Patient Education and Counseling 54 (2004) 95–99

The Groningen Orthopedic Exit Strategy (GOES):a home-based support program for total hip and knee

arthroplasty patients after shortened hospital stayMartin Stevens∗, Inge van den Akker-Scheek, Alette Spriensma, Natalie A.D. Boss,

Ron L. Diercks, Jim R. van HornDepartment of Orthopedics, Groningen University Hospital, P.O. Box 30001, 9700 RB Groningen, The Netherlands

Received 31 January 2003; received in revised form 3 June 2003; accepted 15 June 2003

Abstract

The number of total hip and knee arthroplasty patients in The Netherlands will increase progressively in the coming decades. A reactionto this trend is a shortening of hospital stays. This however should not result in uncertainty, lack of information and unnecessary medicalconsumption by patients at home. Based on insights from social cognitive theory, an exit strategy is described to support patients duringtheir rehabilitation at home. It is our hypothesis that patients who participate in such a support program will be able to resume activitiesof daily living more quickly and effectively, and will reach higher levels of physical activity compared to patients who only follow theshortened program. To determine the effectiveness of the exit strategy, mediating variables (self-efficacy, social support and pain coping)and outcome variables (activities of daily living, physical activity behavior, health-related quality of life and gait analysis) will be measured.© 2003 Elsevier Ireland Ltd. All rights reserved.

Keywords: Total hip; Total knee; Rehabilitation; Orthopedic surgery; Early discharge; Cognitive theory; Home-based support

1. Introduction

Moderate to severe osteoarthritis is the most commonindication for total hip arthroplasty (THA) and total kneearthroplasty (TKA), and although differences in prevalenceand incidence exist between populations, osteoarthritis canbe considered to be a worldwide disease[1]. Most recentavailable data about the prevalence and incidence of os-teoarthritis in The Netherlands are from the mid 1990s. Theprevalence in 1994 of osteoarthritis of the hip was 11.7%(180,800), and osteoarthritis of the knee 19.1% (295,600).The incidence of osteoarthritis of the hip was 190 and os-theoarthritis of the knee 300 per 105 inhabitants. The inci-dences of THA and TKA in 1997 were 112 and 32 per 105

inhabitants, and increased with age. These Dutch numbersare in accordance with other Western countries[2–5].

In The Netherlands, as in other Western countries, thenumber of older adults in the population will increase bothrelatively and absolutely in the coming decades. As a result,an increase is being seen in the demand for THA and TKA.

∗ Corresponding author. Tel.:+31-50-3613271; fax:+31-50-3611737.E-mail address: m.stevens@orth.azg.nl (M. Stevens).

A reaction to this trend is a shortening of hospital stays[6,7].Groningen University Hospital in The Netherlands has de-veloped a shortened program to reduce the average lengthof hospital stay from 9.6 to 5.9 days[6]. A consequenceof shortened hospital stay is an increased responsibility ofpatients towards their rehabilitation. Pilot research into ourshortened program allows us to conclude that a consider-able number of patients lack knowledge about the rules theyhave to follow during rehabilitation, do not feel confident,and miss support from medical professionals once they areat home. This also results in unnecessary medical consump-tion (e.g. consultation of general practitioners or physicaltherapists), which could have been avoided had the supportbeen adequate[6].

Short hospital stays can be considered advantageous tohospitals in terms of shorter waiting lists and financial sav-ings. Shorter waiting lists also benefit patients, because theycan be operated sooner. Shortening of hospital stays how-ever should not result in a transfer of medical consump-tion from the hospital to the local general practitioners andphysiotherapists. Success of major joint arthroplasty is de-termined not just by an effective surgical procedure butalso by adequate post-surgical rehabilitation[8]. Supportprograms are needed to provide sufficient support in the

0738-3991/$ – see front matter © 2003 Elsevier Ireland Ltd. All rights reserved.doi:10.1016/S0738-3991(03)00191-5

96 M. Stevens et al. / Patient Education and Counseling 54 (2004) 95–99

post-hospitalization period. No support program for rehabil-itation in the home situation was available until now in TheNetherlands. This article describes the development of sucha program based on social cognitive theory. It is our hypoth-esis that patients who participate in this support programwill be able to resume activities of daily living more quicklyand effectively, and will reach higher levels of physical ac-tivity compared to patients who only follow the shortenedprogram. Eventually, this will lead to an improvement ofhealth-related quality of life.

