Psychiatry Research 229 (2015) 828–839

Contents lists available at ScienceDirect

Psychiatry Research

http://d0165-17

☆PROn Corr

29009 ME-m

journal homepage: www.elsevier.com/locate/psychres

A systematic review of the benefits of physical therapy within amultidisciplinary care approach for people with schizophrenia: Anupdate$

Elisa Vera-Garcia a,d, Fermín Mayoral-Cleries a, Davy Vancampfort b, Brendon Stubbs c,Antonio I. Cuesta-Vargas d,n

a Hospital Regional Universitario de Malaga, Departamento de Salud Mental, Malaga Spain. Instituto de Biomedicina de Málaga (IBIMA), Spainb University of Leuven, Department of Rehabilitation Sciences, Belgiumc University of Greenwich, School of Health and Social Care, UKd Departamento de Fisioterapia, Facultad de Ciencias de la Salud, Universidad de Málaga, Andalucia Tech, Cátedra de Fisioterapia y Discapacidad, Instituto deBiomedicina de Málaga (IBIMA), Grupo de Clinimetria (FE-14), Spain

a r t i c l e i n f o

Article history:Received 12 January 2015Received in revised form15 June 2015Accepted 30 July 2015Available online 31 July 2015

Keywords:Physical activityPhysical therapyExerciseMental healthYogaTai-chiProgressive muscle relaxation

x.doi.org/10.1016/j.psychres.2015.07.08381/& 2015 Elsevier Ireland Ltd. All rights rese

SPERO Registration number CRD42014014104espondence to: Av/ Arquitecto Peñalosa s/n (álaga, Spain.

ail address: acuesta@uma.es (A.I. Cuesta-Varg

a b s t r a c t

This systematic review summarizes the most recent evidence from randomized controlled trials (RCTs)considering the effectiveness of physical therapy interventions (aerobic exercises, strength exercises,relaxation training, basic body awareness exercises, or a combination of these) within the multi-disciplinary management of schizophrenia. Two authors searched PubMed, PsycINFO, EMBASE, Web ofScience, Physiotherapy Evidence Database (PEDro), and the Cochrane Library considering RCTs publishedfrom July 1, 2011–October 1, 2014. Thirteen RCTs representing 549 participants met the inclusion criteria.Overall, the results demonstrate that aerobic exercise significantly reduces psychiatric symptoms, po-tentially improves mental and physical quality of life and reduces metabolic risk and weight. Specifically,yoga reduces psychiatric symptoms, whilst Tai-chi and progressive muscle relaxation may also havebenefits to patients. Two RCTs reported on adverse events. No adverse event was observed supportingthe notion that physical therapy is safe in people with schizophrenia. There was considerable hetero-geneity in the design, implementation and outcomes in the included studies precluding a meaningfulmeta-analysis. In general, the quality of physical therapy RCTS is improving and current research de-monstrates that physical therapy approaches are valuable interventions and can help improve the psy-chiatric, physical and quality of life of people with schizophrenia.

& 2015 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

People with schizophrenia have a greatly increased number ofphysical comorbidities and experience a substantial prematuremortality gap compared to the general population, with recentestimates suggesting a deficit of 15–20 years (Reininghaus et al.,2014). A number of factors have been proposed to account for themortality gap, including high levels of smoking, alcohol use, asedentary lifestyle (Vancampfort et al., 2010), limited access tohealthcare and exercise facilities (Vancampfort et al., 2011b; DeHert et al., 2011) and in particular high levels of cardio-metabolicdiseases (Vancampfort et al., 2013d; Stubbs et al., 2015). In addi-tion, people with schizophrenia experience increased risk of

rved.

.Teatinos Campus Expansion),

as).

osteoporosis (Stubbs et al., 2014a) and chronic pain (Stubbs et al.,2014b). This increased burden of somatic co-morbidities also has adeleterious impact upon an individual’s mental health and qualityof life (Vancampfort et al., 2013c, 2011).

In non-mental health settings, physical therapy has a re-cognized role in addressing many of these comorbidities. Giventhis, previously Vancampfort et al. (2012) reported that physicaltherapy approaches might have an important role to play in themultidisciplinary treatment of people with schizophrenia. Theauthors concluded that, although still limited, there is evidencethat aerobic and strength exercises and yoga reduce psychiatricsymptoms, state anxiety, and psychological distress and improvehealth-related quality of life (Vancampfort et al., 2012). The au-thors furthermore stated that aerobic exercise improves short-term memory and that progressive muscle relaxation reduces stateanxiety and psychological distress (Vancampfort et al., 2012). Sincethe publication of this systematic review (Vancampfort et al.,2012), interest in managing these physical comorbidities and the

E. Vera-Garcia et al. / Psychiatry Research 229 (2015) 828–839 829

possible role of physical therapy approaches in the treatment ofpeople with schizophrenia increased (Stubbs et al., 2014c, 2014e;Soundy et al., 2014). For instance, a survey of members of the In-ternational Organization of Physical Therapists in Mental Health(IOPTMH, Stubbs et al., 2014d), a subgroup of the World Con-federation of Physical Therapy demonstrated that physical therapyshould have an integral part in the multidisciplinary treatment.Physical therapists should have a focused role on promoting themental and physical health needs of this underserved population.Although physical therapists have an integral and established rolein addressing many of the physical comorbidities seen in peoplewithout mental illness (e.g. cardiovascular disease, diabetes) con-sideration of the role of physical therapy in mental health settingsis still in its infancy. Given this, the IOPTMH (Vancampfort et al.2012, 2012b) stated that more rigorous evidence is needed toconsolidate the role of physical therapy approaches within themultidisciplinary treatment of schizophrenia. In recognition ofthis, we set out to provide an updated critique of physical therapyinterventions for researchers, physical therapists and decisionmakers. The purpose of this systematic review was to update thesystematic review of Vancampfort et al. (2012) and summarize themost recent evidence from randomized controlled trials (RCTs)examining the effectiveness of these physical therapy interven-tions in the multidisciplinary management of schizophrenia.

2. Method

This systematic review was undertaken in accordance withthe PRISMA guidelines (Moher et al., 2009) following a pre-determined, published protocol (PROSPERO registration number:CRD42014014104).

2.1. Eligibility criteria

We included RCTs that met the following criteria:(a) Participants. Included were people with schizophrenia

spectrum disorders according to DSM-IV (Diagnostic and Statis-tical Manual of Mental Disorders, Fourth edition, 1994) or ICD-10(International statistical classification of disease, injuries andcauses of death (10th reversion), 1992) criteria (excluding bipolardisorder or major depression with psychotic features).

