R E S E A R C H A R T I C L E C o n s u l t in g t h e S o u t h A f r i c a n E x p e r t s in P h y s i o t h e r a p e u t i c S t r o k e R e h a b il it a t io n A B S T R A C T : T h is p a p e r p r e s e n ts th e o p in io n s o f e x p e r t p h y s io th e r a p is ts o n w h a t c o n s titu te s o p tim a l s tr o k e r e h a b ilita tio n in S o u th A fr ic a . D a ta w e re c o lle c te d b y th e u s e o f th e D e lp h i te c h n iq u e . C o n s e n s u s w a s r e a c h e d a fte r tw o ro u n d s, a n d th e r e s p o n d e n ts ’ v ie w s a re s u m m a r is e d a n d d is c u s s e d w ith in th e f r a m e w o r k o f S o u th A fr ic a n h e a lth care. R e s u lts s h o w e d th a t p h y s io th e r a p y w a s f e l t to b e v e r y im p o r ta n t a fte r stro ke, a n d th e s u r v e y c r e a te d a p r o file o f th e s k ills th a t p h y s io th e r a p is ts m a y re q u ire in o r d e r to w o r k in th is fie ld . H o w ever, n o n e w o r in n o v a tiv e m e th o d s b y w h ic h a p p r o p r ia te r e h a b ilita tio n A fr ic a w e re g e n e r a te d b y th e su rvey. T h e D e lp h i te c h n iq u e is d e s c r ib e d a n d its u s e in th is s u r v e y c o n sid e re d . K E Y W O R D S : S T R O K E , R E H A B IL IT A T IO N , D E L P H I T E C H N IQ U E This study was part of a dissertation submitted by Mrs. L.A. Hale to the University of the Witwatersrand for a Doctoral degree in Physiotherapy, and was granted ethical clearance from the Committee for Research on Human Subjects, University of the Witwatersrand: No. 950 111. The study was funded by the South African Medical Research Council and the University of the Witwatersrand. HALE LA, MSc’; EALES CJ, Ph.D2 1 lecturer, Department of Physiotherapy, University of the W itwatersrand. Professor o f Physiotherapy, Department of Physiotherapy, University of the W itwatersrand and the Johannesburg Hospital. s e r v ic e s c o u ld b e d e liv e r e d in S o u th Stroke is the second highest cause o f death due to chronic diseases o f lifestyle in South A frica (Fritz, 1997) resulting in between 8 - 10% o f all reported deaths, and 7.5% o f deaths in the country’s w orkforce (25 - 64 years o f age). The age-standardised mortality rate is 125 - 175/100,000 (Neurological A sso ciatio n o f South A frica Stroke W orking Group, 2000). An increase in the incidence and mortality following cerebrovascular accident is predicted as life style risk factors, such as high cholesterol diets and sm oking are added to the high prevalence o f hypertension in South A frica’s black population (Fritz, 1997; Seedat, 1998). M ore recently, the ram pant epidem ic o f HIV /A ID S threatens to increase the incidence o f stroke enorm ously (Hoffm ann, 1998; Seedat, 1998). About 50% o f all CVA survivors have some residual physical C O RRESPO N D EN C E TO: M rs L.A. Hale School o f Physiotherapy, U niversity o f Otago, P O Box 56, Dunedin, New Zealand J ou rn al abstracts are online at: http://w w w .ptglobal.net/ disability (Ashburn, 1999). U nfortuna­ tely, no data are available on the extent o f residual disability am ongst stroke survivors in South A frica but it is rea­ sonable to conclude that it must be high. Given the above observations, there will probably be an increasing number of people with stroke requiring rehabilitation. T he health care system in South Africa is already pressurised without this added burden (Wazakili and M pofu, 2000). To put the problem into perspective, the Chris Hani (C.H.) Baragwanath Hospital, a 3240 bed hospital has 21 Physiothe­ rapists and eight Physiotherapy A ssis­ tants. T he h o s p ita l’s “N euro T eam ” attend to 90-100 new stroke patients every month (The C.H. Baragwanath H ospital Yearbook, 1998 - 1999). In 1998, the average length o f hospitali­ sation following stroke at this hospital was found to be twelve days (Hale et al, 1998). Following the D epartm ent o f National H ealth’s com m itm ent to support Primary H ealth C are (PHC) in South A frica (1992), the U niversity o f the W itw aters­ rand revised its undergraduate physio­ therapy curriculum in order that it met the requirem ents o f PHC (W allner and Stewart, 1994). Subsequent com m uni­ cation w ith both sta ff and students revealed that they often felt ill equipped to deal with com m unity stroke reha­ bilitation. The appropriateness o f the undergraduate neurorehabilitation course was thus questioned (Hale and Wallner, 1996). H istorically the course had been based on models originating from more developed countries such as the U nited K ingdom and A ustralia, and the query arose as to w hether this adequately p rep ared students to cope in their neurological clinical placements such as the C.H. Baragw anath Hospital or the Soweto clinics. As part o f a research study em anating from this question, it was decided to survey the opinions o f qualified South African physiotherapists regarded as experts by their colleagues in the field o f stroke rehabilitation. W hat did these physiotherapists regard as optim al stroke rehabilitation in South A frica? This paper reports on the find­ ings o f this survey, which was conducted using a Delphi technique. The Delphi technique is a method of gathering consensus information about a certain topic from a group o f experts (Cohen and M anion, 1980; Shepard, 1993; Grbich, 1999). The Rand C orpo­ ration developed it during the 1950s. This corporation decided that face-to-face com m ittee m eetings had several lim i­ 32 SA J o u r n a l o f P h y s io th e ra p y 2001 V o l 57 No 2 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. ) http://www.ptglobal.net/ tations: there was intense social pressure to jo in the majority view; strong opi­ nions tended to override weaker voices; people generally avoided conflict; there were clashes o f personality; status pro­ blems m eant that those further up the hierarchy were heard most often; and there tended to be undue emphasis on consensus or on changing opinions as an outcom e. The corporation felt that in order to get m ore representation from its com m ittee mem bers, the interpersonal interaction would have to be eliminated. The Delphi m ethod was thus developed. T he D elphi is characterised by four aspects: 1) anonym ity o f experts; 2) several rounds o f questionnaires; 3) feed b ack betw een rounds; and 4) som e form o f statistically or descrip­ tively analysed group response so that the strength o f consensus can be seen (Seales and Barnard, 1998). The Delphi technique can be used to collate and analyse the opinions o f experts in a specialised field. This is usually achieved by sending sequential questionnaires to a selected group o f people. R esults o f each round o f ques­ tionnaires are fed back to respondents. Jones et al (1999) successfully used this technique to survey specialist physio­ therapists in the treatm ent o f Parkinson Disease, in order to arrive at a fram e­ w ork o f physiotherapy service delivery for this condition. They rated statements relating to physiotherapy m anagem ent according to the context (where/how), the reasons (why), the actions (what) and the effects (how m easured). Statem ents were then rated with adjectives such as h a p p e n e d /d id n ’t happen, desirable/ undesirable and agreement/disagreement. With the Jones et al (1999) study in m ind, the present D elphi survey explored the follow ing statements: 1. The im portance o f physiotherapy in stroke rehabilitation in South Africa. 