2. Intervention: exit strategy

The exit strategy is a home-based program that aims atsupporting patients in the process of rehabilitation at homefor a period of 6 months after shortened hospital stay. Thestrategy consists of an exit video, newsletters, follow-up bytelephone and the involvement of a partner (spouse, relativeor close friend). Participants in the exit strategy are patientsadmitted to our Orthopedic Department for unilateral, pri-mary elective hip or knee arthroplasty and included in theshortened program. Inclusion criteria for the shortened pro-gram are the absence of severe co-morbidity and psycholog-ical dysfunction, and the availability of a partner at homewho is able to support the patient during the rehabilitationprocess.

The exit strategy is based on the social cognitive theorydeveloped by Bandura[9]. This theory assumes that behav-ior is affected by interactions between environmental influ-ences, personal factors and attributes of the behavior itself.A central aspect of the theory is the concept of self-efficacy.There is a general consensus that self-efficacy is one of themain mediating variables of health interventions[10]. The

Treatment OutcomeIntervening mechanism

Action domain Conceptual domain

Pain coping

Self-efficacy

Social supportVideo

Telephonecontact

Partnerparticipation

Physicalactivity

behavior

Qualityof

Life

Newsletters

ADLfunctioning

Fig. 1. The Groningen Orthopedic Exit Strategy.

exit strategy was structured with the help of Chen’s programevaluation theory [11]. Chen urges the development of a the-oretical framework that differentiates the details of the me-diating processes in the program and can serve as a basis toplan and evaluate activities. We have developed a theoreticalframework to be used as the exit strategy program theory.This theory explains the program’s theoretical premises anddescribes the process between input and outcome. Chen dis-tinguishes various theoretical domains within the programtheory. The exit strategy recognizes three domains: treat-ment, intervening mechanism and outcome. Next to thesedomains, Chen distinguishes an action domain and a con-ceptual domain (Fig. 1). The action domain describes howthe mediating variables of the intervening mechanism do-main are manipulated by program activities in the treatmentdomain. The conceptual domain describes how the mediat-ing variables of the intervening mechanism domain affectthe outcome variables as specified in the outcome domain.

2.1. Treatment domain

The treatment domain is made up of the basic, essentialelements that produce the intended changes in a program[11]. The treatment of the exit strategy lasts 6 months andhas four components: an exit video, newsletters, telephonecontact and partner involvement (Fig. 1).

An exit video is shown to the patients the day beforedischarge. Several former patients who participated in theshortened program were followed for 6 months after surgeryto compose the videos. There are separate 20 min videos forhip and knee patients, with the former patients telling abouttheir experiences with rehabilitation at home. This providesa realistic impression about what to expect after leaving thehospital.

M. Stevens et al. / Patient Education and Counseling 54 (2004) 95–99 97

The patients receive four newsletters with informationabout the rehabilitation process over a period of 6 months.The first newsletter arrives 2 days after the operation. Subse-quent newsletters are sent 2 weeks, 2 months and 5 monthsafter discharge. Each newsletter provides relevant informa-tion about the rehabilitation process at that stage.

The follow-up by telephone is two-fold: the department’snurse practitioner in charge of the specific patient makes twotelephone calls, 2 days and 3 weeks after discharge. Infor-mation is gathered on how the patient is doing, and patientquestions are answered. In addition, patients can contact thehospital during a weekly 2 h telephone consultation.

Patients are included in the shortened program only ifthey have a partner (spouse, relative or close friend) whocan support them during the rehabilitation period at home.The partner is invited to the preoperative patient intake andpost-operative instruction sessions (e.g. physical therapy),as well as the presentation of the exit video. The partner isalso expected to join the patient during control moments.

2.2. Outcome domain

A program is created for the purpose of providing a ser-vice or solving a problem. These purposes are formallycalled goals or intended outcome [11]. The intended out-come of the exit strategy can be considered two-fold: theability of these patients to resume activities of daily living,and their post-operative physical activity level. It is our hy-pothesis that both will affect the health-related quality oflife (Fig. 1).

