Patient preoperative expectations of total hip replacement in European orthopedic centers
Patient expectations have been identified as a factor that may account for individual differences in recovery after total hip replacement (THR) surgery. However, patient expectations have not been studied within a valid theoretical framework. This study employed the World Health Organization's model of health, the International Classification of Functioning, Disability and Health (ICF), to classify the content of preoperative patient expectations of THR.
A European cohort of 1,108 patients preoperatively reported 2 types of expectations. Patients reported what they anticipated surgery would enable them to do that they needed to be able to do (“need” expectation), and what they would like to be able to do (“desire” expectation) in a year's time. Free-text responses were independently classified by 2 researchers to 1 or more of the ICF constructs of impairment, activity limitation, and participation restriction.
Interrater reliability was high (κ = ≥0.87). All patient expectations were classified to the ICF constructs. Less than 5% of patient expectations were identified as impairment, 58% of “need” expectations were identified as activity limitations, and 45% of “desire” expectations were identified as activity limitations and participation restrictions combined.
The ICF is a suitable theoretical framework to study patient expectations of THR. THR targets impairment; however, few patient expectations were classified to the ICF definition of impairment. The majority of patient expectations were classified as activity limitation or a combination of activity limitation and participation restriction. Therefore, patient expectations of surgery focus on recovering valued activities rather than reversal of bodily impairments.
Total hip replacement (THR) surgery is an effective and frequently performed procedure for people with severe osteoarthritis of the hip, producing measurable reductions in pain and disability and improvements in quality of life (1–3). Despite the general effectiveness of THR, variability in the success of surgery exists, with some patients not experiencing improvements in pain and function following surgery (4, 5). The variability in the success of THR raises the question of what factors may account for individual differences in outcome and recovery.
Patient expectations have been identified as a factor that may account for individual differences in health outcomes and satisfaction after joint replacement surgery; however, the evidence is mixed. Some studies have found that patient expectations predict health outcomes (6–8), whereas others have not (9, 10). One reason for these mixed findings may lie in the fact that no standardized method of measuring patient expectations is available (11). Consequently, studies have measured patient expectations with differing methodologies that range from asking patients to choose a response from a set of given categories (12) to free-text responses to open-ended questions (13). Moreover, in general, studies have not used an established theoretical framework to define or classify patient expectations, further limiting the development of a cumulative evidence base within the patient expectation literature.
The International Classification of Functioning, Disability and Health (ICF) model of health outcomes (14) is an internationally accepted conceptual framework to understand functioning and disability associated with health and illness. The ICF identifies 3 health outcomes, specifically, impairments to body functions and structure, activity limitations, and participation restrictions. The ICF has already been applied to orthopedics in general and osteoarthritis in particular. For example, the World Health Organization (WHO) has employed the ICF to develop a core measurement set for osteoarthritis (15) and health outcome instruments frequently used in orthopedics, such as the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) and the Oxford Knee Score, that have been analyzed to identify items that measure each of the 3 ICF identified health outcomes (16). Further, the ICF has been used to predict mobility limitations in joint replacement patients (17) and to classify patient concerns following total knee replacement (18). Therefore, the ICF would appear to be a suitable theoretical framework to apply to the study of patient expectations of THR.
The main objective of the current study was to establish whether preoperative expectations of patients with hip osteoarthritis awaiting THR can be reliably classified to the constructs of the ICF in order to understand the content of patient expectations. To achieve this objective, patient expectations were matched to 1 or more or none of the 3 health outcomes identified at the component level of the ICF, specifically, impairment, activity limitations, and participation restrictions, as defined by the WHO.
Significance & Innovations
Patient expectations of total hip replacement (THR) are consistent with the theoretical definitions of impairment, activity limitation, and participation restriction from the International Classification of Functioning, Disability and Health (ICF) framework.
Patient expectations of THR focus on recovering valued activities rather than the reversal of bodily impairments, which is the primary target of THR.
Using the internationally recognized ICF framework to capture the content of patient expectations of THR will facilitate the development of a cumulative evidence base in the osteoarthritis and joint replacement literature.
