The efficacy of computerized cognitive drill and practice training for patients with a schizophrenia-spectrum disorder: A meta-analysis

Abstract

Background

Computerized methods for improving cognitive functioning in schizophrenia have gained popularity during the past decades. Therefore, this study evaluates the available evidence for the efficacy of computerized cognitive drill and practice training for patients with schizophrenia-spectrum disorders.

Methods

A systematic search was carried out using PubMed, Embase, Cochrane Database of Systematic Reviews, and PsycINFO. A meta-analysis was performed to compare cognitive drill and practice training in patients with a schizophrenia-spectrum disorder with non-cognitively oriented control conditions. The primary outcome was cognitive functioning. Secondary outcome measures included psychotic symptoms, depressive symptoms, and functional outcomes. Effect sizes (ES) for all included studies were calculated as Hedges' g.

Results

24 studies were included with 1262 patients in total. Compared to a control condition, patients receiving computerized cognitive drill and practice training showed significantly more improvement on attention (ES = 0.31, p = 0.001), working memory (ES = 0.38, p < 0.001), positive symptoms (ES = 0.31, p = 0.003), and depressive symptoms (ES = 0.37, p = 0.002). Small, marginally significant effect sizes were found for processing speed, verbal and visual learning and memory, and verbal fluency. However, significant effects on functional outcomes and social cognition were absent.

Discussion

The current study showed evidence for the efficacy of computerized cognitive drill and practice training in patients with schizophrenia-spectrum disorders. However, the absence of effects on social cognition and functional outcomes questions the generalization of treatment effects. Together, these results stimulate further development of computerized training programs for schizophrenia that not only improve cognitive functioning, but also generalize cognitive improvement to functional outcomes.

Introduction

During the past decades, an increasing amount of computerized programs have been developed to enhance cognitive performance, such as Nintendo's ‘Brain age’ or PositScience's ‘BrainHQ’. These programs use a principle of repeated practice of tasks (i.e., drill and practice training) to improve cognitive functioning. Interestingly, this principle has also been adopted in the design of computerized cognitive rehabilitation programs for patients with schizophrenia-spectrum disorders (e.g., CogPack and CogRehab). These computer programs have numerous advantages, such as the possibility to automatically adjust practice levels, standardization of instructions, and the possibility to perform the training with only little help from therapist, thereby reducing costs (Burda et al., 1994). However, evidence for the efficacy of such programs in improving (cognitive) functioning remains unclear.

In general, cognitive functioning in patients with schizophrenia-spectrum disorders is between one and two standard deviations below normal and includes dysfunction in the domains of attention/vigilance, reasoning and problem solving, working memory, processing speed, verbal learning and memory, visual learning and memory, and social cognition (Kern et al., 2011; Nuechterlein et al., 2004). Antipsychotic treatment, the main pharmacological treatment for schizophrenia, has a minimal effect on these symptoms (Hill et al., 2010; Nielsen et al., 2015). As cognitive deficits are related to decreased daily functioning and work performance, improvement of cognitive functioning is extremely important for patients´ quality of life (Bowie and Harvey, 2006; Green et al., 2000; McGurk et al., 2007). When developing treatment programs for schizophrenia, the presence of amotivation has to be taken into account, as this is one of the symptoms of the disorder (American Psychiatric Association, 2013). The use of gaming elements in computerized cognitive training might help overcome this problem, as it can make a treatment program increasingly meaningful, appealing, motivating, and feasible (Fleming et al., 2017; Lau et al., 2017; Medalia et al., 2001; Stapleton and Taylor, 2003).

Various studies have investigated the efficacy of drill and practice exercises in the treatment of cognitive deficits in schizophrenia to evaluate whether such programs have potential for this patient population, with inconsistent findings. Importantly, drill and practice training differs from strategic training, in which the goal is to improve performance by explicitly learning and applying cognitive strategies, such as mnemonics (Wykes et al., 2011). Previous meta-analyses showed that cognitive training can improve cognitive functioning in patients with schizophrenia-spectrum disorders, but they did not include computerized training specifically (McGurk et al., 2007; Wykes et al., 2011). In 2011 a meta-analysis was published that did focus on computerized training only and also showed positive results (Grynszpan et al., 2011). To our knowledge, the previous meta-analyses included both drill and practice and strategic training and coaching (Grynszpan et al., 2011; McGurk et al., 2007; Wykes et al., 2011). Importantly, as computerized training programs and games gain popularity, we aim to assess the efficacy of computerized drill and practice cognitive remediation on neurocognitive functioning in schizophrenia. Furthermore, we aim to assess generalization of cognitive improvement to other domains: social cognition, psychotic and depressive symptoms, and functional outcomes.