the Global Assessment of Functioning scale. Increased accuracy WIV was significantly associated with a younger age in people with schizophrenia.

Conclusions:

Measurements of WIV of performance speed can add to the knowledge gained from studies of cognitive dysfunction in schizophrenia in resource-limited settings.

Keywords

SchizophreniaCognition disordersNeuropsychological testsHumansWithin-individual variability
Type
Original Article
Information
Acta Neuropsychiatrica , First View , pp. 1 - 7
Check for updates
Creative Commons
Creative Common License - CCCreative Common License - BYCreative Common License - NCCreative Common License - ND
This is an Open Access article, distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives licence (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided that no alterations are made and the original article is properly cited. The written permission of Cambridge University Press must be obtained prior to any commercial use and/or adaptation of the article.
Copyright
© The Author(s), 2023. Published by Cambridge University Press on behalf of Scandinavian College of Neuropsychopharmacology

Significant outcomes

  • Increased WIV across speed measures was associated with schizophrenia and poorer functional outcomes in the disorder, while no such effects were observed for WIV in accuracy.

  • This finding provides support for the use of measures of speed WIV in future studies exploring the biological basis of cognitive dysfunction in schizophrenia.

  • The finding of a negative relationship between speed WIV and educational attainment provides further evidence for the positive association between years of education and cognitive function, whether measured by WIV or another metric.

Limitations

  • There were substantially more males than females in this cohort, which made accurate assessment of the relationship between sex and WIV challenging.

  • A detailed rating scale was not used to assess symptom severity in schizophrenia which may have limited the analysis of the relationship between symptom severity and WIV.

Introduction

Cognitive dysfunction is characteristic of schizophrenia (Saykin et al., Reference Saykin, Gur, Gur, Mozley, Mozley, Resnick, Kester and Stafiniak1991; Vinogradov et al., Reference Vinogradov, Poole, Willis-Shore, Ober and Shenaut1998; Rentrop et al., Reference Rentrop, Rodewald, Roth, Simon, Walther, Fiedler, Weisbrod and Kaiser2010; Cole et al., Reference Cole, Weinberger and Dickinson2011; Roalf et al., Reference Roalf, Gur, Almasy, Richard, Gallagher, Prasad, Wood, Pogue-Geile, Nimgaonkar and Gur2013a). Within-individual variability (WIV) in cognitive functioning reflects the extent of the variation in an individual’s performance relative to their mean performance, and it has been suggested that WIV provides additional insight into cognitive function over and above mean-based performance measures (MacDonald et al., Reference Macdonald, Li and Bäckman2009; Roalf et al., Reference Roalf, Gur, Almasy, Richard, Gallagher, Prasad, Wood, Pogue-Geile, Nimgaonkar and Gur2013a). WIV indexes risk of psychosis in the general population and is a predictor of functional outcomes in schizophrenia (Vinogradov et al., Reference Vinogradov, Poole, Willis-Shore, Ober and Shenaut1998; Wexler et al., Reference Wexler, Nicholls and Bell2004; Reichenberg et al., Reference Reichenberg, Weiser, Rapp, Rabinowitz, Caspi, Schmeidler, Knobler, Lubin, Nahon, Harvey and Davidson2006; Shin et al., Reference Shin, Kim, Shin, Jung, Hur, Byun, Jang, An and Kwon2013; Wallace and Linscott, Reference Wallace and Linscott2018). Additionally, elevations in WIV have been observed in individuals with high risk of developing psychosis when mean measures of performance are normal (Shin et al., Reference Shin, Kim, Shin, Jung, Hur, Byun, Jang, An and Kwon2013).

WIV may be operationalised in three ways: 1) across-tasks in the same testing session, 2) across-trials of the same task during the same testing session and 3) across-trials of the same task during different testing sessions (MacDonald et al., Reference Macdonald, Li and Bäckman2009; Cole et al., Reference Cole, Weinberger and Dickinson2011). Across-trials and across-task WIV are positively correlated (Hultsch et al., Reference Hultsch, MacDonald and Dixon2002), and elevations in both measures of WIV have been demonstrated in people with schizophrenia (Reichenberg et al., Reference Reichenberg, Weiser, Rapp, Rabinowitz, Caspi, Schmeidler, Knobler, Lubin, Nahon, Harvey and Davidson2006; Cole et al., Reference Cole, Weinberger and Dickinson2011; Roalf et al., Reference Roalf, Gur, Almasy, Richard, Gallagher, Prasad, Wood, Pogue-Geile, Nimgaonkar and Gur2013a; Fassbender et al., Reference Fassbender, Scangos, Lesh and Carter2014). Although it is typical to calculate WIV using measures of performance speed, WIV for performance accuracy may also be used to assess cognitive function (MacDonald et al., Reference Macdonald, Li and Bäckman2009; Roalf et al., Reference Roalf, Gur, Almasy, Richard, Gallagher, Prasad, Wood, Pogue-Geile, Nimgaonkar and Gur2013a; Roalf et al., Reference Roalf, Gur, Ruparel, Calkins, Satterthwaite, Bilker, Hakonarson, Harris and Gur2014a). Previous studies have found elevated across-task WIV for performance speed and accuracy in people with schizophrenia, reflecting an inability to maintain performance across tasks assessing different neurocognitive domains (Roalf et al., Reference Roalf, Gur, Almasy, Richard, Gallagher, Prasad, Wood, Pogue-Geile, Nimgaonkar and Gur2013a; Roalf et al., Reference Roalf, Ruparel, Verma, Elliott, Gur and Gur2013b). Although the neural correlates of WIV require further research, abnormalities of brain white matter and hypoactivity of the dorsolateral prefrontal cortex have been associated with elevated WIV in schizophrenia (Roalf et al., Reference Roalf, Ruparel, Verma, Elliott, Gur and Gur2013b; Fassbender et al., Reference Fassbender, Scangos, Lesh and Carter2014; Panagiotaropoulou et al., Reference Panagiotaropoulou, Thrapsanioti, Pappa, Grigoras, Mylonas, Karavasilis, Velonakis, Kelekis and Smyrnis2019; Ahn et al., Reference Ahn, Yu, Sung, Choi, Lee, Bang and Lee2019). Increased WIV has also been demonstrated in disorders such as attention-deficit hyperactivity disorder (Leth-Steensen et al., Reference Leth-Steensen, Elbaz and Douglas2000; Lin et al., Reference Lin, Hwang-Gu and Gau2015), traumatic brain injury (Stuss et al., Reference Stuss, Murphy, Binns and Alexander2003) and dementia (Holtzer et al., Reference Holtzer, Verghese, Wang, Hall and Lipton2008; Halliday et al., Reference Halliday, Stawski, Cerino, Decarlo, Grewal and MacDonald2018; Webber et al., Reference Webber, Kiselica, Mikula and Woods2022), including human immunodeficiency virus (HIV)-associated neurocognitive disorder (Vance et al., Reference Vance, Del Bene, Frank, Billings, Triebel, Buchholz, Rubin, Woods, W. and Fazeli2021). Further, factors such as ageing, male sex and lower level of education have been associated with increased WIV (Hilborn et al., Reference Hilborn, Strauss, Hultsch and Hunter2009; Cole et al., Reference Cole, Weinberger and Dickinson2011; Roalf et al., Reference Roalf, Gur, Ruparel, Calkins, Satterthwaite, Bilker, Hakonarson, Harris and Gur2014a; De Felice and Holland, Reference De Felice and Holland2018).