2.3. Intervening mechanism

The intervening mechanism domain describes the under-lying mediating processes linking treatment to outcome [11](Fig. 1). The purpose of the strategy is to alter mediatingvariables of the rehabilitation process. To this end, threerelevant mediating variables have been selected for our in-tervening mechanism on the basis of previous research andinsights on social cognitive theory: self-efficacy, social sup-port and pain coping [12–18].

Self-efficacy represents the sum of a person’s processingof many kinds of behavior-related information by means ofwhat is known as observational learning. Self-efficacy en-compasses those beliefs regarding individuals’ capabilitiesfor performance or behavior that will lead to anticipated out-comes [9]. In order to perform a certain behavior, a personmust believe in her capacity to perform the behavior, andmust perceive an incentive to do so. Additionally, the personmust value the outcome expectations or consequences thatshe believes will occur as a result of performing a specificbehavior. Outcome expectations may be classified as havingimmediate or long-term benefits. As these expected out-comes are filtered through a person’s expectations or percep-tions of being able to perform the behavior in the first place,self-efficacy is believed to be the single most important char-

acteristic that determines a person’s behavior. Self-efficacyratings can be changed and thus represent a useful targetfor intervention. The four main sources of efficacy infor-mation are: (1) past and present levels of performance, (2)vicarious experience of observing others perform, (3) verbalpersuasion, and (4) states of physiological arousal [9].

Social support is seen as a multidimensional construct.The distinction is made between four dimensions: socialnetwork, social interaction, subjective support and instru-mental support [19]. The literature distinguishes two dif-ferent theories about the impact of social support on therelationship between stress and health. The stress buffertheory proposes that social support acts as a buffer to pro-tect people from the negative influence of stress on health[14,20]. The impact of disease can be encountered as po-tential stressor. The direct effect hypothesis states that anoptimal amount of social support has beneficial effects onhealth, regardless of whether stress is present [14]. Bothhypotheses are not mutually exclusive. Although social sup-port may be directly helpful in all circumstances, it may beparticularly effective as a buffer during times of stress [21].

Pain coping is the cognitive and behavioral effort indi-viduals use to cope with, deal with, or minimize their pain[22,23]. How a person copes with pain influences how wellshe adjusts to the pain [24]. Perceptions of helplessness pre-dict depression and psychosocial and physical impairment[13,18]. There are also indications that higher levels of painare related to lower levels of self-efficacy [22]. On the otherend of the continuum, perceptions of control over pain pre-dict lower levels of pain and disability, fewer pain behaviors,greater endurance during physical challenges and greaterpsychological well-being [25,26].

2.4. Action domain

In the action domain, we describe how the mediating vari-ables in the intervening mechanism domain are manipulatedby means of the components of the treatment domain [11](Fig. 1).

Moon and Backer [16] argue that in total joint replace-ment, patients’ self-efficacy beliefs need to be consideredwhen planning post-operative care. In the treatment domain,self-efficacy is influenced by several components. In the exitvideo, former patients can be seen as role models at differ-ent stages of their rehabilitation, offering patients a realisticview of the process that lies ahead [27]. Bandura [9] arguesthat realistic role models can have a positive influence onself-efficacy. Newsletters are sent to the patients on a regularbasis to provide them with relevant information about therehabilitation process. Past research indicates that the levelof knowledge and skill pertinent to making the transitionfrom the hospital to the home situation may be insufficient[17]. By giving information about the rehabilitation process,insecurity and lack of knowledge are taken away. It is hy-pothesized that this has a positive influence on self-efficacy.Patients will be contacted by telephone twice after discharge

98 M. Stevens et al. / Patient Education and Counseling 54 (2004) 95–99

from the hospital. This allows us to remain informed abouttheir status. Patients can use these calls to ask additional in-formation about the rehabilitation process if needed. In thatsense, these telephone calls are also hypothesized to have apositive effect on self-efficacy. Research by Weinberger et al.[28] shows that an intervention delivered by telephone had apositive effect on functional status, pain and physical health.