(b) Interventions. We considered a physical therapy approachwhich could comprise physical exercise such as aerobic exercises,strength exercises, relaxation training, basic body awareness ex-ercises, or a combination of these in accordance with the WorldConfederation for Physical Therapy position statement (WorldConference for Physical Therapy, 2007). Specifically physical ex-ercise was defined as physical activity that is planned, structured,repetitive and purposive in the sense that improvement ormaintenance of physical performance or health is an objective(American College of Sport Medicine Position Stand, 1998). Thephysical therapy intervention could be used as single experimentalintervention or along with other interventions (i.e., cognitive re-mediation, occupational therapy) provided that physical therapywas the main intervention and if the specific effects of the physicaltherapy intervention could not be separated from other activecomponents in the intervention.

(c) Control interventions. The control interventions could con-sist of care as usual, a wait-list condition or an intervention inhealthy controls provided that both compared interventions hadsimilar duration. Standard care was defined as care that peoplewould have normally received had they not been included in theresearch trial. Such care would include hospitalization, communitypsychiatric nursing support, and outpatient care.

(d) Outcome measures. We considered any recognized outcome

measure considering participants mental and physical healthparameters using validated assessment tools.

We did not place any restrictions based upon age, sex or na-tionality of sample. If an overlap in data occurred in two studiesfrom the same research team, we used the largest data set. Weconsidered articles published in English and Spanish.

2.2. Data sources and searches

Two independent authors searched PubMed, PsycINFO, EM-BASE, Web of Science, Physiotherapy Evidence Database (PEDro),and the Cochrane Library. Databases were searched from July 1,2011 until October 1, 2014. Medical subject headings used included“schizophrenia” AND “physical therapy” OR “exercise” OR “re-laxation” in the title, abstract, or index term fields.

2.3. Study selection

Search results were screened by title and abstract by two in-dependent authors. In the event that the information in the title orabstract was insufficient, the researchers obtained the full-text.The next phase of the search strategy involved searching for un-published RCTs and for RCTs potentially overlooked or absent fromthe databases. This step involved manually searching the referencelists in all retrieved articles and the available systematic reviews.Furthermore, the authors searched websites housing details ofclinical trials, theses, or dissertations. Citation indexing was usedto track referencing of key authors in the field, and local expertswere contacted for further information.

2.4. Data extraction and quality assessment

In order to compare the quality of the included studies with theprevious systematic review of Vancampfort et al. (2012) and toassess for potential bias in the trials, each study was evaluatedwith the previously validated 5-point Jadad scale (Jadad et al.,1996; Clark et al., 1999; Moher et al., 1995). A score of 0 to 5 isassigned, with higher scores indicating higher quality in the con-duct or reporting of a trial. A trial scoring at least 3 of 5 is con-sidered to be of strong quality. A trial scoring below 3 is con-sidered to be methodologically weak. The Jadad scale focuses onlyon randomization, blinding, and withdrawals and dropouts toevaluate methodological quality of primary research. Only 2 ofthese 3 items are actually applicable to physical therapy becausethe nature of physical therapy interventions does not allow forblinding of the therapists, however blinding of assessors andparticipants is possible (Armijo et al., 2003). For this reason wedecided the use the reliable PEDro scale (Maher et al., 2003) nextto the Jadad scale as well. Each trial report is given a total PEDroscore ranging from 0 to 10. Studies scoring 9–10 were consideredmethodologically to be of “excellent” quality, studies ranging from6–8 were considered methodologically to be of “good” quality,studies scoring 4 or 5 were of “fair” quality while studies thatscore below 4 were felt to be of “poor” quality. After analysis of themethodological quality of the studies, we only included articleswhich score were 3 or high on the Jadad scale and 5 or higher onthe PEDro scale. For the data extraction, we used the data collec-tion form developed and used by the researchers in the previousreview (Vancampfort et al., 2012). Two independent authorscompleted the methodological quality appraisal and dataextraction.

2.5. Data synthesis

Due to the anticipated heterogeneity in the study design, in-terventions and outcome measures used a narrative synthesis to

E. Vera-Garcia et al. / Psychiatry Research 229 (2015) 828–839830

inform clinical practice was deemed the most appropriate mannerto report the data. We did consider a meta-analysis but only ifthere were three or more studies of sufficient homogeneity acrossthe outcomes measures to warrant pooling.

3. Results

3.1. Study selection

The search strategy resulted in 1153 potential articles beingidentified. After the removal of duplicates and screening of titles,abstracts, or full texts, 17 RCTs were considered potentially eligibleof which 13 met the inclusion criteria. Full details of the searchresults including reasons for exclusion are reported in Fig. 1.

3.2. Methodological quality

The Jadad and PEDro scores for the studies are shown in Ta-ble 1. Six publications (Ikai et al., 2013; Scheewe et al., 2013,2013b, Manjunath et al., 2013; Varambally et al., 2012; Ho et al.,2012) were considered to be of strong/good quality on both scales.Two studies (Oertel-Knöchel et al., 2014; Takahashi et al., 2012)were considered to be of weak/fair quality. The other articles(Scheewe et al., 2012; Georgiev et al., 2012; Battaglia et al., 2013;

Records identified: from July 1, 2011 to October 1, 2014(n=1,153)PubMed: 131PsycINFO: 630Web of Science: 367PEDro: 42

A

Records screened after remduplicates and articles that relevant

Full-text articles assessefor eligibility by

independent reviewers(n=17)

Studies included inqualitative synthesis

(n = 13)

Studies included ina meta-analysis

(n=0)

Fig. 1. Flow chart of systematic re

Visceglia et al., 2011 Visceglia et al., 2011; Heggelund et al., 2012)had a suitable quality score in one of two scales. Making an ana-lysis by items, major deficiencies were the lack of masking(especially in patients and therapists), none of the studies in thisreview realized this masking, and no intention-to-treat analysis(Scheewe et al., 2013b; Manjunath et al., 2013; Varambally et al.,2012; Georgiev et al., 2012). Full details are summarized in Table 1.

3.3. Participants

A total of 549 participants (age range¼22–64years) were in-cluded across the 13 RCTs. The majority of the included partici-pants were male (n¼319, 58.1%). Most of the included studiesdiagnosed schizophrenia according to the DSM criteria apart fromtwo studies which used the ICD-10 classification criteria (Ikai et al.,2013; Heggelund et al., 2012). Three studies included both in-patients and outpatients (Scheewe et al., 2013; Ho et al., 2012;Heggelund et al., 2012), two studies relied solely on outpatients(Ikai et al., 2013; Varambally et al., 2012), four studies includedonly inpatients (Manjunath et al., 2013; Georgiev et al., 2012;Oertel-Knöchel et al., 2014; Visceglia and Lewis, 2011), and fourstudies did not specify the setting (Scheewe et al., 2013b, 2012;Battaglia et al., 2013; Takahashi et al., 2012). More detailed char-acteristics of the participants are presented in Table 1.

dditional records identifiedthrough other sources

(n = 0)

oval of were no

Reasons for exclusion:physical therapy no main intervention (n=44)no RCT (n=11)RCTprotocol (n=4)Conference and poster abstract (n=3)Multiple component weight management

d

Full-text articles excluded(n = 4)

Sample no humansNo randomized trial

view inclusion and exclusion.

Table 1Details of included randomized controlled trials.