Despite increasing interest in WIV as a measure of cognitive dysfunction, it has only been studied in a limited number of settings. To our knowledge, no studies of WIV in schizophrenia have been conducted in low- and-middle income countries (LMICs) where differences in educational levels, the prevalence of comorbid conditions (such as HIV) and treatment practices may affect the relationship between WIV and specific disorders. To promote the development of contextually relevant diagnostic and treatment practices, there is a need to assess the generalisability of findings from neuropsychological research on schizophrenia from high-income settings to other contexts. Additionally, the association between symptom severity and WIV in schizophrenia remains unclear as results have been inconsistent with some studies finding a positive association and others finding no relationship (Wexler et al., Reference Wexler, Nicholls and Bell2004; Pellizzer and Stephane, Reference Pellizzer and Stephane2007; Rentrop et al., Reference Rentrop, Rodewald, Roth, Simon, Walther, Fiedler, Weisbrod and Kaiser2010; Shin et al., Reference Shin, Kim, Shin, Jung, Hur, Byun, Jang, An and Kwon2013; Roalf et al., Reference Roalf, Ruparel, Verma, Elliott, Gur and Gur2013b; Akiyama et al., Reference Akiyama, Saito, Saito, Ozeki, Watanabe, Fujii, Honda and Shimoda2016).

Here, we aim to address a gap in current knowledge by examining the relationship between across-task WIV and schizophrenia in a large case–control study from South Africa. Data were originally collected for a genetic study on schizophrenia but includes information on a range of cognitive, demographic and clinical variables. Thus, the study is uniquely placed to explore the relationship between 1) WIV and schizophrenia, 2) WIV and demographic variables (age, sex and level of education), and 3) WIV and clinical variables (HIV status, substance use, symptom severity and functional outcomes) in a LMIC.

Materials and methods

Participants

Patients with schizophrenia and matched controls underwent neuropsychological testing as part of the Genomics of Schizophrenia in the South African Xhosa people (SAX) study (Gulsuner et al., Reference Gulsuner, Stein, Susser, Sibeko, Pretorius, Walsh, Majara, Mndini, Mqulwana, Ntola, Casadei, Ngqengelele, Korchina, van der Merwe, Malan, Fader, Feng, Willoughby, Muzny, Baldinger, Andrews, Gur, Gibbs, Zingela, Nagdee, Ramesar, King and McClellan2020). The SAX study is a case–control study that aimed to identify genetic variants and social exposures contributing to schizophrenia risk in the Xhosa population of South Africa. The Xhosa people are one of the largest black African groups in South Africa and live primarily in the Eastern and Western Cape provinces of the country. The SAX study enrolled 2,849 individuals that self-identified as Xhosa from community healthcare clinics and psychiatric hospitals in the Western and Eastern Cape of South Africa between January 2013 and February 2018. Eligible cases were individuals aged 21–60 years with a diagnosis of schizophrenia or schizoaffective disorder for at least 2 years, with the capacity to provide informed consent for the genomics study. Controls were matched to cases for age, sex and level of education and were individuals presenting to the same community healthcare clinics as cases for non-psychiatric medical conditions.

Instruments and measures

A translated Xhosa language version of the Structured Diagnostic Interview for DSM-IV Axis I Disorders (SCID-I) (First and Gibbon, Reference First and Gibbon2004) was administered to all study participants by trained psychiatric nurses. As previously reported, inter-rater reliability for the Xhosa version of the SCID-I was adequate for the principle psychotic disorder (κ = 0.74, p < .001) (Mall et al., Reference Mall, Platt, Temmingh, Musenge, Campbell, Susser and Stein2020). Where possible, information from medical records, family members and healthcare professionals was incorporated and considered in the diagnostic process. Global functioning was assessed using the Global Assessment of Functioning (GAF) scale (Aas, Reference Aaa2011).

