Stroke maintenance exercise group: pilot study on dailyfunctioning in long-term stroke survivors

Sarah A. PattersonA,B, Benjamin M. Ross-EdwardsA and Hannah L. GillA

ACommunity Integrated Rehabilitation Service, Logan Central Community Health, PO Box 240,Logan Central, Qld 4114, Australia.

BCorresponding author. Email: sarah_patterson@health.qld.gov.au

Abstract. Typical models of stroke rehabilitation usually direct minimal resources for ongoing maintenancebeyond discharge. However, there is increasing recognition of the benefits of community-based rehabilitation tomaintain physical function and health in frail and disabled clients. A stroke maintenance exercise group wasestablished to provide long-term care for stroke survivors. A pilot studywas conducted to explore its effects comparedwith a traditional peer support group. Self-reported questionnaires, measuring daily task participation with the HomeFunctioningQuestionnaire andquality of lifewith theEQ-5D,wereutilised to compare twenty-twoclients in the strokemaintenance exercise group and twenty-one in the peer support group. The results indicated that both these groupsshowed a significant increase with daily task participation over a 3-month period. However, no improvement wasevident in either group on self-rated quality of life or health status, asmeasured by theEQ-5D.This pilot study suggeststhat both stroke maintenance exercise groups and peer support groups are effective with assisting long-term strokesurvivors to improve participation in everyday activities. More research is recommended to further explore thelong-term needs of this clinical group.

Additional keywords: cerebral vascular accident, outpatient care, rehabilitation.

BackgroundIn Australia, long-term stroke-related disabilities are anongoing concern for both stroke survivors and the health caresystem (Senes 2006). There are an estimated 346 000Australians recovering from stroke, with 146 400 of thesepeople continuing to live in the community with a stroke-related disability (Australian Institute of Health and Welfare2003). Approximately half of these clients require assistancewith daily living tasks (Senes 2006).

Furthermore, stroke clients decline over time, especially inrelation to reduced socialisation and activity levels (van dePort et al. 2006). Reduced fitness and mobility are also majorproblems following stroke (Ada et al. 2003; Green et al. 2004;Ouellette et al. 2004). In particular, deterioration of walkingability has been shown to result in decreased ability to engagein activities of daily living, reduced independence and greatersocial isolation (Eng et al. 2003).

The need for ongoing stroke support is often neglected intraditionalmodels of care that focus on the acute, inpatient andoutpatient rehabilitation phases with limited resourcesdirected to ongoing maintenance beyond discharge (Burton2000). It has been suggested that this lack of accessible andappropriate community-based programsmay contribute to the

decline in function for this clinical group (Legg andLanghorne 2004). Emerging consensus in the literatureindicates the effectiveness of organised community-basedrehabilitation services in maintaining functional gainsfollowing stroke (Werner and Kessler 1996; Ada et al. 2003;Eng et al. 2003; Chu et al. 2004; Green et al. 2004; Loganet al. 2004; Ouellette et al. 2004; Greenberg et al. 2006;Sanford et al. 2006). However, the recommended formatof these services is yet to be established (Hopman andVerner 2003).

The Australian National Stroke Guidelines (NationalStroke Foundation 2005) recommend that people living inthe community for 6 months post stroke should have accessto intervention to improve fitness and mobility. Serviceprovision within Australia is inconsistent with theserecommendations. Rehabilitation is typically provided indistinct time limited therapy blocks with little emphasis on thelong-term effects of the condition. Ongoing maintenancegroups for stroke clients are not usually considered within therehabilitation framework.

Conversely, peer stroke support services appear to be wellestablished within the community to provide ongoing supportfor long-term stroke survivors. According to the National

� La Trobe University 2010 10.1071/PY09055 1448-7527/10/010093

CSIRO PUBLISHING Practice & Innovation

www.publish.csiro.au/journals/py Australian Journal of Primary Health, 2010, 16, 93–97

Stroke Foundation (2006), there are many varied forms ofthese groups, although all involve conversational sharing ofstroke-related experiences. There is limited evidence toindicate improvements in quality of life or function associatedwith stroke peer support group attendance; however, manypeople report finding peer support groups helpful (NationalStroke Foundation 2006).