Social support will be manipulated in two ways. First, therole of the spouse or other close family member or friendwho will serve as the partner during the rehabilitation pro-cess is emphasized. Close relatives are the most importantpotential sources of social support [14]. Before, during andafter the hospital stay, the partners are invited to join thepatients in information and instruction sessions. Showal-ter et al. [17] argues that the inclusion of spouses or otherfamily members in the education process is essential. Sec-ond, patients will be contacted by telephone twice after dis-charge. Besides the information-exchange purpose of thesecalls, they are also a type of social support offered by hos-pital professionals. Weinberger et al. [29] have shown thata few telephone calls to elderly patients with osteoarthritisincreased their perceptions of social support, and that thecalls were associated with an improvement in these patients’self-reported level of functioning.

In order to influence pain coping, the newsletter willinclude information about pain and active pain-copingstrategies (e.g. diverting attention, lowering demands, rein-terpreting pain sensations). We also keep in contact with thepatients through the post-discharge telephone calls, and cansupport in the pain-coping process if necessary. During theweekly telephone consultations, patients can also take theinitiative to contact the hospital if they do not succeed in ma-naging their pain or have any additional questions about it.

2.5. Conceptual domain

The conceptual domain describes how the mediating vari-ables of the intervening mechanism domain affect outcomevariables [11] (Fig. 1). It is assumed that the mediating vari-ables can be manipulated through treatment. We hypothe-size that, by manipulating these variables, patients will berehabilitated more quickly and efficiently than through ourtraditional approach.

Empirical evidence indicates that perceived self-efficacypositively affects functional ability in individuals recoveringfrom cardiac surgery and nonelective hip fracture surgeryor arthritis [15,30]. On total joint replacement, Moon andBacker [16] indicated that patient self-efficacy was signifi-cantly and positively related to physical functioning as wellas social and leisure functioning. In their study, self-efficacysignificantly predicted higher levels of performance of ac-tivities of daily living.

Social support from spouses and family is consideredan important mediating variable in the rehabilitation pro-cess [14,17,31]. Kriegsman et al. [14] argue that perceivedfamily support tends to have a beneficial influence on the

consequences of chronic diseases in the aspects of psycho-logical adjustment, well-being, disease control and survivalin elderly people. Systems of partner participation and en-couragement, as well as positive feedback from physicians,nurses and fellow patients have all been suggested to play arole in a successful rehabilitation process [12,14,17,32].

With regard to pain coping, treatment programs that incor-porate interventions designed to decrease maladaptive andincrease adaptive appraisals and coping responses have re-sulted in decreased pain, improved psychological well-beingand enhanced physical functioning [13,18]. Furthermore,there are indications that pain-coping interventions designedto increase the use of adaptive pain-coping strategies couldenhance self-efficacy, reduce pain and improve physical andpsychological functioning in individuals suffering form os-theoarthritis [23]. Based on these studies, it can be arguedthat there is no reason to doubt that this could also be thecase for THA and TKA patients.

3. Practice implications

Our current research efforts are focused on validating theprogram theory of the exit strategy. In order to validate theprogram theory, the distinction is made between the valida-tion of the action domain and the conceptual domain, as aprogram’s overall accomplishment requires success in both[11]. By validating the action domain, we want to obtain in-formation on whether the treatment (the exit strategy) effec-tively manipulates the mediating variables described in theintervening mechanism domain. It is our hypothesis that bymanipulating self-efficacy, social support and pain copingthrough the support program, patients will be rehabilitatedmore quickly and efficiently compared to patients who onlyfollow the shortened program. This must become noticeablein three ways: in the ability of these patients to resume activ-ities of daily living independently, in their physical activitylevel, and as a result of both these effects in an increasedrating of health-related quality of life.

To execute this validation, an experimental study will startin 2003: the results of patients who participate in the exitstrategy (experimental group) will be compared with the re-sults of patients who receive the traditional treatment (con-trol group). Measurements of the mediating and outcomevariables will take place pre-operatively, and 6 weeks and6 months post-operatively. The study protocol is in accor-dance with the guidelines of the Medical Ethical Board ofGroningen University Hospital.