An adapted 10-test version of the Pennsylvania Computerized Neurocognitive Battery (PennCNB) was administered to 544 cases and 861 controls (n = 1,405) in isiXhosa. The PennCNB is designed to assess cognitive domains related to underlying brain networks through the application of “neurobehavioral probes” that have been validated with functional neuroimaging (Gur et al., Reference Gur, Erwin and Gur1992; Roalf et al., Reference Roalf, Ruparel, Gur, Bilker, Gerraty, Elliott, Gallagher, Almasy, Pogue-Geile, Prasad, Wood, Nimgaonkar and Gur2014b). The PennCNB has demonstrated adequate psychometric properties in a number of settings (Gur et al., Reference Gur, Ragland, Moberg, Turner, Bilker, Kohler, Siegel and Gur2001; Gur et al., Reference Gur, Richard, Hughett, Calkins, Macy, Bilker, Brensinger and Gur2010; Moore et al., Reference Moore, Reise, Gur, Hakonarson and Gur2015; Service et al., Reference Service, Vargas Upegui, Castaño Ramírez, Port, Moore, Munoz Umanes, Agudelo Arango, Díaz-Zuluaga, Melo Espejo, López, Palacio, Ruiz Sánchez, Valencia, Teshiba, Espinoza, Olde Loohuis, de la Hoz Gomez, Brodey, Sabatti, Escobar, Reus, Lopez Jaramillo, Gur, Bearden and Freimer2020; Scott et al., Reference Scott, Moore, Stein, Pretorius, Zingela, Nagdee, Ngqengelele, Campbell, Sibeko, King, Mcclellan, Port, Jackson, Ruparel, Susser and Gur2021), and evidence for the validity of the Xhosa version of the CNB is presented in Scott et al. (Reference Scott, Moore, Stein, Pretorius, Zingela, Nagdee, Ngqengelele, Campbell, Sibeko, King, Mcclellan, Port, Jackson, Ruparel, Susser and Gur2021). The tests included in the Xhosa version of the battery were designed to measure domains of neurocognition and social cognition that are known to be affected in SZ (Scott et al., Reference Scott, Moore, Stein, Pretorius, Zingela, Nagdee, Ngqengelele, Campbell, Sibeko, King, Mcclellan, Port, Jackson, Ruparel, Susser and Gur2021). Descriptions of the English versions of the tests have been published. Briefly, the Xhosa version of the battery assesses five neurocognitive domains: 1) executive function, 2) episodic memory, 3) complex cognition, 4) social cognition and 5) sensorimotor speed. Table 1 lists the 10 tests included in the Xhosa version of the battery and the neurocognitive domains that they are designed to assess. Measures of speed and accuracy are calculated for each test except for the Motor Praxis Test and Computerized Finger Tapping Test for which only a measure of speed is obtained.

Table 1. Neurocognitive domains and tests included in the Xhosa version of the PennCNB

Ethics

All participants provided signed informed consent. To assess capacity to consent, The University of California, San Diego Brief Assessment of Capacity to Consent Questionnaire (UBACC) was administered to all participants (Jeste et al., Reference Jeste, Palmer, Appelbaum, Golshan, Glorioso, Dunn, Kim, Meeks and Kraemer2007; Campbell et al., Reference Campbell, Susser, Mall, Mqulwana, Mndini, Ntola, Nagdee, Zingela, Van Wyk and Stein2017). The study was approved by the University of Cape Town Human Research Ethics Committee (reference number – 049/2013).

Data analysis

Demographic and clinical characteristics of the cases and controls that completed the PennCNB were compared using t-tests and Pearson’s chi-square tests for continuous and categorical variables, respectively. An index of across-task WIV was calculated for each participant who completed five or more cognitive tests (n = 1,088) using methodology described in previous studies (Holtzer et al., Reference Holtzer, Verghese, Wang, Hall and Lipton2008; Roalf et al., Reference Roalf, Ruparel, Verma, Elliott, Gur and Gur2013b). First, speed and accuracy measures for each test were z-transformed based on the distribution of the entire sample. Second, across-task WIV for both speed and accuracy was calculated using the following equation:

$$ {\rm WIV}=\sqrt{\sum _{k=1}^{K}{\left({Z_{ik}}-{A_{i}}\right)^{2} \over \left(K-1\right)}} $$

where Z ik is the kth test score for the ith individual and:

$$ A_{i}=\sum _{k=1}^{K}{Z_{ik} \over K} $$

is the individual’s mean z-transformed test score based on all completed cognitive tests in a battery.

For both speed and accuracy WIV, the data were skewed to the right and heteroscedasticity was present. WIV was log-transformed to an approximately normal distribution, and the natural log of WIV was taken as the dependent variable in further analyses. First, the relationship between a diagnosis of schizophrenia and speed and accuracy WIV was assessed using multivariate linear regression in the whole sample. Age, sex, level of education, case–control status, HIV status and history of substance use were included as covariates in the analysis. Second, multivariate linear regression was conducted using cases only with WIV as a dependent variable and age, sex, level of education, HIV status, history of substance use, GAF score, and symptom severity as independent variables. All statistical analyses were performed using IBM SPSS version 28.0.1.0.

Results

Sample characteristics

Of the cases, 96.5% (n = 525) had a principal axis 1 diagnosis of schizophrenia. The term schizophrenia will be used to describe the diagnosis of all cases in the remainder of the text. The demographic and clinical characteristics of the sample are summarised in Table 2. Controls were significantly younger than cases (p = .004). Level of education and sex did not differ significantly between the two groups. There was a significantly higher number of controls with a diagnosis of HIV (p < .001). Street drug use differed significantly between the two groups with a higher number of cases reporting a history of street drug use (p < .001). Scores on the GAF scale were significantly higher amongst controls (p < .001).

Table 2. Sample demographic and clinical characteristics (n = 1,405)

GAF, Global Assessment of Functioning Scale; HIV, human immunodeficiency virus.

*p < 0.05, **p < 0.01, ***p < 0.001.

WIV and schizophrenia

WIV for accuracy and speed measures on the PennCNB is presented in Fig. 1. A multivariate linear regression was used to assess the relationship between WIV for speed and accuracy measures on the PennCNB and a diagnosis of schizophrenia in the entire sample. There was a significant relationship between speed WIV and a diagnosis of schizophrenia (β = 0.255, p < .001), but there was no significant relationship between accuracy WIV and a diagnosis of schizophrenia (β = 0.021, p = .51).