Future work examining the predictive utility of the different types of patient expectations will help to inform clinical practice.
PATIENTS AND METHODS
A cross-sectional study of patient expectations of THR assessed preoperatively was performed. This study is part of a large prospective European cohort study (EUROHIP) of patients undergoing primary THR for osteoarthritis measuring preoperative expectations of surgery and disease, health, and functional status before and 1 year after THR surgery (19).
The EUROHIP cohort comprises a total of 1,327 patients from 20 orthopedic centers across 12 different European countries. Inclusion criteria for patients include a diagnosis of osteoarthritis of the hip, primary THR (i.e., no revision operations), and a willingness and ability to take part in the study. Further detail on the composition of the cohort has been described previously (19).
Part of the preoperative assessment of patients was administered via self-report questionnaire. As previously described in detail (19), the questionnaire was piloted in a UK orthopedic center and then the final version was translated by each of the other participating centers. The questionnaire asked questions about age, sex, employment, education, the WOMAC (20), the EuroQol 5-domain (EQ-5D) index (21), and patient expectations. Patients' expectations of THR were elicited by 2 free-text response questions: one addressing “needs” and the other addressing “desires.” Patients responded to the following 2 open-ended questions: 1) What things do you think you might be able to do in a year's time that you NEED to be able to do, but CANNOT do now, if the operation is a total success? (need expectation) and 2) What things do you think you might LIKE to be able to do in a year's time that you CANNOT do now, if the operation is a total success? (desire expectation). Responses were translated from the patient's native language into English by bilingual representatives from each orthopedic center.
Patients' responses to the expectation questions were quantified and classified using the following protocol by 2 researchers (NH and SR) working independently. To identify distinct expectations, the following rules were adopted: an “and” conjunction between different concepts or actions and sentence punctuation indicated distinct expectations (e.g., the response “to walk without problems and pain” and the response “painfree, more independent” both counted as 2 expectations); however, when the “and” conjunction joined 2 aspects of the same action, then only 1 expectation was counted (e.g., “to put on shoes and socks”). When the response read “nothing,” the number of expectations was counted as 0, and when no response was given, the data were treated as missing and the participant was classified as a “nonresponder” and not included in further analyses.
Second, each of the first 3 expectations expressed by each patient in response to the “need” and “desire” questions was compared to the theoretical definition of each ICF construct (Table 1). Therefore, each expectation was classified as either impairment (I), activity limitation (A), participation restriction (P), or a combination thereof (i.e., IA, IP, AP, IAP).
Table 1. Definitions of the World Health Organization ICF constructs*
|Impairment||Problems in body function or structures such as a significant deviation or loss|
|Activity limitations||Difficulties an individual may have in executing activities|
|Participation restrictions||Problems an individual may experience in involvement in life situations|
All analyses were carried out using SPSS software, version 15. Cohen's kappa statistic assessed the level of agreement between the 2 judges.
Description of sample.
Of the 1,327 patients in the EUROHIP cohort, 1,108 patients (83%) responded to one or other of the expectation questions and thus comprised the sample for the present study. Demographic and clinical differences between responders and nonresponders were tested (Table 2). Responders did not significantly differ from nonresponders in terms of sex, age, employment status, educational qualifications, body mass index, American Society of Anesthesiology status measure of fitness for surgery, Kellgren/Lawrence grade, or WOMAC scores. However, nonresponders did report better health-related quality of life than responders as measured by the EQ-5D (P < 0.001). Data on the demographic and clinical characteristics of the full EUROHIP cohort have previously been reported (19).
