Exercise therapy and cognitive behavioural therapy to improve fatigue , daily activity performance and quality of life in Postpoliomyelitis Syndrome : the protocol of the

Background: Postpoliom yelitis Syndrom e (PPS) is a com p lex o f late onset neurom uscu lar sym ptom s w ith new or increased m uscle w eakness and m uscle fatigab ility as key sym ptom s. Main clin ical com p la ints are severe fatigue, deterioration in functional ab ilities and health related quality o f life. Rehabilitation m anagem ent is the m ainstay o f treatm ent. Tw o different therapeutic in terventions m ay be prescribed (1) exercise therapy or (2) cogn itive behavioural therapy (CBT). H ow ever, the ev idence on the effectiveness o f both in terventions is lim ited . The prim ary aim o f the FACTS-2-PPS trial is to study the efficacy o f exercise therapy and CBT for reducing fatigue and im proving activ ities and quality o f life in patients w ith PPS. Add itionally , the working m echan ism s, patients' and therap ists' expectations o f and experiences w ith both in terventions and costeffectiveness w ill be eva luated . M ethods/Design: A m ulti-centre , s ing le-b linded , random ized contro lled trial w ill be co nducted . A sam ple o f 81 severe ly fatigued patients w ith PPS w ill be recru ited from 3 d ifferent un iversity hosp itals and the ir affiliate rehab ilitation centres. Patients w ill be random ized to one o f th ree groups i.e. (1) exercise therap y + usual care, (2) CBT + usual care, (3) usual care. A t base line , im m ed iate ly post-in tervention and at 3and 6m onths fo llow -up , fatigue , activ ities, qua lity o f life and secondary ou tcom es w ill be assessed. Costs w ill be based on a cost q uestionnaire , and statistical analyses on GEE (generalized estim ated equations). A nalysis w ill also co nsider m echan ism s o f chang e during therapy. A responsive eva luation w ill be conducted to m on ito r the im p lem entation process and to investigate the persp ectives o f patients and therap ists on both in terventions. Discussion: A m ajor strength o f the FACTS-2-PPS study is the use o f a m ixed m ethods design in w h ich a responsive and econom ic evaluation runs parallel to the trial. The results o f this study will generate new evidence for the rehabilitation treatm ent o f persons w ith PPS. Trial registration: Dutch Trial Register NTR1371. * Correspondence: S.Koopman@amc.uva.nl department of Rehabilitation, Academic Medical Centre, University of Amsterdam, Amsterdam, The Netherlands FACTS-2-PPS trial Fieke S Ko o pm an1*, Anita Beelen1, Karin H Gerrits2, Gijs Bleijenberg3, Tineke A A b m a4, Marianne de Visser5, Frans Nollet1 B io Med Central © 2010 Koopm an et al; licensee BioMed Central Ltd. This is an O pen A ccess article distributed under the term s of the Creative C o m m o n s Attribution License (http://creativecom m ons.org/licenses/by/2.0), w hich perm its unrestricted use, distribution, and reproduction in any m edium , provided the original w ork is properly cited. Koopm an et al. BMC Neurology 2010, 10:8 http://www.biomedcentral.com/1471-2377/1078 Page 2 of 10