Figure 1. Means (+ standard error of the mean) for across-task within-individual variability for speed and accuracy in matched controls and people with schizophrenia (cases). Higher variability scores are indicative of worse overall performance.

Predictors of WIV in schizophrenia

Amongst cases, speed WIV was significantly associated with age, level of education, and GAF (Table 3). There was a negative association between level of education and speed WIV (β = −0.11, p = .035). A higher GAF score was associated with a lower speed WIV (β = −0.17, p = .002). Symptom severity, substance use, HIV status and sex did not have a significant relationship with speed WIV. Age was positively correlated with WIV for speed (β = 0.22, p < .001) but contrastingly was negatively correlated with WIV for accuracy (β = −0.13, p = .022). Age was the only significant predictor of accuracy WIV in people with schizophrenia (Table 4).

Table 3. Linear regression analysis of selected predictor variables and WIV speed amongst people with schizophrenia (n = 353)

B, unstandardized regression coefficient; β, standardized regression coefficient; CI, confidence interval; GAF, Global Assessment of Functioning Scale; HIV, human immunodeficiency virus; SE, standard error.

Constant = −0.159, F (7,345) = 6.649, p < 0.001***, adj. R 2 = 0.119.

*p < 0.05, **p < 0.01, ***p < 0.001.

Table 4. Linear regression analysis of selected predictor variables and WIV accuracy amongst people with schizophrenia (n = 353)

B, unstandardized regression coefficient; β, standardized regression coefficient; CI, confidence interval; GAF, Global Assessment of Functioning Scale; HIV, human immunodeficiency virus; SE, standard error.

Constant = −0.185, F (7,345) = 1.720, p = 0.103, adj. R 2 = 0.014.

*p < 0.05, **p < 0.01, ***p < 0.001.

Discussion

In this first study of WIV in cognitive performance in schizophrenia in a LMIC, we provide novel insights into the relationship between WIV and several clinical and demographic variables in this setting. The main findings of this study were as follows: 1) increased speed WIV was significantly associated with a diagnosis of schizophrenia; 2) reduced speed WIV was associated with a higher level of educational attainment and better functional outcomes among people with schizophrenia; and 3) among people with schizophrenia, speed and accuracy WIV were significantly associated with age but with opposite directions of effect, and the magnitude of effect was much larger for speed than accuracy.

The finding of a significant relationship between a diagnosis of schizophrenia and speed WIV in the analysis of the entire sample is in keeping with previous research, which has demonstrated increased WIV in speed measures for cognitive tasks in people with schizophrenia (Roalf et al., Reference Roalf, Gur, Almasy, Richard, Gallagher, Prasad, Wood, Pogue-Geile, Nimgaonkar and Gur2013a; Roalf et al., Reference Roalf, Ruparel, Verma, Elliott, Gur and Gur2013b; Fassbender et al., Reference Fassbender, Scangos, Lesh and Carter2014). However, in contrast with previous studies, we did not find a significant relationship between accuracy WIV and schizophrenia (Roalf et al., Reference Roalf, Gur, Almasy, Richard, Gallagher, Prasad, Wood, Pogue-Geile, Nimgaonkar and Gur2013a; Roalf et al., Reference Roalf, Ruparel, Verma, Elliott, Gur and Gur2013b). These results may suggest that WIV in speed measures is a more sensitive marker of cognitive dysfunction in schizophrenia in this and possibly other LMIC settings. In the analysis restricted to people with schizophrenia, we found an inverse relationship between speed WIV and global functioning. Converging evidence from multiple studies indicates that reduced WIV in reaction time is a predictor of better functional outcomes in schizophrenia (Vinogradov et al., Reference Vinogradov, Poole, Willis-Shore, Ober and Shenaut1998; Wexler et al., Reference Wexler, Nicholls and Bell2004; Rentrop et al., Reference Rentrop, Rodewald, Roth, Simon, Walther, Fiedler, Weisbrod and Kaiser2010). Although abnormal elevations in WIV have been observed in other neuropsychiatric disorders, measures of speed WIV may be of potential use in clinical practice to identify people with schizophrenia at risk of poor functional outcomes and, thus, inform treatment decisions. Further, the results of this study support the significance of measures of speed WIV in our understanding of schizophrenia. Measures of speed WIV may be used in addition to or as an alternative to mean performance measures in future studies exploring the biological mechanisms contributing to cognitive impairment and functional outcomes in schizophrenia.

Cognitive function is an important determinant of socio-economic attainment (e.g. occupation and income), and physical and mental health outcomes in the general population as well as in people with schizophrenia (Kalechstein et al., Reference Kalechstein, Newton and Van Gorp2003; Strenze, Reference Strenze2007; Fioravanti et al., Reference Fioravanti, Bianchi and Cinti2012; Batty et al., Reference Batty, Deary and Zaninotto2016; Ozawa et al., Reference Ozawa, Laing, Higgins, Yemeke, Park, Carlson, Ko, Guterman and Omer2022). Evidence from a recent meta-analysis supports a positive association between educational attainment and cognitive ability (Ritchie and Tucker-Drob, Reference Ritchie and Tucker-Drob2018). Schizophrenia is associated with lower educational attainment in both high-income and low- or middle-income countries (Crossley et al., Reference Crossley, Alliende, Czepielewski, Aceituno, Castañeda, Diaz, Iruretagoyena, Mena, Mena, Ramirez-Mahaluf, Tepper, Vasquez, Fonseca, Machado, Hernández, Vargas-Upegui, Gomez-Cruz, Kobayashi-Romero, Moncada-Habib, Arango, Barch, Carter, Correll, Freimer, McGuire, Evans-Lacko, Undurraga, Bressan, Gama, Lopez-Jaramillo, de la Fuente-Sandoval, Gonzalez-Valderrama, Undurraga and Gadelha2022), and therapeutic guidelines recommend interventions to improve educational outcomes in schizophrenia (National Institute for Health and Care Excellence [NICE], 2014; Early Psychosis Guidelines Writing Group and EPPIC National Support Program, 2016; Norman et al., Reference Norman, Lecomte, Addington and Anderson2017). This study found that higher educational attainment was associated with lower speed WIV amongst people with schizophrenia. This finding provides further support for the positive impact of educational attainment on cognitive function, whether measured by WIV or another metric. This observation emphasises the relevance of interventions to increase educational attainment, and ultimately improve cognitive function, in people with schizophrenia in LMICs.