Study Participants Experimental vsControl interven-tion (No.participants)

Duration Frequency Intensity Relevant Outcomes (Ex-perimental vs Control)

RelevantInstruments

Complementarycointervention

JadadScore

PEDroScore

Ikai et al.(2013)

49 Outpatients (32 men) withschizophrenia or related psychoticdisorder (F20–F29) (ICD-10)

Yoga therapy group(25) vs regular day-care program (24)

8 weeks 60 min1 time/week

Yoga group, improvementon the total length of trunkmotion (F¼ 7789, p¼0.008)and the Romberg ratio(F¼7.849, p¼0.009) andanteflexion standing(F¼6.562, p¼0.014). Im-provement on PANSS nega-tive symptoms subscale(F¼6.566, p¼0.014) andFACT-Sz scale (F¼9.900,p¼0.003)

CSP Pharmacotherapy atstable doses for at least8 weeks and routinenon-structured clinicalmanagement

3 7Anteflexion instandingPANSS

Improvements noted atweek 8 in the yoga groupwere attenuated closer totheir baseline at week 16

DIEPSSFACT-SzEQ-5DMean age:

intervention54.879.0 ycontrol 51.5715.1 y

Scheeweet al.(2013)

63 patients (46 male) with Schi-zophrenia 45, schizoaffective 15,schizophreniform disorder 3.(DSM-IV)

Exercise therapy(29) vs occupationaltherapy (25)

6 months 1 h 2 times/week

Intensity was in-creased gradually(week 1–3: 45%; week4–12: 65%; week 13–26: 75% of heart ratereserve based onbaseline CPET)

Intention-to-treat analyses:a trend-level effect on de-pressive symptoms(p¼0.07) and a significanteffect on cardiovascular fit-ness (Wpeak) po0.01),compared with occupa-tional therapy

PANSS Pharcacotherapy takingthe same dosage for atleast 4 weeks prior toinclusion

3 8

Per protocol analyses: ex-ercise therapy reducedsymptoms of schizo-phrenia (p¼0.001), de-pression (p¼0.012), needof care (p¼0.050), andincreased cardiovascularfitness (po0.001) com-pared with occupationaltherapy. No effect formetabolic syndrome(MetS) factors was foundexcept a trend reductionin triglycerides (p¼0.08)

Mean age:intervention29.277.2 ycontrol33.177.7 y

CPET (Wpeak andVO2peak)Montgomery As-berg depressionrating scale

E.Vera-Garcia

etal./

PsychiatryResearch

229(2015)

828–839

831

Table 1 (continued )

Study Participants Experimental vsControl interven-tion (No.participants)

Duration Frequency Intensity Relevant Outcomes (Ex-perimental vs Control)

RelevantInstruments

Complementarycointervention

JadadScore

PEDroScore

CANBody massindex, body fatpercentage,

and MetS

Scheeweet al.(2013b)

32 patients (26 male) with schi-zophrenia spectrum disorder(DSM-IV) Mean age:

Patients: exercisetherapy (18) or oc-cupational therapy(14)

6 months 1 h 2 times/week

Intensity was in-creased gradually(week 1–3: 45%; week4–12: 65%; week 13–26: 75% of heart ratereserve based onbaseline CPET)

Exercise therapy did notincrease global brain vo-lume, hippocampal volume,or cortical thickness inschizophrenia. Cardio-re-spiratory fitness improve-ment was significantly re-lated to cerebral matter vo-lume increase (0.164 ml/W;p¼0.045), lateral ventricle(�0.018 ml/W; p¼0.035)and third ventricle volumedecrease (�0.0018 ml/W;p¼0.013) and at trend-levelsignificance related to in-crease in cerebral graymatter (0.159 ml/ W;p¼0.059)

MRI Patient were stable onantipsychotic medica-tion (using the samedosage for at least4 weeks prior toinclusion)

3 6

Healthy control:exercise therapy(25) or life as usual(27)

intervention 28.577.3 ycontrol31.178.0 yMinimal 50% compliance andbaseline and follow-upMRI PANSS

CPET

Scheeweet al.(2012)

63 patients (46 make) with schi-zophrenia (DSM-IV) Minimalcompliance demands of 50% of 52offered sessions and compiledwith the maximal exercise testingdemand (RERpeak Z1.1) (33)

Exercise therapy (17)vs occupationaltherapy (16). Controlgroup healthy (53):exercise therapy (26)vs life as usual (27)

6 months 1 hour2 times/week

Exercise intensity wasincreased stepwise(week 1–3, 45%; week4–12, 65%; week 13–26, 75% of HR reservebased on baselineCPET data)

In patients, exercise therapyincreased relative V ̇O2peakcompared with decreasedrelative V ̇O2peak after oc-cupational therapy. In con-trols, relative V ̇O2peak in-creased after exercise ther-apy and to a lesser extentafter life as usual (group, PG0.01; randomization, p¼0.03). Exercise therapy in-creased Wpeak in patientsand controls compared withdecreased Wpeak in nonexercising patients andcontrols (po0.001)

CPET (Wpeak andVO2peak)

The same dosage ofantipsychotic medica-tion for at least 4 weeksbefore inclusion

3 5

Manjunathet al.(2013)

60 in-patients with schizophreniadiagnostic (DSM-IV). Mean age:

Yoga (35) vs exercisetherapy (25)

2 weeks inhospital,4 weeks athome

1 h Not different on the clinicalsyndrome scores at the endof 2 weeks.

PANSSHDRS Patients received anti-psychotic medication attherapeutic doses withanti-parkinsoniandrugs as needed

3 6

At the end of 6 weeks,patients in the yoga grouphad lower mean scores on

30.978.0 (all assessmentscompleted)

E.Vera-Garcia

etal./

PsychiatryResearch

229(2015)

828–839

832

CGIS (po0.01), HDRS(po0.01), total PANSS(po0.05) and PANSSgeneral psychopathologysubscore(po0.05). RMANOVA de-tected an advantage foryoga over exercise in re-ducing the CGIS scoresand HDRS scores

CGIS5 days/week

Varamballyet al.(2012)

95 outpatients Yogasana (39), ex-ercise (22), andwaitlist (34)

1 monthtraining athospital,continue athome3 months

45 min Reductions in PANSS nega-tive symptoms scores be-tween baseline and4 month, the yoga group(33.3% of patients), exercise(9.1%), wait list (8.8%),(p¼0.006); in total PANSSyoga group (38.9% of pa-tients), exercise group(25%), wait list (8.8%),(p¼0.002); in SOFS scores:yoga group (15.4% of pa-tients), exercise group(4.5%), wait list (2.9%),(p¼0.05).Yogasana pro-duced a 5-fold increase inthe chance of obtaining im-provement in PANSS nega-tive scores as comparedwith either exercise (Oddsratio¼5.000; 95% CI¼1.01–24.74) or waitlist (Oddsratio¼5.167; 95% CI¼1.32–20.1). The likelihood of im-provement in yoga group interms of total PANSS scorewas about six times greaterthan in the waitlist group(Odds ratio¼6.576; 95%CI¼1.69–25.66). The inter-ventions had no effect ondrug-induced extra-pyramidal ratings (data notpresented)