Background
Poliomyelitis anterior acuta is an acute viral disease that attacks the anterior horn cells of the spinal cord and the motor neurons of the lower brain stem resulting in flac cid paresis or paralysis. Usually there is partial and sometimes complete recovery from the self-terminating disease. However, many people with a history of polio myelitis report late onset neuromuscular symptoms and a decline in functional abilities. These late symptoms are referred to as Postpoliomyelitis Syndrome (PPS) and include new or increased muscle weakness, abnormal muscle fatigability, generalized fatigue, muscle atrophy, muscle and joint pain, muscle cramps and cold intoler ance [1].
The prevalence of PPS has been reported from 15% to 80% of all patients with previous paralytic polio depend ing on the criteria applied and population studied [2]. Nearly 60% of a sample of Dutch survivors of the 1956 polio outbreak experience late onset polio sequelae [3]. In Western countries, where the large epidemics date back to the 1940s and 1950s, many polio survivors are now experiencing progressive complaints related to PPS. The W orld Health Organization (W HO) estimates that 10 to 20 million polio survivors are alive worldwide, and some estimates suggest that 4 to 8 million of them may develop PPS [4]. Although, the efforts of the Global Polio Eradication Initiative of the W H O initiated in 1988, led to an enormous reduction in the number of acute polio cases globally ever since, polio is still a rele vant problem. In W estern Africa and in South Asia polio is still endemic and spread of the virus from these countries causes new outbreaks in countries that were certified as polio free. Therefore, new cases of PPS can be anticipated in the coming decades.
Fatigue is one of the most frequent complaints of PPS [5][6][7] and it is typically described as tiredness or lack of energy that increases with physical activity and decreases with rest [8]. In a study on disability and health problems in 76 Dutch patients with PPS, 78% of the subjects selected fatigue as their major problem [9]. Subjects with PPS experience higher levels of fatigue than healthy controls [10,11]. Fatigue has a negative impact on activities of daily living and there is evidence that post-polio related fatigue is an important factor for the reduced health related quality of life (HRQoL) in polio survivors [12].
In current practice, rehabilitation management is the mainstay of treatment for PPS. Rehabilitation aims to improve patients' capacities to perform activities of daily living and adapt performance (i.e. actual behaviour) to the available capacities. W e expect that a reduction of the imbalance between patients' capacities and perfor mance will lead to a reduction in fatigue and improvement in activities and HRQoL. There are two possible approaches to achieve this goal; exercise ther apy and cognitive behavioural therapy (CBT). However, evidence to support the effectiveness of either approach is still limited.
The insufficient evidence to support exercise consists of contradictory and incomplete information in the lit erature. O n the one hand, PPS patients are advised to avoid muscular overuse and intensive training as this could worsen symptoms such as muscle weakness and fatigue and provoke a further loss of muscular strength [1]. O n the other hand, physically active PPS patients were found to have less symptoms of fatigue than sedentary patients [13]. It is unclear whether symptoms of fatigue are cause or result of physical inactivity. A systematic review on exercise therapy for neuromuscular diseases included ten studies on muscle strengthening exercises and aerobic exercises in PPS of which five demonstrated significant positive effects on muscular strength and aerobic capacity without generating any adverse effects. However, all these studies had to be qualified as having insufficient or limited methodological quality [14]. This led to the conclusion that there is insufficient evidence for the effectiveness of exercise for patients with PPS and that future, preferably multi-cen tre, studies are needed. A recent study on the short term effectiveness of home-and hospital-based aerobic exercise showed improvement on fatigue and quality of life [15]. However results were not compared with a control group and long-term effectiveness was not eval uated in this study.
There is broad evidence for the effectiveness of CBT in reducing fatigue in chronic fatigue syndrome, fatigued post cancer survivors and m ultiple sclerosis [16][17][18][19]. However, there are no studies of CBT for fatigue in PPS.
The evidence for a cognitive behavioural approach in PPS is currently limited to results from an uncontrolled pilot study in which cognitive behavioural strategies are incorporated in a comprehensive multidisciplinary reha bilitation program [20]. Although significant reductions in fatigue symptoms were found, it is unclear whether these effects can be ascribed to the cognitive behavioural components of the intervention.
We anticipate that exercise therapy and CBT are both effective in reducing fatigue and improving activities and H R Q oL in patients with PPS compared to the usual care. The FACTS-2-PPS study aims to give insight in the efficacy of both interventions for patients with PPS. As secondary outcomes, the working mechanisms, patients' and therapists' expectations of and experiences with both interventions and cost-effectiveness will be evaluated. These aims have led to the following research questions: (1) Does exercise therapy for patients w ith PPS reduce fatigue and improve activities and HRQoL as compared to usual care? (2) Does CBT for patients with PPS reduce fatigue and improve activities and HRQoL as compared to usual care? (3) W hat are the generic and disease-specific deter minants for treatment success of exercise therapy and CBT? (4) W hat are patients' expectations of and experi ences with exercise therapy and CBT? (5) W hat is the cost-effectiveness of exercise therapy and CBT compared to usual care?