2. How a physiotherapeutic service in stroke rehabilitation could be opti­ m ally delivered. 3. W here this service would be best delivered. 4. W hich specific p h y sio th erap eu tic approach or philosophy is preferred for stroke rehabilitation 5. How the effects o f this service deli­ very could best be measured. METHODS The first stage o f the research involved the identification of possible partici­ pants by a peer review process. As the num ber o f physiotherapists in South A frica with a special interest in stroke re h a b ilita tio n is lim ited, all know n interested parties were included. The definition o f “interested” were if the p h y sio th erap ist taught neu ro lo g ical physiotherapy at one o f the eight univer­ sities; if they were post-graduate students in neurological physiotherapy; if they w orked or had w orked in neurology w ards o r n eurological reh ab ilitatio n departm ents; or if they belonged to the Neurological Rehabilitation Group o f the South African Society o f Physiotherapy. In addition, physiotherapists were selected if it was know n that they had an interest in neurological physiotherapy as applied in the com m unity and if it was felt they could make a contribution to the process. Participants were asked if they knew o f any other physiotherapists whom they felt could contribute. There is no consensus as to the optimal size o f the “expert” panel (W illiam s and Webb, 1994), and the matter remains unresolved (Walker and Selfe, 1996). In this case, an attem pt was m ade to obtain the greatest am ount o f “experts” in the field being investigated and from as many areas in the country as possible. A questionnaire was mailed to the respondents by post and included an addressed, stam ped return envelope. T he follow ing questions were asked in the first round o f the survey: 1. P lease state y our opinion o f the im portance o f phy sio th erap y in stroke rehabilitation. 2. In the context o f South Africa, how do you feel stroke rehabilitation can best be delivered in order to achieve an optim al service? 3. T here are m any techniques (for exam ple, Bobath, PNF, and M otor R elearning P rogram m e) d escribed for stroke rehabilitation. In your opi­ nion, w hich o f the techniques, if any, do you feel provides the best practice on which stroke rehabilita­ tion in South A frica can be based. 4. H ow best can we m easure the outcom e o f physiotherapy in stroke rehabilitation in South Africa? T he questions in the Round One questionnaire generated a num ber o f statem ents in response. T hese state­ ments were collated and used to develop a second questionnaire. This second q u estio n n aire w as then sent to the participants. The second questionnaire attem pted to obtain consensus o f opi­ nion on the statem ents generated from Round One. Participants had an oppor­ tunity to see how other experts responded to the open-ended questions, to ensure that their own responses were included in an accurately identified category, and to suggest additional responses that appeared to be missing. The group was also invited to rank the responses to each question in the first questionnaire using a Likert-type scale. The categories: Strongly Agree, Agree, N either A gree nor Disagree, Disagree, and Strongly Disagree were used. The Round Two questionnaires were analysed to obtain the percent o f responses each statement on the questionnaire generated. As consensus was achieved after Round Two, it was felt that a further round o f questionnaires would not serve any specific purpose. RESULTS Round One o f the Delphi survey was sent to 41 physiotherapists. Twenty- seven questionnaires w ere com pleted and returned (a response rate o f 66%). Round Two Questionnaires w ere sent to all 41 participants. Those who had not responded in the first round were invited to jo in in the second round. A total o f 32 questionnaires were com ­ pleted and returned: a response rate o f 78%. As the actual responses were too m any to publish, they have been sum m arised below: 1. Please stale your opinion of the importance of physiotherapy in stroke rehabilitation below. T he statem ents generated by this question produced an extrem ely high level o f consensus am ongst participants. It was felt that phy sio th erap y was extrem ely im portant in all stages o f stroke rehabilitation (acute through to chronic). The im portant roles that phy­ siotherapists play include educating the SA J o u r n a l o f P h y s io th e ra p y 2001 V o l 57 No 2 3 3 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. ) patient, the family, the caregivers and the public; helping people with stroke to re-integrate back into the society (as one respondent stated: “in-patient rehabili­ tation helps patients with stroke adjust to their problem s, and to start interacting with other mem bers o f the rehabilitation team and with people with sim ilar pro­ blem s”); helping patients achieve their m axim al potential and becom e function­ ally independent; helping to improve their quality o f life and assisting clients with stroke to learn to accept their ulti­ mate handicap. Physiotherapists assist the caregivers o f their clients, give clients good hom e program m es, and provide their clients w ith p sychological and em otional support. It was felt that physiotherapy was im portant to improve the quality and norm ality o f m ovem ent and function following stroke and in preventing the secondary com plications o f stroke. Physiotherapists form part o f a team approach. All these statem ents drew agreem ent from more than 77%. Two respondents felt physiotherapists should expand their roles in the com m u­ nity: “there was still not enough follow through o f rehabilitation into the com ­ m unity, and p h y sio th erap ists should be trained in health education,” and “they need to expand their role to involve the