In this study, increased age was significantly associated with both greater accuracy and lower speed WIV amongst people with schizophrenia. In studies of WIV across the lifespan in the general population, increasing age after early adulthood is associated with a progressive increase in WIV in cognitive performance (Hultsch et al., Reference Hultsch, MacDonald and Dixon2002; MacDonald et al., Reference Macdonald, Li and Bäckman2009; Roalf et al., Reference Roalf, Gur, Ruparel, Calkins, Satterthwaite, Bilker, Hakonarson, Harris and Gur2014a). In this study, the finding of a positive relationship between age and speed WIV provides support for a similar relationship between speed WIV and age amongst people with schizophrenia. Contrastingly, age had a negative relationship with accuracy WIV. Although this finding was significant, the model predicting accuracy WIV explained a low amount of variance in accuracy WIV, and age had a small regression coefficient. Thus, the finding should be interpreted with caution and requires replication in other studies. However, that age effects are more pronounced on speed than on accuracy measures on cognitive tests has been established in multiple studies, including those using the PennCNB (Gur et al., Reference Gur, Richard, Hughett, Calkins, Macy, Bilker, Brensinger and Gur2010; Irani et al., Reference Irani, Brensinger, Richard, Calkins, Moberg, Bilker, Gur and Gur2012; Moore et al., Reference Moore, Gur, Thomas, Brown, Nock, Savitt, Keilp, Heeringa, Ursano and Stein2019).

There are certain limitations that need to be considered when interpreting the results of the study. There were substantially more males than females in this cohort, which made accurate assessment of the relationship between sex and WIV challenging. The small number of females may have contributed to the null relationship between sex and WIV in this study when previous research has demonstrated sex differences (Roalf et al., Reference Roalf, Gur, Ruparel, Calkins, Satterthwaite, Bilker, Hakonarson, Harris and Gur2014a). Future studies should aim at recruiting more female participants. Similarly, the lack of association between HIV status and WIV must be interpreted with caution as the number of HIV-positive individuals enrolled in the study was limited. Further, additional clinically relevant variables, such as CD4 count and HIV viral load, should be collected in future studies in order to allow for a more nuanced analysis of the relationship between WIV and HIV. While the inclusion of participants with HIV made it possible to assess the relationship between HIV and WIV in persons with schizophrenia, this may have affected the assessment of the relationship between WIV and a diagnosis of schizophrenia. As HIV is associated with cognitive impairment (Keng et al., Reference Keng, Winston and Sabin2023), future studies aiming to focus on the relationship between WIV and schizophrenia may consider excluding individuals with a diagnosis of HIV. Lastly, the assessment of symptom severity in schizophrenia in this study was limited as it was derived from the SCID and a valid and reliable specific measure, such as the Positive and Negative Syndrome Scale for Schizophrenia (Kay et al., Reference Kay, Fiszbein and Opler1987), would be better suited to assess this relationship.

In conclusion, the results of this study extend current knowledge of WIV in cognitive performance in schizophrenia by exploring its relationship with multiple clinical and demographic variables in a resource-limited setting. Generally, increased speed WIV was associated with schizophrenia and poorer functioning in the disorder, while no such effects were observed for WIV in accuracy. This finding suggests that WIV for speed measures might be a more sensitive marker of cognitive impairment in people with schizophrenia who can still maintain performance accuracy across neuropsychological tasks. The finding of an inverse relationship between educational attainment and speed WIV in this setting provided further evidence for the positive association between education and evening out cognitive function. The significant relationship between speed WIV and schizophrenia as well as global functioning in the disorder highlight the potential for future research into WIV to extend our understanding of cognitive impairment in schizophrenia.

Acknowledgements

We are grateful to the participants who gave their time to make the SAX study possible. We wish to acknowledge the support of Dr Adele Pretorius, project manager of the SAX study, as well as the SAX data teams for their assistance with data management, quality control and access.

Funding statement

This work was supported by the National Institute of Mental Health (NIMH: Grant number U01MH125053). The SAX study was funded by the NIMH (Grant number: 5UO1MH096754). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of Mental Health.

Competing interests

OW, SD, RM, LN, ESS and RCG have no conflicts of interest to declare. DJS is supported by the Medical Research Council of South Africa and has received research grants and/or consultancy honoraria from Abbott, Astrazeneca, Discovery Vitality, Eli-Lilly, GlaxoSmithKline, Jazz Pharmaceuticals, Johnson & Johnson, Lundbeck, Orion, Pfizer, Pharmacia, Roche, Sanofi, Servier, Solvay, Sumitomo, Sun, Takeda, Tikvah, Vistagen, and Wyeth.

Ethical standard

The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008.