Study design
A multi-centre, single-blinded randomized controlled trial (RCT) with 6 months follow-up will be conducted to evaluate the efficacy of exercise therapy and CBT compared to usual care in patients with PPS ( Figure 1).
To gain insight into patients' expectations and experi ences, a responsive evaluation will be conducted. To determine the cost-effectiveness, an economic evaluation will be conducted alongside this RCT. The study proto col was approved by the Medical Ethics Committee of the Academic Medical Centre (AMC) and all participat ing centres granted approval to participate.

Study population
All patients will be recruited from outpatient clinics of the A M C Amsterdam; the University Medical Centre Utrecht (UMCU); the University Medical Centre Nijme gen (UMCN) or the affiliate rehabilitation centres. Case histories are checked to screen for potentially eligible patients. Patients willing to give signed consent are evalu ated by a physician to check in and exclusion criteria ( Table 1). The sample size was calculated for this threearm trial based on the comparison treatment (exercise therapy or CBT) versus usual care, with equal allocation across treatment arms and 4 repeated measurements (at entry (pre-treatment), at 4 months (post-treatment), at 3 F ig u re 1 Stu d y design. (2) severe perceived fatigue (CIS-fatigue > = 35) [47] (3) age between 18 and 75 years (4) life-expectancy longer than one year (5) walking-ability at least indoors with or without a walking aid (6) ability to cycle on a cycle ergometer against a load of at least 25 Watt Exclusion criteria (1) use of psychotropic drugs or other psychiatric treatment (2) clinicaldepression (BDI-PC > 6) [48] (3) disabling co-morbidity interfering with the intervention programs or influencing outcome parameters (including cardiopulmonary disease, epileptic seizures, poorly regulated diabetes mellitus) (4) respiratory insufficiency (FVC < 50%pred or CO 2 retention) or assisted ventilation (5) cognitive impairment (6) insufficient mastery of the Dutch language (7) pregnancy months follow-up (short-term follow-up) and at 6 months follow-up (long-term follow-up), respectively), with an estimated correlation coefficient of the repeated measurements of rho = 0.79 (based on unpublished data from a reproducibility study in 37 PPS patients), a clinical relevant improvement of 8 points on the CIS-fatigue scale (for both treatments) [17] and an estimated standard deviation in each treatment group of 9.3. The total sam ple size needed to detect this difference on a 5% level of significance (two tailed) with a power of 90% was 24 sub jects in each group [21]. We expect a drop-out rate of maximal 10%, based on a previous trial in this patient group [22]. Therefore 81 patients will be recruited.

Randomization and blinding
Patients will be assigned to the rehabilitation centre that is nearest to their home village. Randomization will be stratified per centre. Patients fulfilling both inclusion and exclusion criteria will be random ized to one of three groups: (1) exercise therapy + usual care, (2) CBT + usual care, or (3) usual care. The random ization scheme is computer-generated and will be done by creating random blocks of sequences with variable block sizes of 3 and 6. The investigator who will perform the randomization will inform the patient and, in case of allocation to exercise therapy or CBT also the therapist of the group allocation.
The investigators that are responsible for the inclusion will be blinded. Outcomes will be assessed by blinded and independent outcome assessors. At the beginning of the assessment patients will be instructed not to reveal their group allocation to the investigators. To evaluate the success of blinding, the investigator will be asked at each measurement, to guess the treatment assignment of the patient, with options 'CBT', 'exercise therapy', 'usual care only' and 'unknown'. Analyses will be per formed blinded for treatment allocation.

Usual care
The patients in the control and intervention groups will all receive usual care. Usual care for PPS patients may include usage of assistive devices, orthoses, physical therapy, and medication. Patients will not be restricted in their activities. Co-interventions will be monitored throughout the study.