Table 2. Demographic and clinical characteristics of patients reporting expectations (responders) compared to patients not reporting expectations (nonresponders)*
|Sex, no. (%)|| || ||0.13|
| Male||478 (43.2)||105 (48.8)|| |
| Female||629 (56.8)||110 (51.2)|| |
|Age, mean ± SD years||65.1 ± 11.0||65.7 ± 10.4||0.46|
|Employment status, no. (%)|| || ||0.55|
| Employed||245 (22.8)||42 (20.0)|| |
| Retired||633 (58.8)||122 (58.1)|| |
| Retired early||92 (8.5)||19 (9.1)|| |
| Other||107 (9.9)||27 (12.9)|| |
|Educational qualifications, no. (%)|| || ||0.18|
| None after leaving school||496 (50.5)||86 (49.1)|| |
| College diploma||312 (31.8)||67 (38.3)|| |
| University degree||127 (12.9)||18 (10.3)|| |
| Postgraduate degree||47 (4.8)||4 (2.3)|| |
|BMI, mean ± SD kg/m2||27.6 ± 4.4||27.1 ± 4.2||0.12|
|ASA status, no. (%)|| || ||0.065|
| 1 (normal, healthy)||176 (18.1)||33 (16.3)|| |
| 2 (mild systemic disease)||604 (62.1)||115 (56.7)|| |
| 3 (severe systemic disease)||184 (18.9)||53 (26.1)|| |
| 4 (life-threatening disease)||8 (0.8)||2 (1.0)|| |
|Kellgren/Lawrence grade, no. (%)|| || ||0.75|
| 0 (no features)||4 (0.5)||2 (1.2)|| |
| 1 (minute osteophyte)||5 (0.6)||1 (0.6)|| |
| 2 (definite osteophyte)||28 (3.2)||4 (2.4)|| |
| 3 (moderate diminution joint space)||426 (48.3)||84 (49.7)|| |
| 4 (no joint space)||419 (47.5)||78 (46.2)|| |
|Total WOMAC, mean ± SD||59.5 ± 16.0||57.3 ± 16.3||0.07|
| WOMAC pain||55.6 ± 17.8||55.0 ± 18.1||0.65|
| WOMAC stiffness||60.8 ± 20.8||62.5 ± 20.0||0.27|
| WOMAC function||60.5 ± 16.6||58.8 ± 17.2||0.17|
|EQ-5D, mean ± SD||0.40 ± 0.33||0.59 ± 0.34||< 0.001|
The level of agreement between the 2 researchers regarding the number and ICF classification of patient expectations was high. Cohen's kappa statistic values corresponded to “almost perfect” agreement (22): number of “need” expectations (κ = 0.996), classification of first “need” expectation (κ = 0.908), classification of second “need” expectation (κ = 0.942), classification of third “need” expectation (κ = 0.948), number of “desire” expectations (κ = 0.980), classification of first “desire” expectation (κ = 0.879), classification of second “desire” expectation (κ = 0.886), and classification of third “desire” expectation (κ = 0.873). Discrepancies were discussed until consensus was achieved. The total number of expectations given by each patient varied widely; the mean number of “need” expectations was 2.5 (range 0–12), with 9 patients responding “nothing,” and the mean number of “desire” expectations was 2.4 (range 0–15), with 1 patient responding “nothing.”
All expectations were coded to the constructs of the ICF (i.e., there were no expectations that could not be classified as impairment, activity limitation, or participation restriction, or a combination thereof). The most commonly reported expectations classified into each of the ICF constructs or combined constructs were the following: to be pain free and to be in less pain (impairment), to be able to walk further and perform activities of daily living such as bathing and dressing (activity limitations), to travel and work (participation restrictions), to dance and partake in exercise (activity limitation and participation restriction combined), to walk without pain and to walk without limping (impairment and activity limitations combined), to work without pain (impairment and participation restrictions combined), and to be independent and live a normal life (impairment, activity limitations, and participation restrictions combined).
Table 3 shows the proportion of expectations classified to each ICF construct or combination thereof for the first 3 “need” and the first 3 “desire” expectations. The majority of the patient “need” expectations were classified as activity limitation: 57.8%, 59.0%, and 58.4% of the first, second, and third responses given, respectively. Most of the patient “desire” expectations were classified as activity limitation and participation restriction combined: 40.8%, 47.7%, and 45.6% of the first, second, and third responses given, respectively. Less than 5% of all expectations were classified as impairment.