References

Aaa, IH (2011) Guidelines for rating Global Assessment of Functioning (GAF). Annals of General Psychiatry 10, 2.Google Scholar
Ahn, JI, Yu, ST, Sung, G, Choi, TK, Lee, KS, Bang, M and Lee, SH (2019) Intra-individual variability in neurocognitive function in schizophrenia: relationships with the corpus callosum. Psychiatry Research: Neuroimaging 283, 1–6.CrossRefGoogle ScholarPubMed
Akiyama, K, Saito, S, Saito, A, Ozeki, Y, Watanabe, T, Fujii, K, Honda, G and Shimoda, K (2016) Predictive value of premorbid IQ, negative symptoms, and age for cognitive and social functions in Japanese patients with schizophrenia: a study using the Japanese version of the Brief Assessment of Cognition in Schizophrenia. Psychiatry Research 246, 663–671.CrossRefGoogle ScholarPubMed
Batty, GD, Deary, IJ and Zaninotto, P (2016) Association of cognitive function with cause-specific mortality in middle and older age: follow-up of participants in the English longitudinal study of ageing. American Journal of Epidemiology 183(3), 183–190.CrossRefGoogle ScholarPubMed
Campbell, MM, Susser, E, Mall, S, Mqulwana, SG, Mndini, MM, Ntola, OA, Nagdee, M, Zingela, Z, Van Wyk, S, Stein, DJ (2017) Using iterative learning to improve understanding during the informed consent process in a South African psychiatric genomics study. PLoS One 12, e0188466.CrossRefGoogle Scholar
Cole, VT, Weinberger, DR and Dickinson, D (2011) Intra-individual variability across neuropsychological tasks in schizophrenia: a comparison of patients, their siblings, and healthy controls. Schizophrenia Research 129(1), 91–93.CrossRefGoogle ScholarPubMed
Crossley, NA, Alliende, LM, Czepielewski, LS, Aceituno, D, Castañeda, CP, Diaz, C, Iruretagoyena, B, Mena, C, Mena, C, Ramirez-Mahaluf, JP, Tepper, A, Vasquez, J, Fonseca, L, Machado, V, Hernández, CE, Vargas-Upegui, C, Gomez-Cruz, G, Kobayashi-Romero, LF, Moncada-Habib, T, Arango, C, Barch, DM, Carter, C, Correll, CU, Freimer, NB, McGuire, P, Evans-Lacko, S, Undurraga, E, Bressan, R, Gama, CS, Lopez-Jaramillo, C, de la Fuente-Sandoval, C, Gonzalez-Valderrama, A, Undurraga, J and Gadelha, A (2022) The enduring gap in educational attainment in schizophrenia according to the past 50 years of published research: a systematic review and meta-analysis. Lancet Psychiatry 9, 565–573.CrossRefGoogle Scholar
De Felice, S and Holland, CA (2018) Intra-individual variability across fluid cognition can reveal qualitatively different cognitive styles of the aging brain. Frontiers in Psychology 9, 1973.CrossRefGoogle ScholarPubMed
Early Psychosis Guidelines Writing Group And Eppic National Support Program (2016). Australian Clinical Guidelines for Early Psychosis, Orygen. Melbourne: The National Centre of Excellence in Youth Mental Health.Google Scholar
Fassbender, C, Scangos, K, Lesh, TA and Carter, CS (2014) RT distributional analysis of cognitive-control-related brain activity in first-episode schizophrenia. Cognitive, Affective & Behavioral Neuroscience 14(1), 175–188.CrossRefGoogle ScholarPubMed
Fioravanti, M, Bianchi, V and Cinti, ME (2012) Cognitive deficits in schizophrenia: an updated metanalysis of the scientific evidence. BMC Psychiatry 12, 64.CrossRefGoogle ScholarPubMed
First, MB and Gibbon, M (2004) The structured clinical interview for DSM-IV Axis I Disorders (SCID-i) and the structured clinical interview for DSM-IV Axis II Disorders (SCID-II), Comprehensive Handbook of Psychological Assessment, Vol. 2: Personality assessment, Hoboken, NJ, USA: John Wiley & Sons, Inc. Google Scholar
Gulsuner, S, Stein, DJ, Susser, ES, Sibeko, G, Pretorius, A, Walsh, T, Majara, L, Mndini, MM, Mqulwana, SG, Ntola, OA, Casadei, S, Ngqengelele, LL, Korchina, V, van der Merwe, C, Malan, M, Fader, KM, Feng, M, Willoughby, E, Muzny, D, Baldinger, A, Andrews, HF, Gur, RC, Gibbs, RA, Zingela, Z, Nagdee, M, Ramesar, RS, King, M-C, McClellan, JM (2020) Genetics of schizophrenia in the South African Xhosa. Science 367(6477), 569–573.CrossRefGoogle ScholarPubMed
Gur, RC, Erwin, RJ and Gur, RE (1992) Neurobehavioral probes for physiologic neuroimaging studies. Archives of General Psychiatry 49(5), 409–414.CrossRefGoogle ScholarPubMed
Gur, RC, Ragland, JD, Moberg, PJ, Turner, TH, Bilker, WB, Kohler, C, Siegel, SJ and Gur, RE (2001) Computerized neurocognitive scanning: I. Methodology and validation in healthy people. Neuropsychopharmacology 25(5), 766–776.CrossRefGoogle ScholarPubMed
Gur, RC, Richard, J, Hughett, P, Calkins, ME, Macy, L, Bilker, WB, Brensinger, C and Gur, RE (2010) A cognitive neuroscience-based computerized battery for efficient measurement of individual differences: standardization and initial construct validation. Journal of Neuroscience Methods 187(2), 254–262.CrossRefGoogle ScholarPubMed
Halliday, DWR, Stawski, RS, Cerino, ES, Decarlo, CA, Grewal, K and MacDonald, SWS (2018) Intraindividual variability across neuropsychological tests: dispersion and disengaged lifestyle increase risk for Alzheimer’s disease. Journal of Intelligence 6(1), 12.CrossRefGoogle ScholarPubMed