This ongoing need for long-term exercise stroke care wasobserved by allied health clinicians working within a strokecommunity rehabilitation service. To assist in fulfilling thisidentified service gap, the Queensland Health Metro SouthHealth Service District established a stroke maintenanceexercise group. This group was offered to all stroke clientsupon discharge from individual therapy.

A pilot study was undertaken to evaluate this strokemaintenance exercise group in comparison with a traditionalpeer support group already established in the community.Specifically, the primary aimwas to compare the effect of bothgroups on improvingdaily taskparticipation andquality of lifeoutcomes in long-term stroke survivors. It was hypothesisedthat participants attending the stroke maintenance exercisegroup would significantly improve with their quality of lifeand daily task participation in comparison with thoseattending the peer support program only.

MethodDesign

The design of the study was a non-randomised case-comparison study. Participants in the study were recruitedfrom a pre-existing, ongoing maintenance exercise group andpeer support group in the community. Study participants inboth groups self completed the outcome questionnaires asprovided by the research assistant for the pre and postassessments over a 3-month period during 2007–2008. Thestudy was approved by the Princess Alexandra HospitalHuman Research Ethics Committee in accordance with theNational Health and Medical Research Council Guidelines(Australia). Appropriate written consent for participationwithin the study was obtained for clients within each group.

Participants

Forty-three stroke survivors living within the communitywere recruited to the study, including 22 within the strokemaintenance exercise group and 21 from the peer supportgroup. Participants from both groups were recruitedsimultaneously over 2 weeks at the beginning of the 3-monthresearch period. Owing to the nature of attendance at bothgroups, there was varied attendance of participants duringeach of the recruitment weeks. The initial questionnaires werecompleted during this recruitment stage. Final questionnaireswere completed over another 2-week data collection periodafter 3 months. All participants had been involved with theirrespective groups before the commencement of the research,with all participants attending for a minimum of 3 monthsbefore the recruitment stage. The eligibility criteria for thestudywere community dwellerswith a confirmed diagnosis of

stroke. Clients were excluded from the study if they wereunable to answer the questionnaires owing to cognitive and/orlanguage difficulties related to their stroke or other co-morbidconditions. The maintenance exercise group comprised ofconsecutive admissions to a community-based health serviceexercise training and peer support class already in existence.Participants for the peer support groupwere self-selected fromclients attending a stroke support network group within anadjacent health service district.

Intervention

The stroke maintenance exercise group attended a weekly60-min exercise class conducted in a local community healthcentre rehabilitation gym. This program consisted of bothexercise training and peer support. The exercises includedbalance, gait re-education, strength, endurance, coordinationand fine motor tasks, and were based on functional exerciseslearned previously through traditional rehabilitation. The peersupport component included sharing accounts of personalexperience and adjustments to daily life. Themode of deliverywas active participation from clients with supervision by anoccupational therapist and a physiotherapist. This grouptypically averaged ~17 participants each week.

The peer support group comprised of a weekly60-min program conducted at a local community hall. Thisprogram consisted of peer support only. This peer supportincorporated personal achievement, planning leisureactivities, community-based participation, accounts ofpersonal experience and reports on adjustment to daily lifefollowing stroke. The mode of delivery was activeparticipation with facilitation from a key group member.This group typically averaged 22 group members each week.

Outcome measures

The Home Functioning Questionnaire (HFQ; Kneebone andHarrop 1996) was used to measure daily task participationas an activity limitation outcome measure. The HFQ is a12-item scale measuring levels of confidence to functionindependently at home in the areas of showering, dressing,bed transfer, mobility, toileting, opening doors andwindows, sandwich preparation, drinking, takingmedication,routine planning, contacting help and overall homefunctioning. Subscales were rated by the participant from 1to 4, with 1 = not confident at all, 2 = somewhat confident,3 =moderately confident and 4 = very confident.

The EQ-5D is a health-related quality of life measure,which comprises the following five dimensions: mobility,self-care, usual activities, pain/discomfort and anxiety/depression (The EuroQol Group 1990). Each dimension wasrated by the participant on three levels: no problems, someproblems and severe problems. The EQ-5D also includes avisual analogue scale (VAS), which records self-ratedperceived health status on a vertical scale from 0 (worstimaginable health state) to 100 (best imaginable health state).The VAS was utilised as a measure of health status.