PANSS 3 6with diagnostic of schizophrenia(DSM-IV)Mean age: yoga32.8 710 yexercise 30.677.3 ywaitlist33.679.5 y

SOFSExtra pyramidalsymptoms

25 ses-sions/1°month

Georgievet al.(2012)

59 patients (28 male) PMR (31) restingcontrol condition(30)

Singlesession

Effect sizes for PMR were�0.22 (95%CI¼�0.73 to0.19) state anxiety, �0.96(95%CI¼�0.41 to �1.43)psychological stress, andþ1.01 (95%CI¼0.45 to 1.54)subjective well-being

SAI 3 5Mean age: SEESintervention 43.0378.76 ycontrol41.0779.00 y

Ho et al.(2012)

30 patients (12 male) with schi-zophrenia diagnosis (DSM-IV)

Thai-Chi (15) vswaitlist (15)

6 weeks 1 h There is a significant differ-ence in how the Tai-chi

Qualitativefeedback

Standard residentialcare which includes a

3 6

E.Vera-Garcia

etal./

PsychiatryResearch

229(2015)

828–839

833

Table 1 (continued )

Study Participants Experimental vsControl interven-tion (No.participants)

Duration Frequency Intensity Relevant Outcomes (Ex-perimental vs Control)

RelevantInstruments

Complementarycointervention

JadadScore

PEDroScore

group and the waitlistgroup faired on the displa-cing test (Z¼�2.28;p¼0.023; r¼0.42) andmarginal significance theone-hand test (Z ¼�1.95;p¼0.065; r¼0.36)

30-minute daily morn-ing stretching routinefor all participants

CMDTMean age: intervention51.87710.85 y 2 times/

weekcontrol53.4778.63 y SANSs

Changes to negative symp-toms were not statisticallysignificant.

WHODAS-II

Fewer disruptions in lifeactivities functioning wasobserved for the Tai-chigroup at maintenance(T3) (Z¼�2.14; p¼0.03;r¼0.55). The Tai-chi par-ticipants also found fewerdifficulties with commu-nity participation at T2(Z¼�2.73; p¼0.01;r¼0.70)

Battagliaet al.(2013)

18 male patients Trainig group (TG)(10) vs control group(8)

12 weeks 100–120 min

Moderate and vigor-ous activities bymanaging patients’mean HRs between50%–85% of the esti-mated individualHRmax

TG showed a significantdecrease in BW (beforetraining [BT]: 77.44713.60versus after training [AT]:73.89712.51 kg, Δ¼�4.6%,Po0.001) and BMI (BT:28.5574.06 versus AT:27.2273.70, Δ¼�4.6%,po0.001).

SF-12 3 5Mean age 30-mST3675 y 2 times/

weekWith schizophrenia and/or schi-zoaffective disorder (DSM-IV)Mean age:intervention36.0075.00control35.0074.00 STB

Oertel-Knö-chel et al.(2014)

29 patients (12 males) with schi-zophrenia DSM-IV

Cognitive and ex-ercise training (8) vscognitive and re-laxation trainig (11)vs waitlist (10)

4 week 75 min3 times/week

Aerobic endurance le-vel of 60–70% of in-dividual maximumheart rate.using themaximum heart ratesof the ECG measuredby instructors every10 min during thesessions

Significant main effect ofexercise and relaxation insubscale Negative of thePANSS [F (26)¼8.34,p¼0.02]

MATRICS consensusbattery

Stable medication sinceone last month beforetesting and during theintervention period

2 7

Both intervention groupshad a decrease in stateanxiety (STAI) from pre-to post-testingSF-12: PSK showed a sig-nificant increase betweenfirst and second measure-ment time point [effect oftime: F(46)¼18.62,po0.001] in both the aero-bic exercise and the relaxa-tion group

STAISF-12PANSSRHSMean age: exercise

44.63713.78relaxation34.9179.33control38.3374.51

E.Vera-Garcia

etal./

PsychiatryResearch

229(2015)

828–839

834

Visceglia andLewis(2011)

18 inpatients (12 male) withschizophrenia

Yoga Therapy pro-gram (10) vs Waitlistgroup (8)

8 week 45 min The instructor mat-ched the energy level,attentional ability, andmood state of thegroup members onany given day

Mean decline in PANSS totalscores from pre- to post-assessment significantlygreater in the YT group(�25.20711.24, t¼�4.54,po0.00) versus the WLgroup (1.13712.98); in-cluding PANSS scores onpositive syndrome(t¼�2.64, p¼0.02), nega-tive syndrome (t¼�3.04,po0.01), general psycho-pathology (t¼�3.74,po0.00), activation(t¼�2.29, po0.04), para-noia (t¼�2.89, po0.01),and depression subscales(t¼�2.62, po0.02). Theonly subscale that did notsignificantly decrease forthe YT group was thoughtdisturbance.

PANSS 2 62 times/week

WHOQOL-BREFMean age:intervention 42713.5 y

QOL scores also improvedsignificantly in the YTgroup, specifically in thephysical health(11.30711.09, t¼2.38,po0.04) and psychologicdomains (22.50721.80,t¼2.88, po0.01)

contol48.13711.24 y

Takahashiet al.(2012)

23 outpatients (12 male) withschizophrenia

Exercise module (13)vs

3 months 30–60 min For each participant,the exercise intensitylevel was set at 11–13(fairly light–somewhathard) of the Borg scale(rating of perceivedexertion)

BMI and PANSS generalscale were significantly re-duced in the program group(po0.001, t¼3.84, df¼12;po0.001, t¼4.65, df¼12)but not in the control group(p¼0.68, t¼�0.46, df¼9;p¼0.17, t¼�1.50, df¼9).

PANSS A program primarilyaiming to manageweight gain in chronicpatients

1 5

Cotrols (10) BMI

Basketball participationincreased EBA (extra-striate body area) activa-tion during observation ofbasketball-related actions(peak: x¼42, y¼�74,z¼4, Z score¼4.12, clus-ter size: 43 voxels) com-pared with that before theprogram.

DSM-IV

Increase in EBA activationwas associated with im-provement in the generalpsychopathology scale ofPANSS (r¼�0.78,p¼0.002), in the pro-gramme group

Mean age: 43.5711.8 y (inter-vention group)

2 times/day

39.9713.6 y(control) 6 times/

weekfMRI

E.Vera-Garcia

etal./

PsychiatryResearch

229(2015)

828–839

835

Table 1 (continued )

Study Participants Experimental vsControl interven-tion (No.participants)

Duration Frequency Intensity Relevant Outcomes (Ex-perimental vs Control)

RelevantInstruments

Complementarycointervention

JadadScore

PEDroScore

Heggelundet al.(2012)

13 in- and outpatients (5 male)with schizophrenia, schizotypal ordelusional disorder (ICD-10)

Maximal strengthtraining (MST) (6) vsComputer gametraining (CG) (7)

8 weeks 3 times/week

The training load cor-responded to 85-90%1RM

The MST group improved1RM with 83 kg (p¼0.006)while the CG not changefrom pre-to post-interven-tion (p¼0.465). Enet im-proved by 3.4% more in theMST group than in the CG

Enet The patients’ traditionaltherapy was unchangedduring the interventionperiod

1 6

group (p¼0.046).Mean age: 37.579.6 y(intervention)38.9711.4 y (control) The VO2 cost of walking

at 60 W was also reducedfor the MST group afterintervention (p¼0.028)but not for the CG group(p¼0.0176).