Exercise therapy
Exercise therapy is designed specifically to enhance phy sical capacity. The intervention has a duration of 16 weeks and consists of (1) a home-based aerobic training program three times weekly and a (2) supervised group training consisting of muscle strengthening and func tional exercises once a week. The therapy will be super vised by specifically trained physiotherapists. A n unblinded member of the project group will supervise the therapists and perform integrity checks on each treatment location.
(1) The home-based training program consists of indi vidually tailored aerobic exercise on a cycle ergometer. Patients will be supplied with a cycle ergometer, a log book with training instructions and a training scheme at home. Heart rate will be continuously measured with a heart rate m onitor. Furthermore, in the log book patients will document the num ber and duration of treatment sessions, the training load, perceived exertion on the Borg Rated Perceived Exertion Scale [23] and possible complaints after the training session. Training intensity will be gradually increased from 60% of the Heart Rate Reserve (HRR) to 70% HRR, in accordance with the American College of Sports Medicine guide lines for aerobic training in healthy adults [24] and per sons with chronic diseases and disabilities [25]. The duration of the training sessions w ill gradually be increased from 20 to 30 minutes per session and ses sions are divided into prescribed exercise bouts which are interspersed with short rest periods of unloaded cycling. The duration of exercise bouts will be gradually increased from 2 minutes at the start of the program to 13 minutes at the end of the program.
Warming-up and cooling-down consists of 5 minutes unloaded cycling. Feasibility of the training schemes is weekly checked by one of the therapists by reading out the heart rate monitors and checking the log books. W hen necessary, adjustments to the training schemes will be made.
(2) The supervised group training consists of individu ally tailored muscle strengthening exercises and functional exercises in 1-hour group sessions. Sessions will be divided into a 5-minute warm-up period of aero bic exercises, 30 minutes of muscle strengthening, 20 minutes functional exercises and a 5-minute cool-down period. Muscle strengthening exercise selection and dosage will be determined by the therapist at a separate visit to the treatment location before the start of the program. Muscle groups and accompanying exercises will be selected based on patients MRC-scores [26] and reported problems in daily activities. Exercises are selected based on the best expected effects on physical functioning.
Muscle groups with M RC scores less then 3 are not selected and quadriceps muscles are always selected when this muscle group has a M RC score of at least 3. Training load will be based on the 1-RM test [27]. In the first 4 weeks, training load will be 50% of 1-RM, each exercise consists of 3 sets with 2 minutes rest between a set and num ber of repetitions w ill be increased from 8 to 20 during 4 weeks. W he n the desired number of repetitions with the current load is reached, training load is increased in week 5 and 9 to 60 and 70% respectively, and the number of repetitions is increased in the same manner as prescribed for the first 4 weeks [28]. Cognitive behavioural therapy CBT will be directed at frequently reported perpetuating factors of fatigue in slowly progressive neuromuscular disorders [29]. They may involve dysfunctional cogni tions with respect to the disease itself, to pain and to fatigue [30], dysfunctional attention to pain or fatigue symptoms, deregulation of sleep [30,31], deregulation of physical, social and/or mental activities [30,31], and low social support and negative social interactions [32]. For each factor a standardized module is available as part of the intervention. Because of the variability of relevant perpetuating factors in PPS patients, therapy will be cus tomized to each individual. To determine which m od ules will be necessary, each perpetuating factor will be measured with specific questionnaires ( Table 2). The number of sessions will be determined by the number of modules used and will vary between 12 and 16 ses sions, each with a duration of 1 hour during a 4 months period. Experienced cognitive behavioural therapists will treat the patients. The therapists will be trained in the protocol and the use of instruments to determine which module should be included in the therapy. Role-playing is an important part of this training. In the first CBT session individualized concrete behavioural goals of the therapy are formulated by the patient with help of the therapist. These goals are formulated in terms of con crete behaviours the patient wants to perform if he is not extremely fatigued anymore. Appendix 1 gives an overview of cognitive and behavioural techniques that

Low social support and
Social Support Inventory-Interactions negative socialinteractions (SSL-I_08) [60] Social Support Inventory-Discrepancies (SSL-D_08) [60] can be used during CBT for each perpetuating factor of fatigue.

Compliance and attrition
Compliance will be assessed by recording the number of treatment sessions (CBT or group exercise sessions) attended and, for the patients randomized to exercise therapy, also the total time spent to aerobic exercise on the bicycle ergometer at home will be recorded in a log book.