94 Australian Journal of Primary Health S. A. Patterson et al.

Data analysis

A mixed-model design was utilised to examine both abetween-subjects variable and a within-subjects variable. Thebetween-subjects independent variable (IV) of ‘group’ hadtwo levels: stroke maintenance exercise group and peersupport group. The within-subjects (repeated-measures) IVof ‘time’ also had two levels: pre-assessment and post-assessment. The dependent variables (DVs) were scores on ameasure of daily task participation (HFQ), and a measure ofquality of life (EQ-5D), which included a perceived healthstatus. An alpha level (a) of 0.05 was used for all testing. Thepilot study aimed to recruit sufficient numbers to detect amedium effect size with power at 80%. Power calculationswerepost hoc (Faul et al. 2007), determining that the studyhadsufficient power to detect, at power of 80%, a small–mediumeffect size (F = 0.15).

A series of 2� 2 (group� time) mixed-model analysis ofvariance (ANOVA) were conducted. Before analysis, scoreson the HFQ and EQ-5D were examined for missing values,univariate outliers and whether the data distribution met theassumption of normality. There were no univariate outliers,and the assumption of normality was met.

ResultsThe demographics of the 43 long-term stroke survivors livingwithin the community recruited to the study are outlinedin Table 1. Participants to the study included 22 strokemaintenance exercise group clients (11 right-sided stroke, 10left, 1 missing) and 21 peer support group clients (11 right-sided stroke, 10 left). Twenty-five clients consented to takepart in the maintenance exercise group; however, three wereexcluded owing to the presence of co-morbid psychiatric orother neurological conditions. All twenty-one clientsconsented and participated in the study from the peer supportgroup. The two groups did not differ significantly with regardto age (t(35) = –1.109;P= 0.275) but, diddifferwith regards totime since stroke t(23.537) = –2.358; P= 0.027. However,time since strokewas shown to have no significant correlation

with pre-assessment scores on anyof themeasures.Also, therewere no differences in pre-assessment status based on groupmembership or stroke location.

There was a significant main effect for ‘time’ for dailytask participation as measured by Total Home FunctioningScores, indicating that clients from both groups improvedfrom pre- to post-assessment on this variable. However,there was no main effect for ‘group’ and no significant‘group� time’ interaction. In addition, there were nosignificant changes in either group for ‘time’, ‘group’ and‘time� group’ interactions within the other variables ofTotal Quality of Life Scores and Health Status Score. Theresults are summarised in Table 2.

DiscussionThis pilot study explored the effects of a newly establishedstroke maintenance exercise group and a peer support groupon daily task participation and quality of life. It washypothesised that the stroke maintenance exercise groupparticipants, receiving both exercise and peer support, wouldsignificantly improve with their quality of life and daily taskparticipation compared with those attending the peer supportprogram only.

The results, however, indicated a significant improvementin daily task participation for both groups over time. Thissuggests that both the stroke maintenance exercise group andstroke peer support group assist long-term stroke survivorswith improving their participation in everyday activities. Theresults indicate no improvement in the area of quality of life orhealth status in either group. The results also indicated nilsignificant differences between stroke maintenance exercisegroup and peer support group attendance.

The preliminary findings of this study support the need forappropriate community-based programs for long-term strokesurvivors. It can be concluded that such programs may assistin reducing the typical functional decline of stroke clientsover time as reported by van de Port et al. (2006). Theimprovements in daily task participation noted within bothgroups further strengthens the suggestion by Legg andLanghorne (2004) that the lack of appropriate services for thisclinical group contributes to the decline in function of strokesurvivors over time. It can be suggested that organisedcommunity-based rehabilitation services may be pivotal inmaintaining functional gains over time for communitydwelling stroke survivors.

