Abstract
BackgroundAlthough cognition is a core symptom of schizophrenia and associated with functional impairment, the degree of training for and time associated with its assessment makes it difficult to routinely monitor in clinic care.Smartphone based cognitive assessments could serve as a tool to measure cognition in real time as well as being easily scalable for broad use.Combined with other data gathered from smartphone sensors such as steps, sleep, and self-reported symptoms – capturing ‘cognition in context’ could provide a powerful new tool for assessing the functional burden of disease in schizophrenia. Methods18 participants with schizophrenia and 17 healthy controls completed novel cognitive assessments on their personal smartphones over the course of 12 weeks while also capturing self-reported surveys and step count. No payment or incentives were offered for engaging with the smartphone app. Differing levels of difficulty in cognitive tasks were tested and the results were modeled using a modified Cox proportional hazard model. ResultsOn the smartphone cognitive assessments that involved on simple patterns, both controls and those with schizophrenia achieved similar scores. On the more complex assessment that added task switching in addition to pattern recognition, those with schizophrenia achieved scores lower than controls. Collecting other forms of data such as surveys and steps was also feasible using the same smartphone platform. DiscussionIt is feasible for those with schizophrenia to use their own smartphones to complete cognitive assessments and other measures related to their mental health. While we did not investigate the correlations between these cognitive assessments and other smartphone captured metrics like step count or self-reported symptoms, the potential to longitudinally assess cognition in the context of patients' environments outside of the clinic presents unique opportunities for characterizing cognitive burden in schizophrenia.
Highlights
Schizophrenia is among the most disabling disorders in all of medicine (Haller et al, 2014)
Using an open source and freely available smartphone tools developed by our team (Torous et al, 2016), we have previously demonstrated that it is feasible to capture real time smartphone sensor and survey data from patients with schizophrenia to model sleep (Staples et al, 2017) and functional outcomes like relapse (Torous et al, 2018a)
Eighteen participants with schizophrenia in active treatment were recruited from an outpatient state mental health clinic in Boston and 17 healthy controls were recruited from local colleges to participate in this study (Table 1)
Summary
Schizophrenia is among the most disabling disorders in all of medicine (Haller et al, 2014). Current constructs for clinically assessing cognition produce accurate and valid results but require expert clinical raters and lengthy assessment times (Reichenberg, 2010) These current constructs offer only a single measurement in time that cannot capture the complex temporal dynamics of systems that influence cognition across patients' daily lives. Rather than ignoring this heterogeneity and complexity with a more reductionist view of cognitive deficits in schizophrenia being largely static, we propose it is possible to quantify them through dynamic multimodal modeling. While we did not investigate the correlations between these cognitive assessments and other smartphone captured metrics like step count or self-reported symptoms, the potential to longitudinally assess cognition in the context of patients' environments outside of the clinic presents unique opportunities for characterizing cognitive burden in schizophrenia
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