Hilborn, JV, Strauss, E, Hultsch, DF and Hunter, MA (2009) Intraindividual variability across cognitive domains: investigation of dispersion levels and performance profiles in older adults. Journal of Clinical and Experimental Neuropsychology 31(4), 412–424.CrossRefGoogle ScholarPubMed
Holtzer, R, Verghese, J, Wang, C, Hall, CB and Lipton, RB (2008) Within-person across-neuropsychological test variability and incident dementia. JAMA 300(7), 823–830.CrossRefGoogle ScholarPubMed
Hultsch, DF, MacDonald, SW and Dixon, RA (2002) Variability in reaction time performance of younger and older adults. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences 57(2), P101–15.CrossRefGoogle ScholarPubMed
Irani, F, Brensinger, CM, Richard, J, Calkins, ME, Moberg, PJ, Bilker, W, Gur, RE and Gur, RC (2012) Computerized neurocognitive test performance in schizophrenia: a lifespan analysis. The American Journal of Geriatric Psychiatry 20(1), 41–52.CrossRefGoogle ScholarPubMed
Jeste, DV, Palmer, BW, Appelbaum, PS, Golshan, S, Glorioso, D, Dunn, LB, Kim, K, Meeks, T and Kraemer, HC (2007) A new brief instrument for assessing decisional capacity for clinical research. Archives of General Psychiatry 64(8), 966–974.CrossRefGoogle ScholarPubMed
Kalechstein, AD, Newton, TF and Van Gorp, WG (2003) Neurocognitive functioning is associated with employment status: a quantitative review. Journal of Clinical and Experimental Neuropsychology 25(8), 1186–1191.CrossRefGoogle ScholarPubMed
Kay, SR, Fiszbein, A and Opler, LA (1987) The Positive and Negative Syndrome Scale (PANSS) for schizophrenia. Schizophrenia Bulletin 13(2), 261–276.CrossRefGoogle ScholarPubMed
Keng, LD, Winston, A and Sabin, CA (2023) The global burden of cognitive impairment in people with HIV. AIDS 37(1), 61–70.CrossRefGoogle ScholarPubMed
Leth-Steensen, C, Elbaz, ZK and Douglas, VI (2000) Mean response times, variability, and skew in the responding of ADHD children: a response time distributional approach. Acta Psychologica 104(2), 167–190.CrossRefGoogle ScholarPubMed
Lin, HY, Hwang-Gu, SL and Gau, SS (2015) Intra-individual reaction time variability based on ex-Gaussian distribution as a potential endophenotype for attention-deficit/hyperactivity disorder. Acta Psychiatrica Scandinavica 132(1), 39–50.CrossRefGoogle ScholarPubMed
Macdonald, SW, Li, SC and Bäckman, L (2009) Neural underpinnings of within-person variability in cognitive functioning. Psychology and Aging 24(4), 792–808.CrossRefGoogle ScholarPubMed
Mall, S, Platt, JM, Temmingh, H, Musenge, E, Campbell, M, Susser, E and Stein, DJ (2020) The relationship between childhood trauma and schizophrenia in the Genomics of Schizophrenia in the Xhosa people (SAX) study in South Africa. Psychological Medicine 50(9), 1570–1577.CrossRefGoogle ScholarPubMed
Moore, TM, Gur, RC, Thomas, ML, Brown, GG, Nock, MK, Savitt, AP, Keilp, JG, Heeringa, S, Ursano, RJ, Stein, MB (2019) Development, administration, and structural validity of a brief, computerized neurocognitive battery: results from the army study to assess risk and resilience in servicemembers. Assessment 26(1), 125–143.CrossRefGoogle ScholarPubMed
Moore, TM, Reise, SP, Gur, RE, Hakonarson, H and Gur, RC (2015) Psychometric properties of the Penn Computerized Neurocognitive Battery. Neuropsychology 29(2), 235–246.CrossRefGoogle ScholarPubMed
National Institute for Health and Care Excellence [NICE] (2014) Psychosis and Schizophrenia in Adults: Prevention and Management,Google Scholar
Norman, R, Lecomte, T, Addington, D and Anderson, E (2017) Canadian treatment guidelines on psychosocial treatment of schizophrenia in adults. The Canadian Journal of Psychiatry 62(9), 617–623.CrossRefGoogle ScholarPubMed
Ozawa, S, Laing, SK, Higgins, CR, Yemeke, TT, Park, CC, Carlson, R, Ko, YE, Guterman, LB and Omer, SB (2022) Educational and economic returns to cognitive ability in low- and middle-income countries: a systematic review. World Development 149, 105668.CrossRefGoogle ScholarPubMed
Panagiotaropoulou, G, Thrapsanioti, E, Pappa, E, Grigoras, C, Mylonas, D, Karavasilis, E, Velonakis, G, Kelekis, N and Smyrnis, N (2019) Hypo-activity of the dorsolateral prefrontal cortex relates to increased reaction time variability in patients with schizophrenia. NeuroImage: Clinical 23, 101853.CrossRefGoogle ScholarPubMed
Pellizzer, G and Stephane, M (2007) Response selection in schizophrenia. Experimental Brain Research 180(4), 705–714.CrossRefGoogle ScholarPubMed
Reichenberg, A, Weiser, M, Rapp, MA, Rabinowitz, J, Caspi, A, Schmeidler, J, Knobler, HY, Lubin, G, Nahon, D, Harvey, PD, Davidson, M (2006) Premorbid intra-individual variability in intellectual performance and risk for schizophrenia: a population-based study. Schizophrenia Research 85(1-3), 49–57.CrossRefGoogle ScholarPubMed
Rentrop, M, Rodewald, K, Roth, A, Simon, J, Walther, S, Fiedler, P, Weisbrod, M and Kaiser, S (2010) Intra-individual variability in high-functioning patients with schizophrenia. Psychiatry Research 178(1), 27–32.CrossRefGoogle ScholarPubMed