Table 3. Proportion of expectations in each of the International Classification of Functioning, Disability and Health construct classification groups for the first 3 reported patient “need” expectations and the first 3 reported “desire” expectations*
|Activity limitation (A)||57.8||59.0||58.4||39.5||32.9||36.8|
|Participation restriction (P)||5.2||5.5||5.0||9.2||10.6||9.1|
|Impairment and activity limitation (IA)||11.5||7.1||7.1||5.0||4.8||3.3|
|Impairment and participation restriction (IP)||0.8||0.8||0.5||0.3||0.0||0.0|
|Activity limitation and participation restriction (AP)||18.0||21.5||24.4||40.8||47.7||45.6|
|Impairment and activity limitation and participation restriction (IAP)||3.1||1.9||1.4||2.6||2.6||2.1|
|Total impairment (I, IA, IP, IAP)||19.1||14||12.2||9.7||8.8||8.4|
|Total activity limitation (A, IA, AP, IAP)||90.4||89.5||36.9||87.9||88||87.8|
|Total participation restriction (P, IP, AP, IAP)||27.1||29.7||31.3||52.9||60.9||56.8|
|Total number of patients||1,038||731||438||874||568||329|
The total proportion of impairment, activity limitation, and participation restriction expectations and the total number of patients included in the classification for each of the first, second, and third expectations are also shown in Table 3. The proportion of the first responses that were wholly or in part classified as participation restriction was higher for the “desire” expectation (52.9%) than for the “need” expectation (27.1%; Z = −11.53, P < 0.001).
This study has shown that the ICF can be used to classify patient expectations of THR. All patient expectations could be classified to either 1 or a combination of the ICF constructs, specifically, impairment, activity limitation, or participation restriction. This provides support for the use of the ICF as an existing theoretical framework of health and disability to understand the content of patient expectations.
To date, the development of a cumulative evidence base within the patient expectation literature has been limited by the use of widely different methodologies in the absence of a valid theoretical framework within which the evidence base can be integrated. In the current study, the ICF, which is an internationally accepted framework of functioning and disability, was used to classify patient expectations of THR. The ICF has already been demonstrated to be an appropriate model of health outcomes in osteoarthritis (15) and joint replacement surgery (23). Therefore, the application of the ICF to patient expectations enables those expectations to be defined by the common language provided by the ICF. The use of the ICF supports the integration of the evidence base across studies that employ disparate methods to address different aspects of functioning within orthopedics in general and osteoarthritis in particular.
The current study indicates that patients' needs are different from their desires; the majority of “need” expectations were classified as activity limitation, whereas the majority of “desire” expectations were classified as activity limitation and participation restriction combined. Further, there was a greater proportion of first responses classified as participation restriction for the “desire” expectation question than for the “need” expectation question. The ICF definitions state that activity limitation focuses on the execution of activities, whereas participation restriction emphasizes the individual's social participation and their involvement in life situations (14). The finding that “need” expectations were most often related to the execution of activities while “desire” expectations were most often related both to the execution of activities and to the individual's social participation suggests that what THR patients need to be able to do and what they would like to be able to do may be different. For example, they may need to do basic activities such as walking and bending down, but they may also like to be able to participate in social activities such as playing with grandchildren or sport. Likewise, this explanation is further supported by the finding that participation restriction expectations were reported more often in response to the “desire” expectations than in response to the “need” expectation.
In addition to the semantic difference between the expectation questions by the use of “need” or “would like” to elicit needs or desires, respectively, there is another difference in the wording of the questions that may help to understand the above finding. The “need” item asked patients to identify what they think they will be able to do and need to be able to do, whereas the “desire” item only asked them to identify what they would like to be able to do. Therefore, it could be argued that only the “need” item elicits an expectation and the “desire” item simply elicits a desire or wish.