w hole community, not only caregivers or family.” Two respondents felt that the South African Society o f Physiotherapy should organise a “Stroke W eek” in order to educate the public with regards to stroke. T he only statem ent that was contro­ versial was whether physiotherapeutic stroke intervention is ideally perform ed in the patient’s home. Twelve partici­ pants (39% ) said that they neither agreed nor disagreed, whereas 17 participants (55%) agreed or strongly agreed. One respondent felt that “treatm ent in the home allows for very specific assess­ ment and guidance o f the patient in their own hom e environm ent.” Only two participants disagreed, one o f whom answered: ’’the role o f physiotherapy is best in a hospital or clinic in the acute and sub-acute stages. Hom e care is not cost effective and it is doubtful w hether our skills could be put to best use in the home environm ent. In the chronic stages follow ing stroke, home exercises, m ain ten an ce and stroke exercises classes in an accessib le com m unity centre are required.” 2. In the context of South Africa, how do you feel stroke rehabilitation can best be delivered in order to achieve an optimal service? There was a 78% agreem ent that physiotherapy assistants and com m u­ nity rehabilitation workers (CRW s) can deliver a service to people with stroke, provided that physiotherapists act as consultants. To quote one respondent: “with the shortage o f physiotherapists in South A frica, physiotherapy assis­ tants and com m unity reh ab ilitatio n workers should be used more effectively, but under the supervision o f physiothe­ rapists.” Education was perceived by most respondents as a crucial means o f ser­ vice delivery, and needed to be aim ed at clients, caregivers, the general public, p h y siotherapy assistan ts and CRWs. Table 1 identifies som e o f the issues which respondents felt education pro­ gram m es should address in a stroke rehabilitation programme. Respondents suggested many methods by which stroke rehabilitation could best be delivered, and these are sum m arised in Table 2. Again there were mixed feelings as to w h eth er reh ab ilitatio n should be perform ed in the client’s hom e or at an institution (45% o f respondents were undecided and 13% o f respondents dis­ agreed with hom e visits). O ne respon­ dent stated: “Travelling is expensive and tim e-consum ing and therefore not for South Africa. It would be better to access patients in com m unity centres. There is basically a lack o f money for rehabilita­ tion follow ing stroke.” One respondent suggested “the use o f mobile rehabilita­ tion units or state funded rehabilitation centres” . M any respondents (64%) felt that physiotherapists in the public health sec­ tor as well as those in the private sector should still provide one-on-one interven­ tions. However, one respondent asserted that: “one-on-one rehabilitation is ideal but unrealistic. One would achieve more by em pow ering caregivers in the acute setting, such as the family or nurses” and that “physiotherapy needs to move out o f institutions and the private sector and concentrate m ore on health prom o­ tion and prevention o f stroke.” There was 100% concordance that early rehabilitation intervention after TABLE 1. Issues to be included in an education programme. Issues to Address Respondent Agreement What is a stroke? 100% The prevention of stroke. 74% Physiotherapists' role in stroke rehabilitation. 96% Prevention of caregiver burn-out. 100% TABLE 2. Methods of service provision Suggested Methods of Service Provision Respondent Agreement Dedicated stroke units 100% Individual treatment 64% Group therapy 64% Home programmess 97% Stroke support groups 90% Community Rehabilitation Workers 94% Primary Health Care Clinics 97% Rehabilitation must be outcome-based and client-centred 93% List of the resources available to stroke victims in all regions 87% 3 4 SA J o u r n a l o f P h y s io th e ra p y 2001 V o l 57 No 2 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. ) TABLE 3. Consensus on techniques advocated as the best practice on which stroke rehabilitation in South Africa can be based. Approach to Rehabilitation % Agreement % Disagreement % No Comment Bobath Approach 49% Motor Relearning Programme 24% Proprioceptive Neuromuscular Facilitation (PNF) 19% Brunnstrom approach 35% (for upper limb rehabilitation) 23 % no comment 29% undecided Alfolter concept . 16% 55% had not heard of this concept Bobath concept and PNF are impractical in SA context 65% Manual Skills 67% Functional exercises, and that the quality of this functional return was important 80% Applying Bobath principles to a client that is seen less than once a week is futile. 33% 36% The Bobath concept should be taught to caregivers and CRWs. 65% Motor Relearning Programme (MRP) could be taught at home or in a group setting. 62% Motor Relearning Programme (MRP) could be utilised by CRWs and caregivers under supervision. 65% The MRP concept utilises a cognitive approach that is impractical when there are language barriers. 27% 38% None of the approaches mentioned were practical in SA. 10% 74% stroke is necessary and that one-on-one rehabilitation should be offered in the acute stage following stroke (84% agree­ ment). An inter-disciplinary approach (97%) should also be utilised. O nce the client has been discharged home, both individual and group therapy should continue (71%). One respondent warned that: “group therapy is not ideal when patients are paying a lot o f money to com e for treatm ent” and thus it should be used effectively. O ne respondent felt that continuing rehabilitation once discharged from hos­ pital negates the principles o f Primary H ealth C are and em pow erm ent o f the individual. O ne statem ent proved to be quite controversial: “R ehabilitation should be functionally orientated, and the quality o f function and m ovem ent may have to be sacrificed.” There was only a 48% agreem ent w ith this statem ent. This response was • sim ilar to that elicited from the assertion that the “w estern” approach to rehabilitation is often not appropriate and is often too complicated for the South A frican context (55% agreement). Seventy-one percent o f respondents agreed with the statem ent that certain patients should be targeted for rehabi­ litation, the rationale being based on p ub lish ed studies o f the prognostic indicators follow ing stroke. O ne respon­ dent felt that this could be construed as discrim inatory practise. There should be parliam entary lobby­ ing to prom ote and improve rehabilita­ tion policies (93%), as well as targeting medical aid funders to supply better enum eration packages for