Ritchie, SJ and Tucker-Drob, EM (2018) How much does education improve intelligence? A meta-analysis. Psychological Science 29(8), 1358–1369.CrossRefGoogle ScholarPubMed
Roalf, DR, Gur, RC, Almasy, L, Richard, J, Gallagher, RS, Prasad, K, Wood, J, Pogue-Geile, MF, Nimgaonkar, VL, Gur, RE (2013a) Neurocognitive performance stability in a multiplex multigenerational study of schizophrenia. Schizophrenia Bulletin 39(5), 1008–1017.CrossRefGoogle Scholar
Roalf, DR, Gur, RE, Ruparel, K, Calkins, ME, Satterthwaite, TD, Bilker, WB, Hakonarson, H, Harris, LJ and Gur, RC (2014a) Within-individual variability in neurocognitive performance: age- and sex-related differences in children and youths from ages 8 to 21. Neuropsychology 28(4), 506–518.CrossRefGoogle ScholarPubMed
Roalf, DR, Ruparel, K, Gur, RE, Bilker, W, Gerraty, R, Elliott, MA, Gallagher, RS, Almasy, L, Pogue-Geile, MF, Prasad, K, Wood, J, Nimgaonkar, VL, Gur, RC (2014b) Neuroimaging predictors of cognitive performance across a standardized neurocognitive battery. Neuropsychology 28(2), 161–176.CrossRefGoogle ScholarPubMed
Roalf, DR, Ruparel, K, Verma, R, Elliott, MA, Gur, RE and Gur, RC (2013b) White matter organization and neurocognitive performance variability in schizophrenia. Schizophrenia Research 143(1), 172–178.CrossRefGoogle ScholarPubMed
Saykin, AJ, Gur, RC, Gur, RE, Mozley, PD, Mozley, LH, Resnick, SM, Kester, DB and Stafiniak, P (1991) Neuropsychological function in schizophrenia. Selective impairment in memory and learning. Archives of General Psychiatry 48(7), 618–624.CrossRefGoogle ScholarPubMed
Scott, JC, Moore, TM, Stein, DJ, Pretorius, A, Zingela, Z, Nagdee, M, Ngqengelele, L, Campbell, M, Sibeko, G, King, MC, Mcclellan, JM, Port, AM, Jackson, C, Ruparel, K, Susser, ES, Gur, RC (2021) Adaptation and validation of a computerized neurocognitive battery in the Xhosa of South Africa. Neuropsychology 35(6), 581–594.CrossRefGoogle ScholarPubMed
Service, SK, Vargas Upegui, C, Castaño Ramírez, M, Port, AM, Moore, TM, Munoz Umanes, M, Agudelo Arango, LG, Díaz-Zuluaga, AM, Melo Espejo, J, López, MC, Palacio, JD, Ruiz Sánchez, S, Valencia, J, Teshiba, TM, Espinoza, A, Olde Loohuis, L, de la Hoz Gomez, J, Brodey, BB, Sabatti, C, Escobar, JI, Reus, VI, Lopez Jaramillo, C, Gur, RC, Bearden, CE and Freimer, NB (2020) Distinct and shared contributions of diagnosis and symptom domains to cognitive performance in severe mental illness in the Paisa population: a case-control study. Lancet Psychiatry 7(5), 411–419.CrossRefGoogle ScholarPubMed
Shin, YS, Kim, SN, Shin, NY, Jung, WH, Hur, J-W, Byun, MS, Jang, JH, An, SK and Kwon, JS (2013) Increased intra-individual variability of cognitive processing in subjects at risk mental state and schizophrenia patients. PloS One 8(11), e78354.CrossRefGoogle ScholarPubMed
Strenze, T (2007) Intelligence and socioeconomic success: a meta-analytic review of longitudinal research. Intelligence 35(5), 401–426.CrossRefGoogle Scholar
Stuss, DT, Murphy, KJ, Binns, MA and Alexander, MP (2003) Staying on the job: the frontal lobes control individual performance variability. Brain 126(11), 2363–2380.CrossRefGoogle ScholarPubMed
Vance, DE, Del Bene, VA, Frank, JS, Billings, R, Triebel, K, Buchholz, A, Rubin, LH, Woods, SP, W., LI, Fazeli, PL (2021) Cognitive intra-individual variability in HIV: an integrative review. Neuropsychology Review 32(4), 855–876.CrossRefGoogle ScholarPubMed
Vinogradov, S, Poole, JH, Willis-Shore, J, Ober, BA and Shenaut, GK (1998) Slower and more variable reaction times in schizophrenia: what do they signify? Schizophrenia Research 32(3), 183–190.CrossRefGoogle ScholarPubMed
Wallace, S and Linscott, RJ (2018) Intra-individual variability and psychotic-like experiences in adolescents: findings from the ALSPAC cohort. Schizophrenia Research 195, 154–159.CrossRefGoogle ScholarPubMed
Webber, TA, Kiselica, AM, Mikula, C and Woods, SP (2022) Dispersion-based cognitive intra-individual variability in dementia with Lewy bodies. Neuropsychology 36(8), 719–729.CrossRefGoogle ScholarPubMed
Wexler, BE, Nicholls, SS and Bell, MD (2004) Instability of cognitive processing systems in schizophrenia. Schizophrenia Research 71(2-3), 513–514.CrossRefGoogle ScholarPubMed
Figure 0

Table 1. Neurocognitive domains and tests included in the Xhosa version of the PennCNB

Figure 1

Table 2. Sample demographic and clinical characteristics (n = 1,405)

Figure 2

Figure 1. Means (+ standard error of the mean) for across-task within-individual variability for speed and accuracy in matched controls and people with schizophrenia (cases). Higher variability scores are indicative of worse overall performance.

Figure 3

Table 3. Linear regression analysis of selected predictor variables and WIV speed amongst people with schizophrenia (n = 353)

Figure 4

Table 4. Linear regression analysis of selected predictor variables and WIV accuracy amongst people with schizophrenia (n = 353)