Relationship Between Objectively Measured Sedentary Behavior and Cognitive Performance in Patients With Schizophrenia Vs Controls

Abstract

Objective:

Sedentary behavior (SB) is associated with poor cognitive performance in the general population. Although people with schizophrenia are highly sedentary and experience marked cognitive impairments, no study has investigated the relationship between SB and cognition in people with schizophrenia.

Methods:

A total of 199 inpatients with schizophrenia (mean [SD] age 44.0 [9.9] years, 61.3% male, mean [SD] illness duration 23.8 [6.5]) and 60 age and sex matched controls were recruited. Sedentary behavior and physical activity (PA) were captured for 7 consecutive days with an accelerometer. Cognitive performance was assessed using the Vienna Test System, and the Grooved Pegboard Test. Multivariate regression analyses adjusting for important confounders including positive and negative symptoms, illness duration, medication, and PA were conducted.

Results:

The 199 patients with schizophrenia engaged in significantly more SB vs controls (581.1 (SD 127.6) vs 336.4 (SD 107.9) min per day, P < .001) and performed worse in all cognitive performance measures (all P < .001). Compared to patients with high levels of SB (n = 89), patients with lower levels of SB (n = 110) had significantly (P < .05) better motor reaction time and cognitive processing. In the fully adjusted multivariate analysis, SB was independently associated with slower motor reaction time (β = .162, P < .05) but not other cognitive outcomes. Lower levels of PA were independently associated with worse attention and processing speed (P < .05).

Conclusion:

Our data suggest that higher levels of sedentary behavior and physical inactivity are independently associated with worse performance across several cognitive domains. Interventions targeting reductions in SB and increased PA should be explored.

Introduction

Schizophrenia is a chronic psychiatric disorder with a lifetime prevalence of 1% with a heterogeneous genetic and neurobiological background that influences early brain development, and is expressed as a combination of psychotic symptoms—such as hallucinations, delusions, and disorganization.¹ The burgeoning evidence base has established that schizophrenia is associated with considerable cognitive impairment.² Specifically, research has demonstrated that schizophrenia is associated with deficits across multiple domains including attention, processing speed, working memory, verbal learning and memory, and executive functions.³ Cognitive dysfunction is evident in those with the at risk mental state,^4,5 experiencing their first episode⁶ and cognitive deficits in schizophrenia endure over the course of illness.⁷ Understanding factors that are associated with cognitive performance are important, because cognitive deficits are associated with increased healthcare costs,⁸ lower quality of life, and impaired recovery.⁹ A range of factors have been associated with cognitive performance in schizophrenia including genetic predisposition,¹⁰ hypofunction of N-methyl-d-aspartate (NMDA) receptors,¹¹ negative symptoms, sociodemographic and illness factors (eg, male gender, illness duration).¹² Most recently, novel research has suggested that physical activity appears to be associated with better cognitive performance and in particular attention/concentration.¹³

Physical activity is defined as any bodily movement produced by skeletal muscles that results in energy expenditure.¹⁴ Physical activity sits on a spectrum, ranging from low intensity physical activity (eg, light walking) to high intensity activity (eg, fast running) and is categorized based on energy expenditure. At the opposite end of the spectrum, is sedentary behavior, which is defined as any waking activity characterized by an energy expenditure ≤1.5 metabolic equivalents.¹⁵ Examples of sedentary behaviors may include prolonged sitting or remaining in reclined posture during the day time. In the general population, there is an increasing evidence base to suggest that sedentary behavior is independent from physical activity associated with worse cognitive performance.¹⁶ Higher levels of sedentary behavior are associated with worse cognitive performance across the lifespan including in children¹⁷ and older adults.¹⁸ A recent systematic review of 8 studies concluded that higher levels of sedentary behavior were associated with worse cognitive function.¹⁶