neurological rehabilitation (96%). M ost respondents (96%) felt that in private rehabilitation, b etter planning o f the utilisation of the medical funds available for rehabili­ tation should ensure that funds were available for the later stages o f rehabili­ tation as well. r 3. There are many techniques (for example, Bobath, PNF, Motor Relearning Programme) described for stroke rehabilitation. In your opinion, which of the techniques, if any, do you feel provides the best practice on which stroke rehabilitation in South Africa can be based. This question generated the m ost controversy am ongst respo n d en ts as can be seen in Table 3. However, it was generally felt that a com bination o f ap proaches to reh ab ilitatio n was probably the best strategy to adopt. There was som e agreem ent (42%) that approaches other than the Bobath/ N DT concept w ere difficult to com m ent upon as most training institutes in South Africa taught the Bobath/NDT concept and little was known o f the other methods. Seventy-five percent o f the respon­ dents concurred that m anagem ent of clients should be based on a social model, and should be out-com e based, have functional goals and set tim e frames with which to w ork (87%). One respondent felt that: “the best use m ust be m ade o f the available environment, f o r example, in the home or in the institute, to fa cilita te increased activity which will serve to stimulate their SA J o u r n a l o f P h y s io th e ra p y 2001 V o l 57 No 2 35 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. ) patients to move and involve themselves with their environm ents", and another resp o n d en t concluded: “p h y sio th e ra ­ pists need to be innovative and creative to use their knowledge and skills to best advantage in any situation/context. The principles o f various concepts should be taught, but the physiotherapist should m ove aw ay fro m specific techniques a nd design intervention that suits the specific individual a n d the group or com m unity they are serving. ” 4. How best can we measure the outcome of physiotherapy in stroke rehabilitation in South Africa? M ost o f the respondents felt that im provem ent o f a c lien t’s functional ability and o f their quality o f life were im portant param eters to measure (94%, 96% respectively). They felt that all stroke program m es should have indica­ tors for monitoring and evaluation (94%), and some o f the indicators mentioned were independent function (87%), return to w ork (45%) and re-integration back into the client’s com m unity (65%). Note must be taken o f physical, em otional and social im provem ent (93%). Clients must be regularly assessed, including both functional ability and at an im pair­ m ent level (71 %). However, one respon­ dent felt that “one m ust be careful as to whether one is measuring patient- specific outcom es or program m e evalu­ ation outcomes. Outcomes are also deter­ m ined by the social problems o f the area. ” Specific scales o f m easurem ent were m entioned as being useful and these can be seen in Table 4. O ther tests mentioned by respondents included: the speed o f w alking (45%); the speed, co-ordination and quality of m ovem ent (65%); the use o f specific questionnaires (32%); the use of valid, reliable, effective, easy, quick interna­ tional m otor scales (68% ); and the developm ent o f a South African scale which reflects outcome within the client’s specific environm ent (61%). Assessing caregivers’ opinions as to the success o f intervention was suggested as one m ethod o f gauging outcom e (78%). The use o f photographs and video recordings were other suggestions o f m easurem ent (52%). R eaching realistic goals would be yet another appraisal that could be used (90%). One respondent com m ented that: “it is doubtful whether one scale would be useful in all South A frican situations and communities. It seem s that each specific rehabilitation program m e should use a scale to suit its goals and outcom es ” and another respondent replied: “Do we need to measure it? I f every physiotherapist could treat stroke patients well the o ut­ com es will speak f o r them selves”. There was some concern that func­ tional goals often clash with cultural beliefs (36%) and a few respondents felt that outcom e o f stroke rehabilitation could not be m easured (6% agreed but 74% disagreed). Suggestions for the future were the need for a disability census (55%), and the need for more research, in particular random ised, controlled studies (74%). A further suggestion was: “the depart­ m ent o f Welfare have adapted a well researched disability instrum ent (the D Q 98) to measure in/dependence in elderly people in order to determine whether they require fr a il care. This takes into consideration a biopsycho­ so cia l approach a n d the stress on caregivers. This could be adapted to measure in/dependence in a wide variety o f clients/patients. ” O ne respondent said that unfortunately: “outcomes scales are often difficult to im plem ent due to the lack o f m anpow er and time. ” There was a warning that “over em phasis on measurem ent m ay be threatening f o r patients. ” Two responses sum m arised this sec­ tion: “it is im portant to use international scales, as it is im portant f o r research an d f o r com paring resu lts”, however “scales need to be tailor-made to South A fr ic a ’s cultural issues. ” DISCUSSION Results of the Survey The results o f the survey indicated that physiotherapists believe physiotherapy plays an im portant role in the rehabilita­ tion o f clients with stroke, in all three o f the defined stages: acute, subacute and chronic. Pound and Ebrahim (1997) had sim ilar findings regarding the posi­ tive outlook physiotherapists have on w hat they can offer people with stroke. Ideally m ost respondents in this study would like dedicated stroke rehabilita­ tion units, although a good m any felt rehabilitation should be perform ed in the clien t’s hom e environment. H ow ­ ever, the im practicalities o f home visits w ere h ig hlighted, such as tim e and financial constraints. It was acknow l­ edged that getting the client to a medical facility was difficult due to problems with transport. A com prom ise would be the use o f com m unity centres, for example town halls and churches. M obile rehabilitation clinics were suggested as one m ethod o f reaching out to clients. This m ay be a feasible idea. The Phelophepa Health train has been oper­ ating successfully since 1994, treating nearly 200,000 people around South Africa (Thom, 1998). A mobile reha­ bilitation bus could be used in more localised areas. M ost respondents advocated for the increased use o f