Efficacy and working mechanisms
Outcome measures are presented in Table 3. Our pri mary outcome measure is fatigue. Additionally, daily activity performance and HRQoL will be evaluated. Sec ondary outcomes are categorised in accordance with the International Classification of Functioning (ICF) [33] on the level of body functions, activities and participation and personal factors. As potential effect modifiers the following parameters will be studied; demographic vari ables, main complaints of PPS, disease severity, co-mor bidity, serum creatine kinase activity and the ICF environmental factors; social support, H R Q oL of the partner, coping of the partner and caregiver burden. Two secondary outcome measures on the level of body functions; cardio-respiratory fitness and neuromuscular capacity will be described in greater detail.

Cardio-respiratory fitness
Subjects will perform a submaximal exercise test on a cycle ergometer (Lode Corival, Groningen, The Nether lands). The test starts with unloaded pedalling for three minutes, followed by a 10 W att increment every single minute and will be terminated if one of the following occurs: (1) achieving 80% HRR or, (2) pedalling fre quency dropping below 60 rpm, or (3) not being able to continue the test for any reason. The same workload protocol will be applied at the follow-up measurements. Throughout the test, gas exchange variables (COSMED K4b2, Rome, Italy) and heart rate will be measured con tinuously. Cardio-respiratory fitness will be determined based on changes in heart rate and gas exchange variables.

Neuromuscular capacity
Maximal voluntary contraction (MVC) of the quadriceps muscle will be measured isokinetically between 90° and 30° knee flexion using a fixed dynamometer (Biodex System 3, New York, USA). To determine MVC, sub jects will perform three maximal-effort knee extensions at an angular velocity of 60°/sec. The highest value of peak torque (Nm) will be used for the analyses. In addition, under isometric conditions, voluntary acti vation (VA) will be measured with a modified superim posed stim ulation technique [34]. The quadriceps muscle will be electrically stimulated transcutaneously with surface electrodes placed over the proximal and distal part of the anterior thigh and using a computer controlled constant current stimulator (Digitimer DSH7, Welwyn garden City, UK). A triplet (pulse train of three 200 ^s pulses applied at 300 Hz with a supramaximal stim ulation current) will be superimposed on a 3-4s maximal isometric knee extension at 60° knee flexion. An identical stimulation will be delivered to the relaxed quadriceps muscle to evoke a resting (control) triplet. Voluntary activation will be calculated by the following formula: VA (%) = (1 -(superimposed triplet / control triplet)) x 100% Fatigability of the quadriceps muscle will be deter mined by fatiguing the muscle by a series of electrically stimulated contractions. The current will be reduced such that a single tetanus at 150 Hz evokes approxi mately 30-50% of maximal isometric strength. A series of trains of 50 Hz stimulation (duration 1000 ms, with 1000 ms between trains) will be applied for a period of 300 seconds (150 contractions in total). Recovery of fati gue will be monitored by applying the same 50 Hz train (1000 ms duration) at different times (15 sec up to 3 min) after the end of the protocol. Changes in force, maximal rate of force rise and half relaxation time will be determined. These parameters will be expressed as a percentage of the values obtained in the first contraction of the protocol to correct for differences in muscle strength.

Adverse events
All adverse events reported spontaneously by the parti cipants or observed by the therapists will be recorded. All adverse events will be followed until they have aba ted, or until a stable situation has been reached.

Responsive evaluation
To gain insight into patients' and therapists' expecta tions of and experiences with both interventions, a responsive evaluation will be conducted [35,36]. Thera pists and a subgroup of patients will be interviewed about their experiences with the intervention. The selec tion of respondents will be based on the criterion of ''m axim um variation" [37]. The interviews are semi structured, with open questions guided by a topic list. In order to check the topic list three pilot interviews will be conducted with patients and one with a partner. Then patients engaged in both interventions will be interviewed. Additionally focus groups will be organized to validate, deepen and broaden the issues in the inter views, starting with two homogeneous focus groups (converging interests) with patients and partners, and two homogenous focus groups with therapists (one per intervention). Then all parties will be brought together in a heterogeneous (diverging interests) dialogue group to exchange views. In the cyclic process data from ear lier phases will be used as input for the next phase; this hermeneutic dialectic interaction prevents one-sidedness and fosters the validation of data [38]. During all phases the interviews and focus groups will be tape-recorded, transcribed and analyzed. The analysis will focus on the recurring issues and concerns of all parties, and the comparison of perspectives. Member checks will be held to check the credibility of the analysis [37]. In order to consult and gain advice from patient representatives with PPS the research team will collaborate with the Vereniging Spierziekten Nederland (Dutch patient sup port group), in all phases of the research process [39].