Despite the growing interest in the potential deleterious impact of sedentary behavior in people with schizophrenia, no study has to our knowledge investigated the relationship with cognitive performance. Nonetheless, a recent meta-analysis¹⁹ established that people with schizophrenia engage in very high levels of sedentary behavior. To date, only 2 studies^20,21 have investigated the relationship between sedentary behavior and health outcomes in people with schizophrenia, and both established a negative relationship with cardiometabolic markers. Although helpful, Stubbs et al²⁰ and Vancampfort et al²¹ relied upon self-report sedentary behavior (which is prone to bias¹⁹) and did not include a control group. Only 5 published studies have utilized objective measures of sedentary behavior in people with schizophrenia,^22–26 of which, only one study included a control group,²⁴ and all included modest sample sizes (n < 50). There is also provisional evidence that higher light physical activity may be associated with better cognitive performance.¹³ It remains unclear if the relationship between physical activity and cognitive performance is independent from sedentary behavior and vice versa. Nonetheless, given the relationship between sedentary behavior and cognition in the general population, the high levels of sedentary behavior and cognitive deficits in people with schizophrenia, research is required to explore if sedentary behavior and low physical activity are associated with cognition in this population. In addition, it is important that research considers if a relationship between sedentary behavior and cognition would also be evident after adjusting for important confounders known to influence both factors such as negative symptoms.^25,27

The aims of the current study were to (1) investigate sedentary behavior levels in people with schizophrenia compared to controls using accelerometer, (2) investigate the relationship between sedentary behavior and cognitive performance in people with schizophrenia and compare this to controls, and (3) Explore the possible independent associations between sedentary behavior and low physical activity and cognitive performance in people with schizophrenia.

Methods

Participants

The current study recruited participants from the long stay psychiatric wards at Jianan Mental Hospital, Taiwan. Inclusion criteria were (1) diagnosis of schizophrenia (according to DSM IV made by an independent psychiatrist), (2) stable and on the same antipsychotic medicine for at least 3 months. Patients who were unable to communicate, walk independently or had any neurological disorders were excluded.

Healthy Control Group.

The healthy control group consisted of age-, gender-, and BMI-matched comparison participants recruited among the staff of 2 hospitals and universities. Eligibility criteria for controls included (1) no present or past history of any mental illness and (2) not taking any psychotropic medication. A total of 60 participants were selected to ensure comparable gender balance, age, and BMI ranges to the schizophrenia group.

The study was approved by the Institutional Review Board of Jianan Mental Hospital. All participants provided informed written consent.

Measures

Outcome Variable: Cognitive Function Assessments Vienna Test System (VTS).

All participants (including controls) completed the cognitive outcome measures. The VTS is a computerized neurocognitive function assessment consisting of various personality, intelligence, and psychomotor abilities.²⁸ The VTS has been used in previous studies,^29,30 including those with schizophrenia³¹ to determine cognitive performance. Specifically, within the current study, 2 subsets of the VTS were used: the Reaction test (RT) and Cognitrone test (COG). The RT measures the determination of reaction time to visual stimulus of individuals and the test last between 5 and 10min. Participants place their hands on the rest button, and are asked to push it as soon as possible when they detect a visual stimulus. The reaction and motor time is calculated in milliseconds (msec), which is the time between the start of the relevant stimulus and the moment the finger leaves the rest button.

The COG test captures the attention and concentration through the comparison of figures with regard to their congruence on a computer screen. Within the COG test, participants are presented with an abstract figure which they have to match to a model. The variable of the number of reactions made within the total working time of 7min provides information on the respondent’s processing speed.²⁸

Grooved Pegboard Test (GPT).

The GPT is a test of manual dexterity, upper-limb motor speed, hand-eye coordination, and speed of processing and has been widely used in previous research.^32–35 The GPT consists of 25 holes with randomly positioned slots. The pegs have a key along one side and must be put in the board in a fixed order and in the correct direction with only one hand being used. Participants are encouraged to perform the task as quickly as possible. Each participant is scored by the total time in seconds from when they starts the task until the last peg is put in place. For the purposes of the current study, each participant was tested once with their dominant and nondominant hand (ie, a total of 2 times) and the average was calculated to give an overall score.