physiotherapy assistants or CRW s, but w ith physiotherapists acting as consultants, a view held by other physiotherapy service providers (Taukobong, 2000; Wazakili and Mpofu, 2000), and one which is probably worth exploring further. R espondents felt that support groups should be encouraged. Group therapy itse lf w as co n tro v ersial w ith som e believing it was very useful, and others TABLE 4. Suggested scales of measurement of stroke rehabilitation in South Africa Measurement Scale % Agreement % No Opinion Canadian Neurological Scale (acute stage) 6% 61% Motor Assessment Scale (MAS) (subacute stage) 23% 29% Rivermead Scale (subacute stage) 23% 29% Functional Independent Measure (chronic stage) 45% 23% The Reintegration to Normal Living (RNL) 58% 19% 3 6 S A J o u r n a l o f P h y s io th e ra p y 2001 V o l 57 No 2 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. ) feeling that it may be efficient (to see many clients at once) but not effective in ach iev in g the d esired results. However, structured group therapy can provide positive group dynam ics such as psychological support, com petition and m otivation as well as creating self- reliance am ongst group m embers in a participatory process (Ada, 1999; Stewart and Bhagw anjee, 1999). P ro v id in g individual therapy to clients was still considered important by respondents both in private and govern­ m ent settings. This view how ever may not be possible given the shortage o f physiotherapists and time constraints under w hich physiotherapists have to work, w hich probably prom pted the co m m en t from one resp o n d en t to “one-on-one rehabilitation is ideal but unrealistic. One would achieve more by em pow ering acute caregivers, such as the fa m ily or nurses ”. It would depend on w hat was defined as “individual” therapy, and the frequency and duration o f this therapy. U nfortunately only people with a disability due to stroke w ho present to hospitals where adequate physiotherapy services are available, will in fact benefit from individual the­ rapy. For m any stroke victims, rehabi­ litation is either not available, or only available for a very short time. Follow-up once discharged is impossible in many cases. O ne resp o n d en t felt that: “Physiotherapists need to set ‘id e a l’ goals but also need to reconcile these with the reality o f health care in South Africa, a n d make services realistically available to the larger community, i.e. the best service fo r all, not the ideal ser­ vice f o r a f e w ”. The results o f the survey created a picture o f what is required in a physio­ therapist involved in stroke rehabili­ tation. A ttributes include an intimate know ledge of: stroke (what it is, its prevention and the prognostic indicators o f outcom e); the social model o f reha­ bilitation in order to be client-centred and culturally sensitive; the latest in stroke research to ensure practice is outcom e-based; skills to operate in a team; know ledge o f adult education and teaching skills; and the ability to promote health and wellness. Creative thinking is required in order to make rehabilitation efforts appropriate when w orking in an institute or if giving stroke therapy in a group. None o f these attributes are novel, they are often espoused, and training institutes are addressing these issues. However, it was noticeable in the last two statem ents o f the survey, that m any th erapists had no notion o f many o f the latest techniques or of recent research. This may be due to a lack o f exposure o f therapists, unless they are in an academic situation, to the latest literature, and it might be som e­ thing university departm ents or special interest groups could address. This lack o f exposure is how ever not unique to South African physiotherapists. A recent survey o f physiotherapists’ reasons for selection o f treatm ent techniques in the United Kingdom and A ustralia found that there was a virtual absence o f the use o f journal literature as a basis for selection. Instead there was an over­ w helm ing reliance on formal education, such as initial undergraduate training and practice-related courses (T urner and W hitfield, 1999). The im portant role physiotherapists have in health education was widely acknow ledged, w ith one respondent going so far as to say: “physiotherapy needs to move out o f institutions and the private sector and concentrate more on health prom otion and prevention o f stroke. ” Physiotherapists are often in a unique position to provide preventative education. However, as rehabilitationists, our educational role is far greater. W hile prevention is desirable, our defined role is the “restoration o f optim al levels o f physical, psychological and social ability ..... “ (Andrew, 1987) after the disabling event. We need to educate, to provide advice, w hich would enhance our clients’ optim al return o f functional ability, and thereby, their quality o f life. It was interesting that one respondent felt that continuing rehabilitation once discharged from hospital negates the principles o f Prim ary Health Care (PHC) and em pow erm ent o f the individual. The declaration at A lm a A ta defined PHC as: “essential health care made universally accessible to individuals and fam ilies in the com m unity by means acceptable to them, through their full participation and at a cost that the com m unity and the country can afford” (W HO, 1979). This does not mean that rehabilitation should be discontinued after discharge. R ather it implies that rehabilitationists are required to provide a service in w hich there is equal partner­ ship and ow nership o f the process both with the client and with their community. This means providing a rehabilitation service that the com m unity defines as opposed to that w hich the health profes­ sional feels o u ght to be p ro v id ed (Petrick et al, 1999). This would require a client-centred approach w here the health professional listens carefully to w hat the client w ishes and there is inform ed negotiation o f what the profes­ sional can provide and the likely result o f that intervention. The national govern­ m ent’s W hite Paper on an Integrated National D isability Strategy (1997) is aimed at facilitating the em pow erm ent o f all people with disabilities and pro­ moting their fullest development as equal citizens w ithin society (Stew art and Bhagwanjee, 1999). This may require the physiotherapy profession to recon­ sider their role in the rehabilitation from one o f the “restoration o f optim al levels o f physical, psychological and social ability ..... “ (Andrew, 1987) to one o f enabling people with disability to have increased self-aw areness and