Economic evaluation
Cost-effectiveness and cost-utility will be evaluated from a societal perspective. The costs will include direct HRQoL EuroQol-5D [41] X X X X

Resource use
Cost diaries X X X X * In the exercise group two extra blood samples will be obtained at 5 and 10 weeks. T1; pre-treatment, T2; post-treatment, T3; short-term follow-up, T4; long-term follow-up.
health care costs of visits to general practitioners, medi cal specialists, therapists, medications and assistive devices, direct non-health care costs i.e. non-reimbursable costs for complementary medicine and over-thecounter medication, and indirect, non-health care costs of changes in paid and unpaid work. All these data will be collected via cost diaries. Resource use will be valued following the procedures outlined by the Dutch M anual of Costing [40]. Where standard cost prices are not available, tariffs or professional fees will be used to estimate costs. The costs of medications will be estimated using the prices reported by the Royal Dutch Society for Pharmacy. Productivity loss costs from paid work will be estimated using both the human capital approach and friction cost method. The productivity loss costs from unpaid work will be deter mined using a shadow price. Cost prices of the interven tions will be determined by a bottom-up calculation.
The effect measures used in the economic evaluation will be our primary outcome measure fatigue and HRQoL m easured w ith the EuroQol-5D [41] at the four different tim e m easurem ents. U tilities will be derived using the D utch tariffs and used to determ ine qualityadjusted life years (QALYs) [42,43].

Statistical Analyses
GEE (generalized estim ated equations) analyses will be used to investigate differences in th e effects (prim ary and seco n d ary outcom es) b etw een b o th in te rv en tio n groups and the usual care group and to investigate asso ciations betw een potential effect modifiers and effect of in terv en tio n s. D ata will be analyzed according to the in te n tio n -to -tr e a t p rin c ip le . M issing d a ta w ill be im puted by carrying the last observation forward.
T h e eco n o m ic e v alu atio n w ill also be p e rfo rm e d according to the intention-to-treat principle. First, mean differences in QALYs will be tested using p aram etric te sts, and u n c e rta in ty ex p ressed by 95% confidence intervals. Second, we will conduct betw een-group com parisons of the m ean costs for each of the resource use categories and total costs. Confidence intervals around the m ean cost differences will be obtained by a bias-cor re c te d and accelerated (Bca) b o o tstra p p in g based on 5000 replications [44]. Third, the increm ental cost-effec tiveness ratio (ICER) will be determ ined by dividing the m ean difference in total costs by the m ean difference in effects. Insight in to the u n certain ty aro u n d the m ean ICER will be obtained by bootstrapping based on 5000 replications, the generation of cost-effectiveness planes [45] and cost-effectiveness acceptability curves [46]. Sen sitivity analyses involving the m ost im p o rtan t cost dri vers will be perform ed in order to assess the robustness of the results.