Independent Variable: Sedentary Behavior Time.

Sedentary behavior was captured using the ActiGraph (wActiSleep), a tri-axial accelerometer. The ActiGraph has been validated previously among people with schizophrenia.^36–38 The measurement of sedentary behavior with accelerometers is the optimal free living measure in people with schizophrenia, with self-report measures such as the IPAQ lacking accuracy.³⁹ Standardized instructions of wearing an accelerometer were provided by research assistants. Participants were asked to wear the accelerometer on the wrist of the nondominant hand for 7 consecutive days and to remove it during bathing or water activities. The accelerometers were initialized, downloaded, and analyzed using ActiLife software version 6 (ActiGraph LLC). Sedentary behavior was defined according to the cut-off point outlined by Freedson⁴⁰ as activities ≦100 counts per minute (cpm), representing a threshold corresponding with sitting, reclining, or lying down. The sedentary behavior time was only measured during the waking day of each individual. The total energy expenditure of physical activity per week was also calculated by the software. Sedentary behavior and total physical activity were categorized into 2 levels by the mean: “low and high” in a binary split at the mean.

Covariates.

A range of sociodemographic information was collected including data on age (<40 or ≥40), sex, smoking habits, alcohol consumption, and education (<9 schooling years or ≥9 schooling years). Based on self-report answers, participants were categorized into 2 groups: “yes” (current/former smoker or drinker) and “no” (never). Body weight and height were measured and body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters (kg/m²). BMI classifications of obesity reflect risks for type 2 diabetes and cardiovascular diseases and varies among ethnic groups.⁴¹ Therefore, the national criteria of obesity in Taiwan were used to determine weight status (normal/underweight <24, overweight/obesity ≥24).⁴²

Metabolic Syndrome and Associated Metabolic Risk Factors.

Details regarding the metabolic syndrome (MetS) criteria and associated risk factors were collected including waist circumference, systolic/diastolic blood pressure (SBP/DBP), serum triglyceride (TG), high-density lipoprotein cholesterol (HDL-C), and fasting glucose (FG). Abnormalities in MetS were classified based on the following criteria: waist circumference ≥90cm in men and ≥80cm in women; SBP/DBP ≥130/85 mmHg; TG ≥150mg/dl; HDL-C <40mg/dl in men and <50mg/dl in women; and FG ≥100mg/dl. We used the number of abnormalities in MetS in the analyses and it requires meeting 3 or more of the five criteria to be diagnosed with MetS.

Positive and Negative Syndrome Scale.

All participants completed the Positive and Negative Syndrome Scale (PANSS), a tool specifically developed to assess the severity of symptoms and measure general psychopathology among patients with schizophrenia.⁴³ It is a 30-item rating scale, consisting of 3 subscales: 7-item Positive Symptoms, 7-item Negative Symptoms, and 16-item General Psychopathology. For each item, there are 7 rating points with increasing levels of psychopathology severity from 1 (asymptomatic) to 7 (extremely symptomatic). The PANSS score is the sum of ratings across items, with ranging from 7 to 49 for the Positive and Negative Scales and 16 to 112 for the General Psychopathology Scale. Higher scores indicate more severe symptoms.

Medication.

Information regarding antipsychotics medication and sleeping pills use was collected through the hospital records. Antipsychotics medication use was converted into a daily equivalent dosage of chlorpromazine⁴⁴ and the daily equipotent dosage of Lorazepam were also calculated for each patient according to the defined daily dose (DDD) of WHO Collaborating Centre for Drug Statistics Methodology (http://www.whocc.no/ddd/definition_and_general_considera/).