Considering the similar benefits reportedwithin eachgrouptype, the stroke maintenance exercise group format may offer

Table 2. Effects of participation in a stroke maintenance exercise group and a peer support group for time, group and group� time interactions

Outcome Main effect for time Main effect for group Group� time interaction

Total home functioning F [1, 39] = 13.391, P= 0.001,partial h2= 0.256

F [1, 39] = 0.512, P= 0.479 F [1, 39] = 1.269, P= 0.267

Total quality of life F [1, 36] = 0.169, P= 0.683 F [1, 36] = 0.797, P= 0.378 F [1, 36] = 0.032, P= 0.859Health status score F [1, 33] = 0.553, P= 0.462 F [1, 33] = 0.003, P= 0.957 F [1, 33] = 0.126, P= 0.724

Table 1. Characteristics of 43 long-term stroke survivors who wererecruited to the study

Characteristic Exercise (n= 22) Peer support (n= 21)

Mean age (years) 62.5 (s.d. 7.8) 66.5 (s.d. 11.25)Sex (M : F) 13 : 9 12 : 9Location of stroke (R : L) 11 : 10 (1 missing) 11 : 10Mean time since stroke (years) 3.3 (s.d. 2.1) 5.3 (s.d. 3.8)

Daily functioning in long-term stroke survivors Australian Journal of Primary Health 95

an ‘alternative’ for clientswhodonotwish to attend traditionalpeer support groups. As the stroke maintenance exercisegroup offers both an exercise and peer support component, itmay appeal to wider range of stroke survivors. In addition,stroke maintenance exercise groups could assist clients withlanguage difficulties because traditional conversationalsupport groups may be inadequate for their needs. Althoughthere were no differences between left and right stroke clientsin this study, exercise-based activities may be of particularbenefit to this client group.

Further research is required to explore the needs of long-term stroke survivors and the community-based servicesrequired to best meet their needs. Owing to a non-randomisedcomparison of a small sample size and potential reporter bias,the true effect of the stroke maintenance exercise group mayhave not been captured in this study. It is also likely that otherexternal factors would have affected the participants in bothgroups, influencing the outcomes irrespective of theintervention provided. In addition, the study is also likely to belimitedbyapossible ceiling effect on theEQ-5D, confoundingthe results on this variable. Furthermore, utilising self reportmeasures may not capture objective measures of functionalperformance.

Future research is recommended to include the use of acontrol group to investigate whether peer support and/orstrokemaintenance exercise groups are effective in improvingdaily task participation compared with clients livingunsupported in the community. In addition, long-term follow-up of these participants is recommended to determinewhetherthe improvement in participation is sustained over time.Further investigation into the effectiveness of the strokemaintenance exercise group could also involve qualitativeresearch regarding client satisfaction and perceived groupbenefits. Exploring the cost effectiveness of these ongoingstroke servicesmay assist with strategic planning of long-termcare for this clinical group.

ConclusionThe need for ongoing support services for long-term strokesurvivors is becoming recognised. This pilot study into anewly established a stroke maintenance exercise groupprovides further suggestions that ongoing stroke services,either exercise-based or in a peer support format, appear toimprove daily task participation in stroke survivors. It is hopedthat continued research into the effectiveness of long-termstroke programs may assist in the development of evidence-based models of care for stroke rehabilitation services.Considering the reported functional decline for stroke clientsover time, it is anticipated that establishment of such servicescould ease the burden of this disease in the community.

Conflicts of interestThe researchers completing this study were employed byQueensland Health and responsible for the establishment ofthis stroke maintenance exercise group.

AcknowledgementsThe authors would like to thank Gail Hetherington, therapy assistant, for herassistance in data collation and theQueenslandHealth Community IntegratedRehabilitation Service at Logan Central Community Health Centre for theirfunding and support of this research.

ReferencesAda L, Dean CM, Hall JM, Bampton J, Crompton S (2003) A treadmill

and overground walking program improves walking in persons residingin the community after stroke: a placebo-controlled, randomized trial.Archives of Physical Medicine and Rehabilitation 84(10), 1486–1491.doi:10.1016/S0003-9993(03)00349-6

Australian Institute of Health and Welfare (2003) ‘Stroke in Australia.’Available at http://www.aihw.gov.au/cvd/stroke.cfm [Verified 24 July2009]

Burton CR (2000) Re-thinking stroke rehabilitation: the Corbin and Strausschronic illness trajectory framework. Journal of AdvancedNursing 32(3),595–602. doi:10.1046/j.1365-2648.2000.01517.x