Oxygen update(l min�1,ml kg�1 min�1)

No significant changeswere observed from pre-to post-intervention inthe two groups in totalPANSS (MST: p¼0.115,CG: p¼0.753), SF-36physical health score(MST: p¼0.173, CG:p¼0.116) or mentalhealth (MST: p¼0.463,CG: p¼0.345)

HRVE

RERPANSSSF-36

ICD-10¼ International Classification of Diseases, the 10th edition; CSP¼ Clinical stabilometric platform; PANSS¼Positive and Negative Syndrome Scale; DIEPSS¼Drug Induced Extrapyramidal Symptoms Scale; FACT-Sz¼FunctionalAssessment for Comprehensive Treatment of Schizophrenia; EQ-5D¼Euro Quality of Life questionnaire; CPET¼ cardiopulmonary exercise test; CAN¼ Camberwell Assessment of Need; DSM-IV¼ Manual of Mental Disorders,Fourth Edition; HDRS¼Hamilton Depression Rating Scale; CGIS¼Clinical Global Impression Severity; SOFS¼ Social and Occupational Functioning Scale; PMR¼progressive muscle relaxation; SAI¼State anxiety inventory ofSpielberger; SESS¼ Subjective exercise experiences scale; CMDT¼Minnesota Rate of Manipulation Test; SANSs¼ Scale for the Assessment of Negative Symptoms; WHODAS-II¼World Health Organization Disability AssessmentSchedule; HR¼heart rate; SF-12¼ Assessment of self-reported quality of life was evaluated by the Short Form, PSK (psychic scale); 30-mST¼ 30 m sprint run test; STB¼slalom test with ball; RHS¼ Revised Hallucination Scale;WHOQOL-BREF¼ World Health Organization Quality of Life BREF questionnaire; BMI¼Body mass index; fMRI¼Functional Magnetic Resonance; EBA¼extrastriate body area; Enet¼Net metabolic rate; VE¼Total pulmoraryventilation; RER¼Respiratory exchange ratio; 1RM¼one repetition máximum.

E.Vera-Garcia

etal./

PsychiatryResearch

229(2015)

828–839

836

E. Vera-Garcia et al. / Psychiatry Research 229 (2015) 828–839 837

3.4. Experimental and control interventions

In four studies (Ikai et al., 2013; Manjunath et al., 2013; Var-ambally et al., 2012; Visceglia and Lewis, 2011), yoga was the mainintervention, which was compared with a regular day program(Ikai et al., 2013), aerobic exercise (Manjunath et al., 2013; Var-ambally et al., 2012) and/or waitlist (Varambally et al., 2012; Vis-ceglia and Lewis, 2011).

Aerobic exercise was the main intervention in the other sixtrials (Scheewe et al., 2013, 2013b, 2012; Battaglia et al., 2013;Oertel-Knöchel et al., 2014; Takahashi et al., 2012). Comparisonwas made with occupational therapy (Scheewe et al., 2013, 2013b,2012) and/or exercising in healthy controls (Scheewe et al., 2013b,2012; Battaglia et al., 2013) and patient control group (Vancamp-fort et al., 2011b). In the study of Oertel-Knöchel et al. (2014) acombination of exercise and cognitive training was compared withrelaxation and cognitive training versus a control waitlist group.Heggelund et al. (2012) a maximal strength training was comparedwith a computer game training group. Georgiev et al. (2012)analysed the effectiveness of progressive muscle relaxation com-pared with a resting control intervention. Ho et al. (2012) eval-uated the effectiveness of Tai-Chi sessions compared with a waitlist.

3.5. Outcomes

3.5.1. Effectiveness of aerobic exercises, strength exercises, or both inthe multimodal care of people with schizophrenia

In three studies (Scheewe et al., 2013; Oertel-Knöchel et al.,2014; Takahashi et al., 2012) combined aerobic exercise andstrength training significantly reduced the total Positive and Ne-gative Syndrome Scale (PANSS) score. The study of Oertel-Knöchelet al. (2014) showed that both intervention groups (cognitive plusexercise and cognitive plus relaxation) reduce the PANSS negativescore (p¼0.02). Takahashi et al. showed that the PANSS generalscore significantly reduced in exercise group (po0.001) and not incontrol group (p¼0.17). Three trials (Battaglia et al., 2013; Oertel-Knöchel et al., 2014; Heggelund et al., 2012) showed improve-ments in physical and mental quality of life scores, two (Battagliaet al., 2013; Oertel-Knöchel et al., 2014). Oertel-Knöchel et al.(2014) found as well similar improvements in the physical domainof quality of life in the relaxation groups (po0.001). Heggelundet al. (2012), in contrast, did not find significant changes in qualityof life following high-intensity training. Only one study (Scheeweet al., 2013) assessed effects of cardiovascular exercise on meta-bolic syndrome parameters and did find only a trend reduction intriglycerides (p¼0.08). The latter study however also showed thatindividuals with schizophrenia following aerobic exercise did needless care (p¼0.05). Reductions in body mass index (BMI) wereshown in the studies of Battaglia et al. (2013) and Takahashi et al.(2012). Finally, Scheewe et al. (2013) found that exercise therapyincreased relative cardiorespiratory fitness (VO2peak and peak,po0.001) in exercising patients and controls, and had other po-sitive neurobiological effects. Details are shown in Table 1.

3.5.2. Effectiveness of yoga in the multimodal care of people withschizophrenia

Three studies reported yoga as the primary intervention(Manjunath et al., 2013; Varambally et al., 2012; Visceglia andLewis, 2011) and found significant reductions in the total PANSSscore. Another study (Ikai et al., 2013) found that yoga only re-duced negative symptoms (p¼0.014). Manjunath et al. (2013)compared yoga with an exercise intervention (stretching andaerobic exercise), and showed that yoga was more efficacious thanexercise in decreasing Clinical Global Impression Severity scale(CGIS) and Hamilton Depression Rating Scale (HDRS) scores.

Varambally et al. (2012) found that yoga produced a five-fold in-crease in the chance of obtaining reduction in the PANSS negativesubscale compared with exercise group. Yoga also improves phy-sical and psychological quality of life, Visceglia and Lewis (2011)showed in their study that quality of life improves significantly inthe yoga group (in both physical health (po0.04) and psycholo-gical domain (po0.01). Finally, Ikai et al. (2013) found improve-ments in stability (p¼0.008 for total length of trunk motion andp¼0.009 for the Romberg ratio) and trunk flexibility (p¼0.014).