Statistical Analyses

Independent t-tests and chi-square tests were performed to compare the percentages or differences between levels of sedentary time in participants with schizophrenia. Differences of cognitive function, sedentary time, and physical activity energy expenditure between participants with and without schizophrenia were examined by independent t-tests. To examine the association of sedentary levels on cognition function, a 2-step forced entry multivariate regression analysis was performed for each cognitive test. In the multivariate linear regression, the dependent variable was sedentary behavior levels and variables entered into the model were demographic variables (age, sex, education), weight status, smoking, alcohol consumption, medications, PANSS, and MetS, followed by physical activity energy expenditure. All analyses were performed with IBM SPSS statistics 22 and a 2-tailed P-value less than .05 was considered as statistically significant.

Results

A total of 200 potential inpatients were identified as suitable for the study. Overall, 199 patients with schizophrenia gave informed consent and were recruited with a mean age of 44.0 years (SD 9.9), 61.3% were male with a mean illness duration of 23.8 years (SD 6.5). Sixty age and sex matched controls were also recruited (mean age 41.1 (9.6) years, 56.7% male).

Sedentary Behavior, Clinical, and Cognitive Outcomes in Patients

Patients with schizophrenia were divided into low levels of sedentary behavior (n = 110, <581.1min per day) and high levels of sedentary behavior (n = 89, ≥581.1min per day). Table 1 shows the demographic and clinical outcomes for both groups. Briefly, no difference in gender, age, or education was observed, although those who were more sedentary were less likely to smoke (table 1). No difference in psychiatric symptoms, chlorpromazine equivalents or illness duration was noted, although the less sedentary group had been admitted for longer. Patients who were more sedentary had worse motor reaction time and speed processing (P < .05) and engaged in less physical activity.

Table 1.

Characteristic Comparisons of Sedentary Levels in Patients With Schizophrenia

Variables % or Mean (SD)	Low Sedentary	High Sedentary	P Value
Age (%)			.322
<40	59.7	40.3
≥40	52.4	47.6
Sex (%)			.647
Female	53.2	46.8
Male	56.6	43.4
Education (%)			.973
High (≥9 years)	55.4	44.6
Low (<9 years)	55.1	44.9
Weight status (%)			.398
Normal (BMI < 24)	52.1	47.9
Obese (BMI ≥ 24)	58.1	41.9
Smoke (%)			.004
Never	46.5	53.5
Yes	67.1	32.9
Alcohol (%)			.500
Never	54.2	45.8
Yes	60.6	39.4
No. of meeting MetSa	1.8 (1.3)	1.7 (1.3)	.754
PANSSb	61.4 (20.1)	64.9 (20.0)	.209
PANSS_P	15.2 (6.8)	15.7 (4.0)	.529
PANSS_N	16.6 (5.8)	17.3 (6.6)	.459
PANSS_G	29.6 (10.4)	32.0 (11.0)	.114
Chlorpromazine equivalent doses (mg/d)	859.4 (856.0)	833.1 (688.6)	.815
Lorazepam equivalent doses (mg/d)	1.1 (1.4)	1.1 (1.1)	.886
Time since illness onset (year)	23.2 (7.1)	24.5 (5.8)	.193
Duration of hospitalization (month)	16.1 (19.5)	11.5 (13.1)	.047
Cognitive function
RTc _Reaction (msec)	669.5 (532.2)	652.3 (410.0)	.803
RTc _Motor (msec)	355.2 (170.8)	421.3 (252.7)	.037
COGd	176.9 (95.1)	165.8 (107.1)	.442
GPTe (sec)	131.6 (44.1)	145.4 (46.7)	.034
Physical activity (Kcal)	644.3 (384.3)	471.4 (294.4)	<.001

Note: Sedentary behavior total time was split at the mean to develop low levels of sedentary behavior (n = 110, <581.1min per day) and high levels of sedentary behavior (n = 89, ≥581.1min per day).