Chu KS, Eng JJ, Dawson AS, Harris JE, Ozkaplan A, Gylfadottir S (2004)Water-based exercise for cardiovascular fitness in people with chronicstroke: a randomized controlled trial. Archives of Physical Medicine andRehabilitation 85(6), 870–874, 1039–1040. doi:10.1016/j.apmr.2003.11.001

Eng JJ, Chu KS, Kim CM, Dawson AS, Carswell A, Hepburn KE (2003)A community-based group exercise program for persons with chronicstroke. Medicine and Science in Sports and Exercise 35(8), 1271–1278.doi:10.1249/01.MSS.0000079079.58477.0B

Faul F, Erdfelder E, Lang AG, Buchner A (2007) G*Power 3: a flexiblestatistical power analysis program for the social, behavioural, andbiomedical sciences. Behavior Research Methods 39(2), 175–191.

Green J, Young J, Forster A, Collen F, Wade D (2004) Combined analysisof two randomized trials of community physiotherapy for patientsmore than one year post stroke. Clinical Rehabilitation 18(3), 249–252.doi:10.1191/0269215504cr747oa

Greenberg E, Treger I, Ring H (2006) Rehabilitation outcomes inpatients with brain tumours and acute stroke: comparative study ofinpatient rehabilitation. American Journal of Physical Medicine &Rehabilitation 85(7), 568–573. doi:10.1097/01.phm.0000223218.38152.53

Hopman WM, Verner J (2003) Quality of life during and after inpatientstroke rehabilitation. Stroke 34(3), 801–805. doi:10.1161/01.STR.0000057978.15397.6F

Kneebone I,HarropA (1996) Patient confidence in return home: developmentof a measurement instrument. Australian Occupational Therapy Journal43(1), 19–23.

Legg L, Langhorne P (2004) Rehabilitation therapy services for strokepatients living at home: systematic review of randomised trials. Lancet363(9406), 352–356. doi:10.1016/S0140-6736(04)15434-2

Logan PA, Dyas J, Gladman JRF (2004) Using an interview study oftransport use by people who have had a stroke to inform rehabilitation.Clinical Rehabilitation 18(6), 703–708. doi:10.1191/0269215504cr742oa

National Stroke Foundation (2005) ‘Clinical guidelines for strokerehabilitation and recovery.’ (National Stroke Foundation: Melbourne)

National Stroke Foundation (2006) ‘Survey of stroke support programs inAustralia and overseas.’ (National Stroke Foundation: Melbourne)

Ouellette MM, LeBrasseur NK, Bean JF, Phillips E, Stein J, Frontera WR,Fielding RA (2004) High-intensity resistance training improves musclestrength, self-reported function, and disability in long-term strokesurvivors. Stroke 35(6), 1404–1409. doi:10.1161/01.STR.0000127785.73065.34

96 Australian Journal of Primary Health S. A. Patterson et al.

Sanford JA,Griffiths PC,RichardsonP,HargravesK,ButterfieldT,HoenigH(2006) The effects of in-home rehabilitation on task self-efficacy inmobility-impaired adults: a randomized clinical trial. Journal of theAmerican Geriatrics Society 54(11), 1641–1648. doi:10.1111/j.1532-5415.2006.00913.x

Senes S (2006) ‘Howwemanage stroke inAustralia.’AIHWcat.No.CVD31.(Australian Institute of Health and Welfare: Canberra)

The EuroQol Group (1990) EuroQol – a new facility for the measurement ofhealth-related quality of life.Health Policy (Amsterdam) 16(3), 199–208.doi:10.1016/0168-8510(90)90421-9

van de Port IGL, Kwakkel G, Van Wijk I, Lindeman E (2006) Susceptibilityto deterioration of mobility long-term after stroke: a prospective cohortstudy. Stroke 37(1), 167–171. doi:10.1161/01.STR.0000195180.69904.f2

Werner RA, Kessler S (1996) Effectiveness of an intensive outpatientrehabilitation program for post-acute stroke patients. American Journalof Physical Medicine & Rehabilitation 75(2), 114–120. doi:10.1097/00002060-199603000-00006

Daily functioning in long-term stroke survivors Australian Journal of Primary Health 97

http://www.publish.csiro.au/journals/py