self-confidence w ithin their level o f ability (Brohier, 1998). In the book Em pow erm ent o f the Blind, Websons (1997, p .27) w rites: “once w e are empowered, we have personal power and control over our lives. We have po s­ itive self-im ag es th at are affirm ed and sustained.” A num ber o f theoretical a p proaches and m ethod o lo g ies on em p o w erm en t have been d eveloped by social scien tists (S tew art and Bhagw anjee, 1999). These may help health professionals develop the neces­ sary m in d -sh ift from the histo rical medical model o f rehabilitation to the national governm ent’s envisaged reha­ bilitation model set within a fram ew ork o f hum an rights and developm ent (S tew art and B hagw anjee, 1999). Densen described this model o f reha­ bilitation w ithin a social context as “independent living” and defined it as ” a concept, a policy, a set o f community- based services and program m es, and a SA J o u r n a l o f P h y s io th e ra p y 2001 V o l 57 No 2 3 7 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. ) civil-rights movem ent ... it is freedom to participate in the community fully and to have access to housing, transporta­ tion, health care, em ploym ent and edu­ cation .... it is reflective o f a self-deter­ mined and self-directed lifestyle that permits the individual to make m eaning­ ful choices” (Densen, 1988, p. 18). The use o f the N D T/Bobath approach appeared to be the technique or approach well favoured by m any respondents fo r stroke rehabilitation in South Africa. However, the general feeling was that an eclectic approach based on the attainm ent o f functional goals was the best strategy. The im provem ent in the quality of this func­ tional m ovement was deem ed important by many, however one respondent felt that “the quality o f function and m ove­ m ent m ay have to be sacrificed. ” A lthough this proved to be a fairly controversial statement, it may be prag­ matic given the lack o f contact time in stroke rehabilitation. Published litera­ ture on the effects o f rehabilitation after stroke have shown that there are weak associations betw een im proved func­ tional outcom e and 1) the intensity of rehabilitation services, 2) the early onset o f rehabilitation and 3) task-specific therapy services versus more generalised therapy (C ifu and S tew art, 1999; Kwakkel et al, 1999; Rice-Oxley, 1999). H ow ever, most in tervention studies assess the presence or absence o f certain functional skills rather than on the q u ality o f co -ordination patterns o f m ovem ent o f theses skills (Kwakkel et al, 1999). There seems to be some evidence for the use o f physiotherapy to im prove functional ability follow ing stroke, but a paucity o f evidence on the positive effect o f physiotherapy on the quality o f movement. M easuring the outcome o f physio­ therapy in stroke rehabilitation in South Africa did not generate much consensus or particularly strong points o f view. It was generally felt that it was very important to monitor stroke rehabilita­ tion program m es, and that the return of independent function and the im prove­ ment o f the client’s quality o f life were the most im portant indicators. It must be em phasised, however, that any. pro­ gram m e instituted for stroke rehabilita­ tion should be monitored and outcome m easured to ensure the efficacy of such a program m e. In order for stroke reha­ bilitation to exist, and for physiothera­ pists to be involved in it, physiotherapists have to justify their work. Justification will be dem anded not only by the clients and the public but also by all involved funders, whether they are private or governm ental (Turner and W hitfield, 1999). The use o f valid and reliable out­ come measures are a vital com ponent of rehabilitation and should be taught in under-graduate training program m es to ensure the survival o f the profession. The Use of a Delphi Survey The resultant report o f a Delphi survey should indicate areas o f consensus: spread o f opinion, minority arguments and deviant responses. This was cer­ tainly attained in this survey, in spite o f it only consisting o f two rounds. Further questionnaires would not have added to the body o f knowledge obtained or formed any further consensus o f the more controversial issues. Because the experts do not interact, the use of the Delphi prevents the poten­ tial problems that often occur in small, consen su s-d riv en groups (S hepherd, 1993). The success depends on the dili­ gence and m otivation o f the respondents in form ulating considered opinions, and in their proficiency in com m unicating. The respondents o f this survey showed a great deal o f motivation with a good return rate to both rounds. M any o f them went to a great deal o f trouble com m u­ nicating their thoughts. The Delphi method has been criticised for its lack of “ methodological rigour” , for exam ple, individuals may be influ­ enced to conform because views are fed back to each member. The study may also fail to get a representative sample of experts. Different panels with similar expertise may produce different results (lack o f reliability) and the validity o f the results may be influenced by the response rates. In addition there is little agreem ent on the optim um num ber for the panels or on what constitutes “consensus” (Seales and Barnard, 1998). Jones and H unter (1995) argue that the existence o f a consensus does not mean that the “correct” answ er has been found. The Delphi technique has an equal chance of producing collective ignorance as well as collective wisdom. The validity and reliability o f the Delphi m ethod have not been well-eva- luated (W alker and Selfe, 1996). One m ethod o f achieving a high concurrent validity is by achieving a consensus (W illiams and Webb, 1994), however there is no standard threshold for con­ sensus. Some authors have arbitrarily defined their own limits, for example, B oyce et al (1993) set consensus at 66%, and M cK enna (1994) suggested 51%. In this survey, if the categories “agree” and “strongly agree” are com bined into “agreed” , and those o f “disagree” and “strongly disagree” into “disagreed”, then there was a consensus o f over 70% in 62 out o f the 100 statements made by the participants. Statem ents reaching between 50% - 70% consensus num ­ bered 21. O f those that were under 50% consensus, six statements had a “no opinion” response greater than 56%. The majority o f the controversial state­ ments were generated by the third and fourth questio n s w here p articipants had to com m ent on which treatment technique and m easuring tool they felt were appropriate for stroke rehabilita­ tion in South A frica. M