3.5.3. Effectiveness of progressive muscle relaxation (PMR) in themultimodal care of people with schizophrenia

One study (Georgiev et al., 2012) showed that PMR significantlydecreased state anxiety, psychological stress, and increased sub-jective well-being following PMR with clinical relevance (effectsize¼�0.22 to �0.96).

3.5.4. Effectiveness of Tai-chi in the multimodal care of people withschizophrenia

One study showed that tai-chi buffered deteriorations inmovement coordination but effects were not sustained 6-weekspost-intervention (Ho et al., 2012). However, no changes in ne-gative symptoms were found. The authors also found that tai-chiresulted in fewer disruptions in daily life activities (p¼0.03) andfewer difficulties with community participation (p¼0.01).

3.6. Adverse effects

Only two studies (Manjunath et al., 2013; Visceglia and Lewis,2011) reported whether adverse effects were present or not, yetboth reported no incidents during the trials.

3.7. Motivational strategies used

Only two studies (Scheewe et al., 2013; Manjunath et al., 2013)described motivation interventions to avoid drop-outs and in-crease adherence. Scheewe et al. (2013) used telephone remindersin order to minimize the attrition rate, while in the study ofManjunath et al. (2013) social support (throughout encourage-ments) was provided by relatives. Participants also noted theiryoga-practices at home in a logbook which were discussed duringmonthly follow-up visits.

4. Discussion

4.1. General findings

This systematic review provides a detailed exploration of themost recent evidence concerning physical therapy as an adjuncttreatment in the multidisciplinary care of people with schizo-phrenia. The current evidence builds upon the body of evidence upto 2011 summarized by Vancampfort et al. (2012, 2012b, 2013) andconfirms that aerobic, strength exercises and yoga reduce psy-chiatric symptoms while improving health-related quality of life.Also there is preliminary evidence that progressive muscle re-laxation reduces state anxiety and psychological distress is con-firmed in more recent rigorous studies. The current review clearlyshows that aerobic exercise improves cardio-metabolic fitness,which has been demonstrated in a recent meta-analysis in peoplewith severe mental illnesses (Rosenbaum et al., 2014) and adds tothe current knowledge that tai-chi shows promising as an evi-dence-based intervention.

In a previous systematic review (Vancampfort et al., 2012) a callwas made for rigorous RCTs to be developed. Our review showsthat this call has been met in recent years and that the quality of

E. Vera-Garcia et al. / Psychiatry Research 229 (2015) 828–839838

physical therapy RCTs is improving. Next to this our systematicreview confirms that physical therapy is safe and well-tolerated inpeople with schizophrenia. Nevertheless, adherence and drop-outrates are still not routinely reported in the literature. Engagementin physical therapy for people with schizophrenia does howeverrequire careful consideration. Previous qualitative research ex-ploring the psychosocial experiences during physical activity andexercise in people with schizophrenia provided a useful under-standing of the processes involved in initiating and maintainingphysical activity, as well as establishing barriers and facilitators(Soundy et al., 2014b, 2012).

Given the calls made to utilize physical activity interventions toreduce the burden of cardiovascular disease and premature mor-tality in this group (McNamee et al., 2013, Stubbs et al. 2014a), ourreview demonstrates that physical therapy should form a centralrole in daily clinical practice for people with schizophrenia.

4.2. Limitations

A number of important limitations must be considered withthese new results. First, the outcome measures and interventioncharacteristics (frequency, intensity, time) of the different physicaltherapy modalities used are heterogeneous. This diversity pre-cludes a meta-analysis. A consensus is urgently needed on whichvalid outcome measures and intervention characteristics are use-ful in future physical therapy research. Second, as with the pre-vious systematic review (Vancampfort et al., 2012), there is still apotential for selection bias; however, we used a comprehensivesearch strategy. In addition, two independent reviewers analysedthe research data, and reasons for study exclusions were docu-mented in the flow diagram. Third, performance bias may stilllimit the current findings. None of the new RCTs were double-blind, although this is hard to achieve, in particular providing aconventional placebo or control group that receives no active in-tervention in the control group is not practical and unethical.Many of the ‘treatment as usual groups’ still receive an active in-tervention, so it may be that physical therapy treatments have tosignificantly out perform the control groups to demonstrate asuperior effect. As indicated previously (Vancampfort et al., 2012),although researchers may not always be able to mask participantsto physical therapy interventions to remove the chance of per-formance bias, every attempt should be made to collect researchdata in a masked manner. Finally, there are still no RCTs availableinvestigating the role of basic body awareness exercises. As pre-viously reported (Vancampfort et al., 2012), the use of basic bodyawareness exercises as an adjunct treatment to improve bodybalance and self-esteem may be relevant for people with schizo-phrenia, but there is currently no RCTs to demonstrates its efficacy.

4.3. Future research

Since adherence and drop-out rates are still not well reportedin the literature and that people with schizophrenia do have im-portant motivational deficits (Vancampfort et al., 2015), futureresearch should focus on understanding the motivational dy-namics of physical therapy initiation and maintenance. Recently itwas demonstrated that the self-determination theory (SDT) (Deciand Ryan, 1985, 2000; Deci et al., 1994) might assist physicaltherapists in motivating people with schizophrenia to adopt andmaintain lifestyle changes (Vancampfort et al., 2013b, 2015b).Within SDT, it is suggested that all individuals have three keypsychological needs that should be satisfied to lead to relativelyautonomous types of physical activity motivation (Deci and Ryan,1985, 2000; Deci et al., 1994), i.e; need for autonomy, competenceand social relatedness. Future physical therapy research shouldinvestigate whether physical therapists can support participants’

autonomy by offering clear choices, supporting initiatives, avoid-ing the use of external rewards, offering relevant information forcomplying with the physical therapy exercises and using autono-my supportive language (e.g. “could” and “choose” rather than“should” and “have to”). This research should also explore whetherfeelings of competence are attained when people with schizo-phrenia experience benefits while following physical therapy.Additionally, incorporating mediator analyses into future physicaltherapy research will help confirm if any action theory links orconceptual theory links exist between theoretical frameworks anddesired physical therapy related goals. Qualitative research willalso help identify the barriers and facilitators for the successfulimplementation of physical therapy approaches in people withschizophrenia (Soundy et al., 2014).