^aMetabolic syndrome.

^bPositive and Negative Syndrome Scale.

^cReaction time.

^dCognitrone: number of total reactions.

^eGrooved Pegboard Test.

Sedentary Behavior and Cognition in Patients and Controls

Patients with schizophrenia engaged in significantly more sedentary behavior vs controls (581.1min per day vs 336.4min per day, P < .001) equating to a mean difference of 244.6min per day or 4h and 7min per day. Patients with schizophrenia also engaged in less total physical activity (567.0 vs 904.1 Kcal per day, P < .001; mean difference 337.1) and had worse cognitive performance on all measures vs controls (all P < .001) (figure 1).

Fig. 1.

Cognition performance, sedentary time, and physical activity in participants with and without schizophrenia. All P < .001. GPT, Grooved Pegboard Test (total time); COG, Cognitrone test (number of total reactions); RT, Reaction time test (total time); Sedentary, sedentary time (time of sitting, reclining, or lying down per day excluding sleep); PA, physical activity (total energy expenditure of physical activity per week).

Exploring the Independent Relationships Between Sedentary Behavior and Low Physical Activity and Cognition

The multivariate associations between sedentary behavior and other outcomes with cognitive performance in the adjusted model with and without adjusting for physical activity are outlined in tables 2 and 3, respectively. Patients with schizophrenia who engaged in higher levels of sedentary behavior had a slower motor reaction time (β = .175, P = .026), while no relationship was observed with the Cognitrone or Grooved Pegboard Test. After adjusting for physical activity, the relationship between higher levels of sedentary behavior was independently associated with slower motor reaction times (β = .162, P = .042). Lower levels of physical activity were independently associated with worse attention and concentration (Cognitrone scores) (β = −.222, P = .008) and poorer processing speed and manual dexterity (GPT scores, β = .309, P < .001) in the fully adjusted model.

Table 2.

Multivariable Linear Regressions of Cognitive Function in People With Schizophrenia (Without Adjusting Physical Activity)

Variables	COG			GPT			RT-Reaction			RT-Motor
	R 2	Beta	P Value	R 2	Beta	P Value	R 2	Beta	P Value	R 2	Beta	P Value
	18.6			15.7			15.8			14.0
Age (≥40)a		−.235	.002		.201	.009		.083	.277		.121	.120
Sex (Male)a		.108	.248		−.108	.254		−.247	.010		−.184	.056
Education (low)a		−.029	.689		.076	.308		.045	.548		−.002	.982
Weight status (obese)a		.180	.027		−.096	.244		−.131	.111		−.054	.513
Smoke (yes)a		.031	.733		−.094	.305		.024	.797		.064	.488
Alcohol (yes)a		−.091	.247		.143	.074		.036	.650		−.122	.129
MetS		−.009	.907		.116	.151		.135	.094		.001	.994
PANSS		−.223	.010		.231	.008		.366	<.001		.228	.010
Chlorpromazine equivalent doses		.152	.044		.147	.056		.001	.992		−.134	.085
Lorazepam equivalent doses		−.143	.062		−.043	.578		.068	.382		.074	.345
Time since illness onset (year)		.038	.613		−.001	.993		−.045	.559		−.134	.085
Duration of hospitalization (month)		−.025	.739		−.060	.436		.020	.795		.037	.633
Physical activity (low)a	—	—	—	—	—	—	—	—	—	—	—	—
Sedentary (high)a		−.010	.897		.062	.424		−.031	.690		.175	.026

Note: Bold values are statistically significant.

^aDummy variable.

Table 3.