any had no opinion with regards to statements in these areas. Thus it was felt that although consensus in these areas was not as high as the other sections, continued rounds o f questioning would not produce any further useful information. M ost studies limit the num ber o f rounds to two or three to prevent respondent fatigue and increased attrition (W hitman, 1990). In spite o f this method o f uninflu­ enced, equal and anonym ous participa­ tion, there has been debate regarding its ability to reduce psychosocial variables (O ’Brien, 1978). Johnston (1970) was concerned on the capabilities o f the “experts” to develop future, alternative solutions to problem s whose current practice they well may have influenced. The Delphi method has been criti­ cised for its m ethodology that does not allow participants a chance to discuss th eir responses (Jones and H unter, 1995). M any o f the respondents in this survey felt that they would have liked to clarify their rating o f statem ents. 38 SA J o u r n a l o f P h y s io th e ra p y 2001 V o l 57 No 2 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. ) Thus the results o f this survey may not produce the “correct” answers to the problem s facing stroke rehabilitation in South Africa. They can only be used to generate ideas and to gauge opinions. Readers must guard against placing too much reliance on quantitative conclu­ sions from the results of Delphi surveys (Jones and Hunter, 1995). These sur­ veys are generally used to produce consensus statements that must then be tested with more rigorous m ethodologi­ cal m ethods. In-depth interviews may have been a better alternative to the Delphi in this study. A deeper under­ standing o f the “experts” thoughts would have been attained, allowing them the chance to clarify and explain their ideas more accurately. However, due to finan­ cial constraints, this would have limited the num ber o f “experts” interviewed. Response Rate The response rate of Round One (66%) was deem ed adequate, given that a cou­ ple o f the selected physiotherapists were out the country at the time, and that a few o f the envelopes were returned with the address o f the recipient unknown. In addition, the survey was conducted at a time when the South African postal services were still considered ineffi­ cient. Unfortunately, not all o f the iden­ tified “experts” had e-mail addresses. A response rate o f 70% or more is consi­ dered acceptable for a postal question­ naire in a Delphi survey (Bork, 1993). The second round questionnaire gener­ ated a higher response rate o f 78%. This survey invited non-respondents from the first round to contribute to the second round if they so wished. This is an acceptable practice (Walker and Selfe, 1996). CONCLUSION This article presents the results o f a Delphi survey o f physiotherapists who have experience or an interest in neuro­ rehabilitation. It summarises their opi­ nions o f what constitutes optim al stroke rehabilitation in South Africa. A Delphi technique was used in an attem pt to obtain the participating p h y sio th era­ pists’ independent views, without these opinions being influenced by the collec­ tive thought o f a com m ittee. This paper does not offer a solution o f what should be taught regarding stroke rehabilitation at an undergraduate level, but is rather a synopsis o f w hat q u alified p h y sio ­ therapists currently view as optim al practise, and in this role, the survey has generated some ideas which could be debated further. Are these ideas realistic? Is what is viewed as an optimal approach able to provide rehabilitation to all people with stroke who require it within the socio-econom ic constraints prevailing presently in South Africa? O r do we require more innovative ideas, perhaps a fresh approach to stroke rehabilitation in South Africa? The only way these questions can be answered is if data regarding the incidence o f disability following stroke and the outcom e of intervention, are closely monitored and reported. REFERENCES A da L, M ackay F, H eard R 1999 Stroke rehabilitation: does the therapy area provide a physical ch allen g e ? A ustralian Journal o f Physiotherapy 45: 33 - 39 A nd rew s K 1987 R e h abilitation o f the O lder A dult. I st e dition, L ondon: E dw ard A rnold A shburn A 1999 Physical recovery follow ing stroke. P h ysiotherapy 83(9): 4 80 - 490 Bork C E 1993 R esearch in Physical Therapy. 1st e dition, P h iladelphia, J B L ippincott C o m ­ pany. B o y c e W, G o w la n d C , R u ssell D 1993 C o n se n su s m ethodology in the d e v elopm ent a nd c o n te n t v alid atio n o f a gross perform ance measure. Physiotherapy C anada 45(2): 9 4 - 100. B rohier W G 1998 E m p o w erm en t or im po­ verishm ent. A ction A id D isability N ew s 9(2): 5 6 - 5 8 . C h ris Hani B arag w a n a th H ospital Yearbook 1998 - 1999 Johannesburg: T ransw orld G roup. Cifu D X, S tew art D G 1999 Factors affecting fu n c tio n al ou tco m e after stroke: a critical re v ie w o f re h a b ilita tio n in te rv e n tio n s. A rchives o f P hysical M edicine and R e h a ­ b ilitation 80: s-35 - S-39. C ohen L, M anion L 1980 R esearch M ethods in E ducation. L ondon; C room Helm. D ensen C R 1988 In d ependent living: public p o lic y issu e s. A m e ric a n R e h a b ilita tio n 1: 1 2 - 16. Fritz V U 1997 S troke incidence in South A frica. South A frican M edical Journal 87(5): 5 8 4 - 5 8 5 . G rb ic h , C. 1999. Q u a lita tiv e R e se arc h in H ealth; an introduction. 1st ed., A ustralia: A llen and U nw in, pp 116 - 120. H ale L A , E ales C J 1998 R e co v e ry o f w a lk ­ ing function in stroke patients after m inim al rehabilitation. P hysiotherapy R esearch In ter­ national 3(3): 1 9 4 - 2 0 5 Hale L A, W allner P J 1996 T he c h allenge o f s e rv ic e p ro v isio n in S o u th A fric a for patients with hem iplegia. Physiotherapy 82(3): 1 5 6 - 158 H offm ann M W 1998 T he D urban stroke d ata bank w ith special e m p h a sis on h ig h er cortical function deficits [thesis]. D urban: D e partm ent o f N eurology, U n iversity o f N atal. Joh n sto n D 1970 F o recasting m ethods in the social sciences. T echnological F o re c astin g and Social C hange: 173 - 187 Jones D, Plant P, L ovegreen B 1999 A fram e ­ w ork for p h y siotherapy service d e liv e ry in P arkinson disease [abstract]. P roceedings o f the 13th International C ongress o f the W orld C o n federation for Physical T herapy: 128 Jones J, H unter D 1995 C o nsensus m ethods fo r m edical and h e alth se rv ice s re sea rc h . B ritish M edical Journal 311: 376 - 380 K w akkel G, W agenaar R C, K oelm an T W 1997 E ffects o f intensity o f re h abilitation after stroke: a research synthesis. Stroke 28: 1550 - 1556 M cK enna H P, 1994 T he D elphi technique: a w o rthw hile research a p proach fo r nursing? Journal o f A d v a n ce d N ursing 19: 1221 - 1225 M inistry in the O ffice o f the P resident 1997 W h ite P a p e r on an In te g ra te d N a tio n a l D isability Strategy. G auteng, South A frica N e u ro lo g ic al A sso c ia tio n o f S outh A frica. 