Discussion
In the FACTS-2-PPS study, the efficacy of exercise ther apy and CBT in reducing fatigue and im proving activ ities and HRQoL in patients w ith PPS will be evaluated in co m p a riso n w ith u su al care. T h is stu d y ca p tu re s some im portant strengths.
Firstly, the majority of studies on exercise in PPS eval uated the effects of strength or aerobic training in PPS separately. However, to execute norm al every day activ ities, m o re in te g ra te d fu n ctio n al tasks are necessary requiring a com bination of strength, endurance, flexibil ity, balance and co o rd in atio n . T his study evaluates a m u lti-c o m p o n e n t exercise train in g in w hich aerobic, strength and functional exercises are combined.
Secondly, a lth o u g h it is w idely re c o m m e n d e d th a t behavioural adaptation is an im portant com ponent in the m anagem ent of fatigue in PPS, there are no studies con firming the effectiveness of this approach. This is the first study th a t will evaluate the efficacy of CBT in reducing fatigue in PPS. The treatm ent protocol was developed by investigators with broad experience with CBT for chronic fatigue syndrome and fatigued cancer survivors, and cus tomized to the specific aspects of fatigue in PPS.
Thirdly, the long-term follow-up, the broad arsenal of outcom e m easures on different dom ains of the ICF and the in clu sio n of a responsive evaluation and an eco nomic evaluation gives this study a unique mixed m eth ods design. In v estig atin g p a tie n t p ersp e ctiv e s and possible discrepancies betw een perspectives of patients and professionals will help to elaborate strong and weak p o in ts of b o th in te rv e n tio n s, w hich in tu rn m ig h t explain specific results obtained in the study. A ddition ally, this mixed m ethods design will facilitate the deci sion m aking w ith re sp e c t to im p le m e n ta tio n of the results when the interventions turn out to be effective.
How ever, the study has also som e lim itations. The sam ple size was based on the detection of an approxi mately equal effect size of exercise training and CBT in comparison to usual care. Detecting differences in effec tiv en ess b etw een b o th in te rv e n tio n s w ould req u ire m uch larger sam ple sizes. F urtherm ore, the design of this study, will n o t give insight into the indicators th at are h elpful in decisio n m aking a b o u t ta ilo rin g interventions.
In conclusion, the results of this study will provide greater insight on evidence-based treatm ent options for PPS. Successful outcom e from the study has the poten tial to result in reduced fatigue and im proved activities and quality of life for patients w ith PPS. Results m ight lead to alterations of existing international guidelines.

Appendix 1: Overview of CBT intervention
Perpetuating factors: dysfunctional cognitions with respect to the disease, pain and fatigue A patient can have dysfunctional cognitions, for example unrealistic expectations on prognosis or difficulties with the acceptation of the consequences of PPS. M ore fre quent are dysfunctional cognitions w ith respect to fati gue a n d /o r pain. By non helping cognitions as 'I can't c o n tro l th e fatig u e' and 'th e fatigue is te rrib le ', the p atient catastrophizes about fatigue thereby enlarging the experience of fatigue. By means of cognitive restruc turing (Socratic dialogue and prom pting new self-talk) the patient learns to change these non helpful thoughts into more helpful one in daily life. Many patients are alert to symptoms of fatigue (or pain). P erso n s in th e ir social e n v iro n m e n t ask th e p a tie n t about the fatigue (or pain). Inform ation and explanation will be given th at the effect of attending to fatigue (or pain) se n sa tio n s is in creasing and intensifying these symptoms. By daily exercises the patient can learn how to shift attention away from fatigue (or pain). Also les sening the a tte n tio n from oth ers for these sym ptom s w ill be realized . T h e p a rtn e r of th e p a tie n t m ay be involved in this module.

Perpetuating factor: deregulation of sleep
A n irregular sleep-wake rhythm can perpetuate fatigue. To restore the biologic rhythm , patients will be encour aged to adhere to fixed b ed tim es and w ake-up tim es and will be discouraged from sleeping during the day or adapt fixed rest period(s).

Perpetuating factors: deregulation of physical, social and mental activities
Polio survivors are know n to have strong perseverance. F rom c h ild h o o d on they w ere le a rn e d to deny th e ir sym ptom s in o rd er to achieve a no rm al life. M any of them have difficulties adapting their physical, social and m ental activities to their decreasing abilities. First, these patients will be tho u g h t to pace or reduce the am ount of activities. N ext, graded increase of certain physical, social or m ental activities will help the patient to reach step by step their set goals.

Perpetuating factors: low social support and negative social interactions
Unrealistic expectations or non helpful cognitions about the social environment (for example 'why do I not get help from others?') can perpetuate fatigue. After explanation, the therapist will help to change non helpful cognitions and install more realistic expectations toward others in the p a tie n t's social en v iro n m en t. T he p a rtn e r or relevant others may be involved in this part of the therapy.