Multivariable Linear Regressions of Cognitive Function in People With Schizophrenia (Adjusting Physical Activity)

Variables	COG			GPT			RT-Reaction			RT-Motor
	R 2	Beta	P Value	R 2	Beta	P Value	R 2	Beta	P Value	R 2	Beta	P Value
	22.1			22.5			17.1			14.5
Age (≥40)a		−.192	.012		.140	.065		.057	.468		.104	.190
Sex (Male)a		.056	.551		−.036	.700		−.215	.027		−.165	.094
Education (low)a		−.030	.680		.076	.288		.045	.544		−.002	.984
Weight status (obese)a		.120	.146		−.013	.877		−.095	.266		−.032	.714
Smoke (yes)a		.023	.795		−.083	.345		.028	.756		.067	.469
Alcohol (Yes)a		−.098	.203		.153	.047		.040	.608		−.119	.138
MetS		−.008	.920		.114	.143		.134	.094		.001	.999
PANSS		−.175	.043		.164	.057		.336	<.001		.209	.021
Chlorpromazine equivalent doses		.148	.046		.153	.039		.003	.964		−.132	.089
Lorazepam equivalent doses		−.167	.028		−.010	.891		.083	.290		.083	.294
Time since illness onset (year)		.073	. 332		−.049	.512		−.066	.394		−.147	.063
Duration of hospitalization (month)		−.015	.840		−.074	.317		.014	.857		.033	.669
Physical activity (low)a		−.222	.008		.309	<.001		.136	.111		.084	.330
Sedentary (high)a		.025	.745		.014	.854		−.052	.505		.162	.042

Note: Bold values are statistically significant.

^aDummy variable.

Discussion

The current study is the largest to investigate objective sedentary behavior in people with schizophrenia and controls and has several novel findings. First, our data suggest that sedentary behavior is associated with worse motor reaction time, independent from multiple confounders. Second, our data show that among patients, those with higher levels of sedentary behavior appear to have worse speed processing and motor reaction times compared with those who are less sedentary. Third, lower levels of physical activity appear to be independently associated with poorer processing speed, attention, and dexterity. Fourth, our data are the first to suggest that sedentary behavior and physical inactivity (less physical activity) are independently associated with neurocognition. Finally, our data also confirm that people with schizophrenia engage in significantly more sedentary behavior (mean difference 4h and 7min per day) and less total physical activity than age and sex matched controls.

The potential mechanisms through which sedentary behavior may influence cognitive performance among people with schizophrenia are yet to be elucidated. One potential explanation is that the high levels of obesity and glucose/lipid metabolism alterations in people with schizophrenia,^45,46 which are known to influence cognition in the general population through atrophy in several brain regions and hippocampal dysfunction,⁴⁷ could play an integral role. Excess sedentary behavior in the general population is known to increase the risk of cardiometabolic disease⁴⁸ with around an extra hour a day being associated with a 22% and 39% increased risk of diabetes and MetS, respectively.⁴⁹ Some preliminary work has demonstrated that excess self-reported sedentary behavior is associated with elevated C reactive protein²⁰ and MetS²¹ in people with schizophrenia. There is also emerging evidence that obesity is associated with worse cognitive performance among people with schizophrenia.^50,51 Therefore, the hazards of obesity and metabolic alterations may be influenced by sedentary behavior, and extend beyond contributing to the excess mortality⁵² and also negatively influence cognition. An alternative explanation might also be that cognitive impairment may exacerbate negative symptoms (such as avolition) and reduce social functioning, both of which would increases the likelihood of sedentary behavior. Clearly, given the cross sectional nature of our work, more longitudinal research is required to disentangle these potential relationships and confirm the directionality to test these hypotheses.