2000 S troke therapy c linical guideline. South A frican M edical A sso c ia tio n - N eurological A ssociation o f South A frica S troke W orking G roup. S outh A frican M edical Jo u rn a l 90 (3 Pt 2): 2 76 - 278, 2 8 0 - 2 9 8 . O ’B rien P 1978 T he D elphi tec hnique: a re v ie w o f th e re se a rc h . S o u th A u s tra lia Jo u rn a l o f E d u cation R esearch 1(1): 57 - 75 SA J o u r n a l o f P h y s io th e ra p y 2001 V o l 57 No 2 3 9 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. ) P etrick M , H o m e r S, E vans R 1999 A re th era ­ pists a w are o f the needs o f p e ople w ith d is ­ a b ilitie s ? S o u th A fric a n Jo u rn a l o f P h y sio th e rap y 55(1): 26 - 28 P ou n d P, E brahim S 1997 R e d efin in g ‘doing so m e th in g ’: H e alth P ro fe ssio n a ls’ view s on th e ir ro le in the c a re o f stro k e pa tie n ts. P h y sio th e ra p y R e se arc h In tern a tio n a l 2(2): 1 2 - 2 8 . R ice -O x ley M 1999 E ffec tiv e n ess o f brain injury re habilitation. C linical R e h ab ilitation 13(suppl 1): 7 - 24 S eales J, B arnard S 1998 T h era p y R esearch: P ro c e sse s a nd p ra c tic a litie s . O x fo rd : B u tte rw o rth -H e in e m an n , 1 st edition: 3 6 - 4 0 S eedat Y K 1998 T he pre v alen c e o f h y p e r­ tension a nd the status o f c ard io v ascu la r health in S outh A frica. E thn D is 8(3): 394 - 397 Shepard K F 1993 Q u e stio n n aire design and use. In: R esearch in Physical Therapy. B ork C E (editor). 1st e dition, P hiladelphia: J.B . L ippi- ncott C om pany: 176 - 204 S te w a rt R, B ha g w a n je e A 1999 P rom oting g ro u p e m p o w erm en t a nd self-re lia n c e through particip a to ry research: a c ase study o f people w ith p h y sic a l d isa b ility . D is a b ility a nd R eh ab ilitatio n 21(7): 338 - 345 T aukobong N P 1999 T he role o f the c o m m u ­ nity b a se d re h abilitation w o rk e r w ithin the p ri­ m ary h e alth care service o f the O di D istrict. S outh A frican Journal o f P h y sio th e rap y 55 (1): 1 9 - 2 2 T h e D e p a rtm e n t o f N a tio n a l H e a lth a nd P o p ulation D e velopm ent. 1992: A new health d ispensation fo r S outh A frica. T hom A 1998 H ealth T rain: a rural winner. T he Star Newspaper, Johannesburg, July 21: 13 T u rn er P, W h itfield T W A 1999 P h y sio ­ th e ra p is ts’ reaso n s for selection o f treatm ent te c h n iq u e s: a c ro s s -n a tio n a l survey. P h y ­ sio th era p y T h eo ry a nd P ra c tice 15: 235 - 246 W alker A M , Selfe J 1996 T he D elphi m ethod: a useful tool for the a llied h ealth researcher. B ritish Journal o f T h era p y and R e h abilitation 3(12): 677 - 681 W allner P J, S te w a rt A 1994 A com m u n ity a p p ro a c h to p h y sio th era p y . S o u th A frica n Journal o f P h y sio th e rap y 50(3): 54 - 56 W azakili M , M pofu R M B 2 000 P h y sio ­ th era p y service p ro v id e rs’ view s on issues o f a ssista n ts : are p h y sio th e ra p y a ssista n ts re q u ire d ? S outh A frican Journal o f P h y sio ­ th era p y 56(4): 22 - 25 W ebson W A 1997 E m p o w erm en t o f the B lind. A H a n d b o o k for O rg anisations o f and fo r the B lin d a nd V isually Im paired. W atertow n: W orld B lind Union Institutional D e v e lo p m e n t P ro je c t, T h e H ilto n /P e rk in s Program . W hitm an N 1990 T he co m m itte e m eeting a lte rn a tiv e : u sin g th e D e lp h i te c h n iq u e . C a n ad ian Journal o f N ursing A dm inistration 20(6/7): 3 0 - 36 W illiam s P L, W ebb C. 1994 T he D elphi tec h ­ nique: a m eth o d o lo g ic al d iscussion. Journal o f A d v a n ce d N ursing 19: 8 0 - 1 8 6 W orld H ealth O rganisation 1979 H ealth for A ll. S eries N o. 2. S trateg ies fo r health fo r all by the y e ar 200. G eneva: W H O . ^ ■ l o c o f Locum and Perm anent vacancies in the U K for: P h y sio th e ra p ists Occupational Therapists Speech & Language Therapists UK OFFICE, A N G E L A S H A W ASSOCIATES Call us fre e fro m South A frica on: 0800 990 767 SA OFFICE, D A N IE L KOTZE Call: 021 975 1556 call: + 4 4 (0 )2 0 -8 5 5 4 7691 fax: + 44 (0)20-8554 9900 email: locums@ asa-locums.co.uk □ □ □ 9 S 9 : :□ locums a member of the maleh group M o n tro s e House, 412-41 6 Eastern A v e n u e , Ilfo r d , Essex, IG2 6NQ, UK. GOVERNMENT OF THE SULTANATE OF O M A N M IN ISTRY OF HEALTH Tax-free s a la r y + accom m odation + 4 8 d a y s' paid holid ay + excellen t benefits PHYSIOTHERAPY SUPERVISOR Applications are invited for the above posts by the Ministry of Health in the Sultanate of Oman. O m an is a progressive country with breathtaking scenery and m a ny opportunities far outdoor activities as swimming, scuba diving, p hotography etc. Requirements: • Degree ar Diplom a in Physiotherapy followed by at least 6 y e a rs' post degree/diplom a experience with at least two years at a senior level with a relevant validated postgraduate course. The successful candidate will underlake the supervision and education o f junior staff and be on active participant in an ongo ing inservice education program m e. Salary & Scale: R .0 .7 8 7 / - per month with onnuol increments of R .0 .2 0 / - per month to m axim u m of scale, which is R .0 .9 8 1 / - per month. One Riyol O m oni is equivalent to 2.57 U.S. Dollars approximately. Benefits include free, furnished (hard furnishings only) air-conditioned occommodation. Economy class air tickets ore provided to em ployee an initial appointment and on final exit. 7 5 % af the cast o f on Economy Class air ticket is paid for onnuol leave travel once during the contractual year. Paid onnuol leave is 4 8 days. Allowances will be given for transport, water and electricity when these ore not provided free by the Ministry. Medical treatment within the Sultanate is free. The contracts are for one ye ar and ore renewable annually if agreeable to both parties. Please apply with the nam es of 3 referees, a full curriculum vitae, copies af professional and experience certificates and a recent passport-sized photograph to: Director of Personnel Affairs Ministry of Health Post Box 393, Muscat, Postal Code 113 Sultanate ol Oman. 4 0 SA J o u r n a l o f P h y s io th e ra p y 2001 V o l 57 No 2 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. ) mailto:locums@asa-locums.co.uk