Existing instruments for measuring mediating variableshave been selected, and wherever necessary translated fromEnglish into Dutch. Research has been done into validity andreliability. In order to measure self-efficacy, we use a Dutchversion of the Behavior-Specific Self-Efficacy scale [33,34].The scale for measuring social support is an adjusted andtranslated version of the Duke Social Support Index [35].Pain coping will be measured with a Dutch version of the

M. Stevens et al. / Patient Education and Counseling 54 (2004) 95–99 99

TAMPA [36]. To measure the outcome variables, activitiesof daily living will be measured with the SF-36 and by meansof gait analysis with the DynaPort® system [37,38]. Physicalactivity behavior will be measured with the Physical ActivityQuestionnaire for the Elderly, and health-related quality oflife with the SF-36 [38,39]. Through the present study, weaim at enlarging our insight into the rehabilitation processfollowing total hip and knee arthroplasty.

References

[1] Saase van JLCM, Romunde van LKJ, Cats A, Vandenbroucke JP,Valkenburg HA. Epidemiology of osteoarthritis: Zoetermeer survey.Comparison of radiological osteoarthritis in a Dutch population withthat in 10 other populations. Ann Rheum Dis 1989;48:271–80.

[2] Observations on the future of public health, part 1, update on thehealth situation. Bilthoven: RIVM, 1997.

[3] Ostendorf M, Johnell O, Malchau H, Dhert WJ, Schrijvers AJP, Ver-bout AJ. The epidemiology of total hip replacement in The Nether-lands and Sweden: present status and future needs. Acta OrthopScand 2002;73:282–6.

[4] Prismant, van en voor de zorg (NL). Yeardisk for hospitals. Utrecht,2002.

[5] Robertsson O, Dunbar MJ, Knutson K, Lidgren L. Past incidence andfuture demand for knee arthroplasty in Sweden: a report from theSwedish Knee Arthroplasty Register regarding the effect of past andfuture population changes on the number of arthroplasties performed.Acta Orthop Scand 2000;71:376–80.

[6] Spriensma A. The total hip patient. Groningen: Hanze University,2001.

[7] Van de Aast M, Van Haastregt J. Operation hip: a pilot researchinto possibilities and impossibilities of shortening of hospital stay inpeople who underwent a hip operation. Leiden: TNO, 1994.

[8] Peerbhoy D, Keane P, Maciver K, Shenkin A, Hall GM, Salmon P.The systematic assessment of short-term functional recovery aftermajor joint arthroplasty. J Qual Clin Pract 1999;19:165–71.

[9] Bandura A. Self-efficacy: the exercise of control. New York: Freeman,1997.

[10] DeVellis BM, DeVellis RF. Self-efficacy and health. In: Baum A,Revenson TA, Singer JE, editors. Handbook of health psychology.Mahwah, London: Erlbaum (Lawrence), 2001. p. 235–47.

[11] Chen HT. Theory-driven evaluations. Newbury Park: Sage Publica-tions, 1990.

[12] Broos P, Fourneau I. Host factors that affect outcome of total hiparthroplasty. Lancet 2000;355:1479–80.

[13] Keefe FJ, Williams DA. A comparison of coping strategies in chronicpain patients in different age groups. J Gerontol 1990;45:161–5.

[14] Kriegsman DM, Van Eijk JT, Penninx BW, Deeg DJ, Boeke AJ. Doesfamily support buffer the impact of specific chronic diseases on mobi-lity in community-dwelling elderly? Disabil Rehabil 1997;19:71–83.

[15] Kurlowicz LH. Perceived self-efficacy, functional ability, and de-pressive symptoms in older elective surgery patients. Nurs Res1998;47:219–26.

[16] Moon LB, Backer J. Relationships among self-efficacy, outcomeexpectancy, and postoperative behaviors in total joint replacementpatients. Orthop Nurs 2000;19:77–85.

[17] Showalter A, Burger S, Salyer J. Patients’ and their spouses’needs after total joint arthroplasty: a pilot study. Orthop. Nurs.2000;19:49–57, 62.

[18] Turner JA, Clancy S. Strategies for coping with chronic low backpain: relationship to pain and disability. Pain 1986;24:355–64.

[19] Koenig HG, Westlund RE, George LK, Hughes DC, Blazer DG,Hybels C. Abbreviating the Duke Social Support Index for use inchronically ill elderly individuals. Psychosomatics 1993;34:61–9.