The objective sedentary behavior data from the current study includes data on more people with schizophrenia than all of the previous 5 published studies pooled in a prior meta-analysis.²⁰ Our results found that that people with schizophrenia were sedentary for on average 581.1min per day, which is comparable with the recent meta-analysis in this patient group (599min per day).¹⁹ It is worth noting this amount of sedentary behavior is among the highest reported in any clinical population. The difference between patients with schizophrenia and controls in sedentary behavior was very high (over 4h) and is clearly concerning, not only for the potential relationship with poor cognition, but also due to the robust evidence in the general population that sedentary behavior is associated with cardiovascular disease and associated mortality.⁴⁸ Previously, a study found that people with schizophrenia that spent 1.5h of self-report sedentary behavior more than another group of patients, was associated with an increased level of C reactive protein (>5.0mg/μl). Our data also suggest that the difference in sedentary behavior between low and high sedentary patients (mean difference =196.4min per day) is associated with reduced cognitive function. An important area for future research is to explore with longitudinal/interventional data if reducing sedentary behavior in patients can improve cognition and indeed the inflammatory profile of people with schizophrenia.

Interestingly, we found that physical inactivity (lower levels of physical activity) is independently associated with worse processing speed and attention among our population with established psychosis. Although the relationship between physical activity and cognition is reasonably well established in the general population,⁵³ our data offer further evidence of the potential benefits of physical activity in people with schizophrenia. The relationship between low physical activity and these cognitive performance measures appears independent of sedentary behavior and other important confounders (eg, positive and negative symptoms, antipsychotic medication) and is the first data to demonstrate this independent relationship. There have been pressing calls for members of the multidisciplinary team to reduce sedentary behavior in people with psychosis due to the potential deleterious impact on cardiometabolic health⁵⁴ and an important future research question is whether replacing sedentary time with physical activity can improve health outcomes. There is an emerging evidence base from RCTs that aerobic exercise can improve cognition,⁵⁵ as to whether specifically targeting reductions in sedentary behavior can prevent decline in cognitive performance and/or improve cognition is an important area for future research. The potential mechanism by which physical activity could improve cognitive performance may be related to hippocampal neurogenesis, although the literature is equivocal among people with schizophrenia to date.⁵⁶ Regardless, given the established benefits of physical activity on various health dimensions among people with schizophrenia,⁵⁷ and relatively low levels of activity,⁵⁸ physical activity should be encouraged among this group.

Although our data are the largest study investigating objective sedentary behavior and include multiple novel results, some limitations exist. The primary limitation is that the data is cross sectional and therefore it is important that future longitudinal work disentangles the relationships we observed. Second, the study included people with established psychosis who were inpatients which limits the generalizability. Future research in the earlier stages of illness is required to see if similar relationships exist. In particular, interventions targeting reductions in sedentary behavior in those with first episode psychosis should consider if cognitive performance subsequently improves. There is some provisional evidence that physical exercise can improve cognition in people with first episode psychosis⁵⁹ and whether or not reducing sedentary behavior in this group can improve cognition is an important future research question. Third, the variance explored by our models was relatively modest, which does suggest that other factors not accounted for in our study may also account for the relationships we observed. Finally, we did not conduct any additional post hoc tests to adjust for multiple testing; therefore there is possibility of type I errors in the data. Clearly, future longitudinal research is required to confirm/refute our exploratory data.

In conclusion, the largest study using objective measures of sedentary behavior in schizophrenia to date, suggests that sedentary behavior may be related to reduced motor reaction time. Moreover, patients with higher levels of sedentary behavior appear to have worse cognitive performance than those engaging in less sedentary behavior. Higher levels of physical activity appear to be associated with better cognitive performance independent of sedentary behavior and multiple other factors. Clearly, future longitudinal data are required to confirm/refute our findings. Nonetheless, given the deleterious impact of sedentary behavior on multiple health outcomes, interventions specifically targeting reducing sedentary behavior are required among this highly sedentary population.

Funding

This work was in part supported by the National Taiwan University of Sport and the Taiwan Ministry of Science and Technology (104-2410-H-018-028). B.S. is supported by the National Institute for Health Research (NIHR) Collaboration for Leadership in Applied Health Research and Care South London at King’s College Hospital NHS Foundation Trust. The views expressed are those of the author[s] and not necessarily those of the NHS, the NIHR or the Department of Health.