4.4. Conclusion

The current scientific evidence highlights that physical therapyapproaches can improve the mental and physical health of peoplewith schizophrenia. Moreover, the quality of physical therapybased research is improving and is safe for this population andphysical therapists should be considered an important resource toimprove the health and wellbeing of people with schizophrenia. Inparticular, and the more recent higher quality RCTs demonstratethat physical therapy might have important cardiorespiratory,neurobiological and psychiatric benefits. More theoretically-drivenresearch is however required to examine how to reliably assistpeople with schizophrenia to adhere to the physical therapy pro-tocol and maintain benefits obtained following the intervention,and this in the face of significant motivational deficits that areinherent to schizophrenia. There is cause for cautious optimismthat if this kind of research can confirm the current evidence inlarge sample sizes and if benefits are maintained on the long-termthat physical therapy will be considered in due course as a cor-nerstone of the multidisciplinary treatment of people with schi-zophrenia. More physical therapy research therefore is needed ifthe potential mental and physical health benefits of physicaltherapy are to be maximized for this underserved population.

Conflict of interest

The authors declare they have no conflict of interest.

Appendix A. Supplementary material

Supplementary data associated with this article can be found inthe online version at http://dx.doi.org/10.1016/j.psychres.2015.07.083.

References

American College of Sport Medicine Position Stand, 1998. The recommendedquantity and quality of exercise for developing and maintaining cardior-espiratory and muscular fitness, and flexibility in healthy adults. Med. Sci.Sports Exerc. 30 (6), 975–991.

Battaglia, G., Alesi, M., Inguglia, M., Rocaccella, M., Caramazza, G., Bellafiore, M.,Palma, A., 2013. Soccer practice as an add-on treatment in the management ofindividuals with a diagnosis of schizophrenia. Neuropsychiatr. Dis. Treat. 9,595–603.

Clark, H.D., Wells, G.A., Huët, C., McAlister, F.A., Salmi, L.R., Fergusson, D., Laupacis,A., 1999. Assessing the quality of randomized trials: reliability of the Jadadscale. Control Clin. Trials 20 (5), 448–452.

Deci, E.L., Eghrari, H., Patrick, B.C., Leone, D.R., 1994. Facilitating internalization: theself-determination theory perspective. J. Personal. 62, 119–142.

Deci, E.L., Ryan, R.M., 1985. Intrinsic Motivation and Self-determination in HumanBehaviour. Plenum Press, New York.

E. Vera-Garcia et al. / Psychiatry Research 229 (2015) 828–839 839

Deci, E.L., Ryan, R.M., 2000. The “what” and “why” of goal pursuits: human needsand the self-determination of behavior. Psychol. Inquiry 11, 227–268.

De Hert, M., Cohen, D., Bobes, J., Cetkovich-Bakmas, M., Leucht, S., Ndetei, D.M.,Newcomer, J.W., Uwakwe, R., Asai, I., Möller, H.J., Gautam, S., Detraux, J., Correll,C.U., 2011. Physical illness in patients with severe mental disorders. II. Barriersto care, monitoring and treatment guidelines, plus recommendations at thesystem and individual level. World Psychiatry 10 (2), 138–151.

Georgiev, A., Probst, M., De Hert, M., Genova, V., Tonkova, A., Vancampfort, A., 2012.Acute effects of progressive muscle relaxation on state anxiety and subjectivewell-being in chronic bulgarian patients with schizophrenia. Psychiatr. Danub.24 (4), 367–372.

Heggelund, J., Morken, G., Helgerud, J., Nilsberg, G.E., Hoff, J., 2012. Therapeuticeffect of maximal strength training on walking efficiency in patients withschizophrenia- a pilot study. BioMed Central Res. Notes 5, 344.

Ho, R.T., Au Yeung, F.S., Lo, P.H., Law, K.Y., Wong, K.O., Cheung, I.K., Ng, S.M., 2012.Tai-Chi for residential patients with schizophrenia on movement coordination,negative symptoms, and functioning: a pilot randomized controlled trial. Evid.-Based Complement. Altern. Med. 2012 1-10 pp..

Ikai, S., Uchida, H., Suzuki, T., Tsunoda, K., Mimura, M., Fujii, Y., 2013. Effects of yogatherapy on postural stability in patients with schizophrenia-spectrum dis-orders: a single-blind randomized controlled trial. J. Psychiatr. Res. 47,1744–1750.

International statistical classification of disease, injuries and causes of death (10threversion), World Health Organization, Geneva, 1992.

Jadad, A.R., Moore, R.A., Carroll, D., 1996. Assessing the quality of reports of ran-domized clinical trials: is blinding necessary? Control Clin. Trials 17, 1–12.

Maher, C.G., Sherrington, C., Herbert, R.D., Moseley, A.M., Elkins, M., 2003. Relia-bility of the PEDro scale for rating quality of randomized controlled trials. Phys.Therapy 83, 713–721.

Manjunath, R.B., Varambally, S., Thirthalli, J., Basavaraddi, I.V., Gangadha, B.N., 2013.Efficacy of yoga as an add-on treatment for in-patients with functional psy-chotic disorder. Indian J. Psychiatry 55 (3), 374–378.

McNamee, L., Mead, G., MacGilivray, S., Laurie, S.M., 2013. Schizophrenia, poorphysical health and physical activity: evidence-based interventions are re-quired to reduce major health inequalities. Br. J. Psychiatry 203 (3), 239–241.

Moher, D., Liberati, A., Altman, D.G., 2009. Preferred reporting items for systematicreviews and meta-analyses: the PRISMA statement. Ann. Intern. Med. 151 (4),264–269.

Moher, D., Jadad, A.R., Nichol, G., 1995. Assessing the quality of randomized con-trolled trials: an annotated bibliography of scales and checklists. Control Clin.Trials 16, 62–73.

Oertel-Knöchel, V., Mehler, P., Thiel, C., Steinbrecher, K., Malchow, B., Tesky, V.,2014. Effects of aerobic exercise on cognitive performance and individualpsychopathology in depressive and schizophrenia patients. Eur. Arch. Psy-chiatry Clin. Neurosci. 264, 589–604.

Reininghaus, U., Dutta, R., Dazzan, P., Doody, G.A., Fearen, P., Lappin, J., 2014.Mortality in schizophrenia and other psychoses: a 10-year follow-up of theESOP First-Episode Cohort. Schizophr. Bull. 41 (3), 664–673.

Rosenbaum, S., Tiedemann, A., Sherrington, C., Curtis, J., Ward, P.B., 2014. Physicalactivity interventions for people with mental illness: a systematic review andmeta-analysis. J. Clin. Psychiatry 75 (9), 964–974.

Scheewe, T.W., Takken, T., Kahn, R.S., Cahn, W., Backx, F.J.G., 2012. Effects of exercisetherapy on cardiorespiratory fitness in patients with schizophrenia. Med. Sci.Sports Exerc. 44 (10), 1834–1842.

Scheewe, T.W., Backx, F.J.G., Takken, T., Jörg, F., van Strater, A.C.P., Kroes, A.G., Kahn,R.S., Cahn, W., 2013. Exercise therapy improves mental and physical health inschizophrenia: a randomised controlled trial. Acta Psychiatr. Scand. 127,464–473.

Scheewe, T.W., van Haren, N.E.M., Sarkisyan, G., Schnack, H.G., Bruower, R.M., Glint,M., Hulshoff Pol, H.E., 2013b. Exercise therapy, cardiorespiratory fitness andtheir effect on brain volumes: a randomised controlled trial in patients withschizophrenia and healthy control. Eur. Neuropsychopharmacol. 23 (7),675–685.