[20] Sarason IG, Levine HM, Basham RB, Sarason BR. Assessing so-cial support: the Social Support Questionnaire. J Pers Soc Psychol1983;44:127–39.

[21] Penninx BW, Van Tilburg T, Kriegsman DM, Boeke AJ, Deeg DJ,Van Eijk JT. Social network, social support, and loneliness in olderpersons with different chronic diseases. J Aging Health 1999;11:151–68.

[22] Rosenstiel AK, Keefe FJ. The use of coping strategies in chronic lowback pain patients: relationship to patient characteristics and currentadjustment. Pain 1983;17:33–44.

[23] Keefe FJ, Kashikar-Zuck S, Robinson E, Salley A, Beaupre P, Cald-well D, et al. Pain coping strategies that predict patients’ and spouses’ratings of patients’ self-efficacy. Pain 1997;73:191–9.

[24] Tan G, Jensen MP, Robinson-Whelen S, Thornby JI, Monga TN.Coping with chronic pain: a comparison of two measures. Pain2001;90:127–33.

[25] Buescher KL, Johnston JA, Parker JC, Smarr KL, Buckelew SP,Anderson SK, et al. Relationship of self-efficacy to pain behavior. JRheumatol 1991;18:968–72.

[26] Rejeski WJ, Craven T, Ettinger Jr WH, McFarlane M, Shumaker S.Self-efficacy and pain in disability with osteoarthritis of the knee. JGerontol B Psychol Sci Soc Sci 1996;51:24–9.

[27] Krouse HJ. Video modelling to educate patients. J Adv Nurs2001;33:748–57.

[28] Weinberger M, Tierney WM, Booher P, Katz BP. The impact ofincreased contact on psychosocial outcomes in patients with os-teoarthritis: a randomized, controlled trial. J Rheumatol 1991;18:849–54.

[29] Weinberger M, Hiner S, Tierney W. Improving functional status inpatients with osteoarthritis: the effect of social support. ArthritisRheum 1986;29:S155.

[30] Shoor SM, Holman HR. Development of an instrument to explorepsychological mediators of outcome in chronic arthritis. Trans AssocAm Physicians 1984;97:325–31.

[31] Mutran EJ, Reitzes DC, Mossey J, Fernandez ME. Social support,depression, and recovery of walking ability following hip fracturesurgery. J Gerontol B Psychol Sci Soc Sci 1995;50:S354–61.

[32] Greenfield S, Apolone G, McNeil BJ, Cleary PD. The importanceof co-existent disease in the occurrence of postoperative complica-tions and one-year recovery in patients undergoing total hip replace-ment. Comorbidity and outcomes after hip replacement. Med Care1993;31:141–54.

[33] Waldrop D, Lightsey OR, Ethington CA, Woemmel CA, Coke AL.Self-efficacy, optimism, health competence and recovery from ortho-pedic surgery. J Couns Psychol 2001;48:233–8.

[34] Van den Akker-Scheek I, Stevens M, Spriensma A, Van Horn JR.Measuring self-efficacy in patients with a total hip or knee arthro-plasty, unpublished observations.

[35] Van den Akker-Scheek I, Stevens M, Spriensma A, Van Horn JR.Measuring social support in patients with a total hip or knee arthro-plasty: the Groningen Othopedic Social Support Scale (GO-SSS),submitted for publication.

[36] Vlaeyen JW, Kole-Snijders AM, Boeren RG, van Eek H. Fear ofmovement/(re)injury in chronic low back pain and its relation tobehavioral performance. Pain 1995;62:363–72.

[37] Dikkenberg van den N, Meijer OG, Slikke van der RMA, Lum-mel van RC, Dieën van JH, Pijls B, et al. Measuring functionalabilities of patients with knee problems: rationale and constructionof the Dynaport knee test. Knee Surg Sports Traumatol Arthrosc2002;10:204–12.

[38] Van der Zee KI, Sanderman R. Measurement of general healthstatus with the RAND-36: manual. Groningen: Northern Centre forHealthcare Research, 1996.

[39] Voorrips LE, Ravelli AC, Dongelmans PC, Deurenberg P, VanStaveren WA. A physical activity questionnaire for the elderly. MedSci Sports Exerc 1991;23:974–9.