Soundy, A., Kingstone, T., Coffee, P., 2012. Understanding the psychosocial processesof physical activity for individuals with severe mental illness: a meta-etno-graphy. In: Abate, L. (Ed.), Mental Illnesses- Evaluation, treatments and im-plications. InTech, Croatia.

Soundy, A., Freeeman, P., Stubbs, B., Probst, M., Cofee, P., Vancampfort, D., 2014. Thetranscending benefits of physical activity for individuals with schizophrenia: asystematic review and meta-ethnography. Psychiatry Res. 15–200 (1–2), 11–19.

Soundy, A., Stubbs, B., Probst, M., Hemmings, L., Vancampfort, D., 2014b. Barriers toand facilitators of physical activity among persons with schizophrenia: a surveyof physical therapist. Psychiatr. Serv. 65 (5), 693–696.

Stubbs, B., De Hert, M., Sepehry, A.A., Correll, U., Mitchel, A.J., Soundy, A., 2014a. Ameta-analysis of prevalence estimates and moderators of low bone mass in

people with schizophrenia. Acta Psychiatr. Scand. 130 (6), 470–486.Stubbs, B., Mitchell, A.J., De Hert, M., Correll, C.U., Soundy, A., Stroobants, M., 2015.

The prevalence and moderators of clinical pain in people with schizophrenia: asystematic review and large scale meta-analysis. Schizophr. Res. 160 (1-3), 1–8.

Stubbs, B., Probst, M., Soundy, A., Parker, A., De Herdt, A., De Hert, M., 2014c.Physiotherapists can help implement physical activity programmes in clinicalpractice. Br. J. Psychiatry 204 (2), 164.

Stubbs, B., Soundy, A., Probst, M., De Hert, M., De Herdt, A., Vancampfort, D., 2014d.Understanding the role of physiotherapists in schizophrenia: an internationalperspective from members of the International Organisation of PhysicalTherapists in Mental Health (IOPTMH). J. Ment. Health 23 (3), 125–129.

Stubbs, B., Soundy, A., Probst, M., Parker, A., Skjaerven, L.H., Lundvik Gyllensten, A.,2014e. Addressing the disparity in physical health provision for people withschizophrenia: an important role for physiotherapists. Physiotherapy 100 (3),185–186.

Stubbs, B., Vancampfort, D., De Hert, M., Mitchell, A.J., 2015. The prevalence andpredictors of type two diabetes mellitus in people with schizophrenia: a sys-tematic review and conmparative meta-analysis. Acta Psychiatr. Scand. 132,144–157.

Takahashi, H., Sassa, T., Shibuya, T., Kato, M., Koeda, M., Murai1, 2012. Effects ofsports participation on psychiatric symptoms and brain activations duringsports observation in schizophrenia. Transl. Psychiatry 20 (2).

Vancampfort, D., Knapen, J., Probst, M., van Winkel, R., Deckx, S., Maurissen, K.,2010. Considering a frame of reference for physical activity research related tothe cardiometabolic risk profile in schizophrenia. Psychiatry Res. 177 (3),271–279.

Vancampfort, D., Probst, M., Scheewe, T., Maurissen, K., Sweers, K., Knapen, J., DeHert, M., 2011. Lack of physical activity during leisure time contributes to animpaired health related quality of life in patients with schizophrenia. Schi-zophr. Res. 129 (2–3), 122–127.

Vancampfort, D., Sweers, K., Probst, M., Mitchell, A.J., Knapen, J., De Hert, M., 2011b.Quality assessment of physical activity recommendations within clinical prac-tice guidelines for the prevention and treatment of cardio-metabolic risk fac-tors in people with schizophrenia. Community Ment. Health J. 47 (6), 703–710.

Vancampfort, D., Probst, M., Helvik Skjaerven, L., Catalán-Matamoros, D., Lundvik-Gyllensten, A., Gómez-Conesa, A., Ijntema, R., De Hert, M., 2012. Systematicreview of the benefits of physical therapy within a multidisciplinary care ap-proach for people with schizophrenia. Phys. Therapy 92 (1), 11–23.

Vancampfort, D., Vansteelandt, K., Scheewe, T., Probst, M., Knapen, J., De Hert, A., DeHerdt, M., 2012b. Yoga in schizophrenia: a systematic review of randomisedcontrolled trials. Psychiatr. Scand. 126 (1), 12–20.

Vancampfort, D., Correll, C.U., Scheewe, T., Probst, M., De Herdt, A., Knapen, J., DeHert, M., 2013. Pregressive muscle relaxation in persons with schizophrenia: asystematic review of randomized controlled trials. Clin. Rehabil. 27 (4),291–298.

Vancampfort, D., De Hert, M., Vansteenkiste, M., De Herdt, A., Scheewe, T.W.,Soundy, A., 2013b. The importance of self-determined motivation towardsphysical activity in schizophrenia: a multicentre study. Psychiaty Res. 210 (3),812–818.

Vancampfort, D., Probst, M., Scheewe, T., De Herdt, A., Sweers, K., Knapen, J., 2013c.Relationships between physical fitness, physical activity, smoking and meta-bolic and mental health parameters in people with schizophrenia. PsychiatryRes. 207 (1–2), 25–32.

Vancampfort, D., Wampers, M., Mitchell, A.J., Corell, C.U., De Herdt, A., Probst, M.,2013d. A meta-analysis of cardio-metabolic abnormalities in drug naïve, first-episode and multi-episode patients with schizophrenia versus general popu-lation controls. World Psychiatry 12 (3), 240–250.

Vancampfort, D., De Hert, M., Stubbs, B., Ward, P.B., Rosenbaum, S., Soundy, A.,Porbst, M., 2015. Negative symptoms are associated with lower autonomousmotivation towards physical activity in people with schizophrenia. Compr.Psychiatry (in press).

Vancampfort, D., Vansteenkiste, M., De Hert, M., De Herdt, A., Soundy, A., Stubbs, B.,Buys, R., Probst, M., 2015b. Self-determination and stage of readiness to changephysical activity behaviour in schizophrenia. Ment. Health Phys. Act. 7 (3) (inpress).

Varambally, S., Gangadhar, B.N., Thirthalli, J., Jagannathan, A., Kumar, S., Venkata-subramanian, G., 2012. Therapeutic efficacy of add-on yogasana intervention instabilized outpatient schizophrenia: randomized controlled comparison withexercise and waitlist. Indian J. Psychiatry 54 (3), 227–232.

Visceglia, E., Lewis, S., 2011. Yoga therapy as an adjunctive treatment for schizo-phrenia: a randomized, controlled pilot study. J. Altern. Complement. Med. 17(7), 601–607.

World Confederation for Physical Therapy, 2007. WCPT Description of PhysicalTherapy. WCPT, London, United Kingdom.