Less than 5% of patient expectations were classified as impairment. This finding is somewhat surprising, given that pain is a primary symptom of osteoarthritis and within the ICF pain is defined as a measure of impairment (15, 16, 24). Further, when asked about expectations in relation to pain, more than 75% of patients reported the expectation of complete pain relief after joint replacement (7). There are several possible explanations for the discordance between the current finding and previous research. First, pain expectations may not be spontaneously reported by patients. Unlike previous studies, the current study did not specifically ask about expectations in relation to postoperative pain but rather provided patients with the opportunity to respond to the expectation questions in any way they chose. While pain relief may be expected by patients, such expectations may be more likely to be expressed in response to a specific question about postoperative pain status. Similarly, participants in a study of osteoarthritis in the community did not spontaneously refer to pain when asked to compare activities of daily living such as walking outside and getting into and out of a chair (25). Therefore, it would seem that even though pain is a primary symptom of osteoarthritis, pain-related cognitions are not necessarily spontaneously expressed by patients. It is perhaps the case that pain is less important to patients than imagined and that issues related to engagement in activities and social situations are of greater concern. For example, the Oregon Death with Dignity Act 2009 reported that terminally ill patients had more concerns about being able to engage in activities that make life enjoyable (86.4%) than about pain (10.2%) (26).
Pain expectations may also not be commonly reported because rather than focusing on pain itself, patients may focus on the impact of pain, i.e., how it inhibits their ability to engage in valued activities. Therefore, pain may be important for patients; however, their attention and efforts are directed toward managing the pain in order to continue living their life and maintaining their social relationships. Qualitative studies have found that osteoarthritis patients commonly modify their lifestyle to cope with pain (27) and that the functional, psychological, and social consequences of pain are important factors that lead to the decision to undergo joint replacement (28).
An alternative explanation for the lack of impairment expectations may be found in the wording of the expectation questions in the current study. The questions asked the patient to report what they cannot do but need or would like to be able to do; hence, the repeated use of the verb “do” may have shaped patients' thinking toward actions and tasks, and resulted in them reporting activity limitation and participation restriction expectations. This may explain the finding that the majority of expectations were classified as activity limitation and participation restriction. However, even if the wording of the questions had primed patients toward giving activity limitation and participation restriction expectations, if patients held strong impairment expectations, then we might have expected to see more expectations classified as activity limitation or participation restriction combined with impairment, for example, to be able to “walk (activity) without pain (impairment)” or to be able to “watch a film in the cinema with my grandchildren (participation) without being in pain (impairment),” but this was not the case.
Future work should provide patients with the opportunity to respond to open-ended expectation questions that do not employ the verb “do.” Such studies would allow further investigation of the content of spontaneously reported expectations free from the demand characteristics of question wording and predetermined tick-box response categories. It is also acknowledged that earlier questions in a questionnaire may prime responses for later questions. Patient expectations of the EUROHIP cohort were elicited in the preoperative assessment questionnaire. The expectation questions were preceded by the WOMAC, which assesses dimensions of pain, stiffness, and function, and the EQ-5D, which assesses the dimensions of mobility, self-care, usual activities, including social and leisure activities, pain and discomfort, and anxiety and depression. The holistic nature of the dimensions addressed by these 2 measures would suggest that patients would have been equally primed to report expectations relating to any of the 3 ICF constructs (impairment, activity limitation, or participation restriction) and thus it is likely that if a significant priming effect occurred it occurred equally across all 3 ICF constructs.
When patients did not respond to an expectation question, the data were treated as missing and these cases were not included in the analysis. Approximately 20% of patients were categorized as nonresponders in this way. It is possible that nonresponders did not have any expectations of THR and thus could have been aggregated with the approximately 1% of patients who responded that they expected “nothing.” However, the actual reason for patients not responding could not be deciphered, and therefore a conservative approach was adopted and the data were treated as missing.
The current study is based on data from the EUROHIP cohort, which is essentially a convenience sample of THR patients recruited from orthopedic centers with an interest in the prospective investigation of THR patients. Therefore, it is possible that the participating centers differ in some important way from nonparticipating centers. Similarly, it is possible that patients who agreed to participate differ on some demographic, clinical, or psychological variable from those who refused to take part, potentially further weakening the representativeness of the sample. We found that patients responding to the expectation questions had worse health-related quality of life as measured by the EQ-5D than patients who did not respond. This finding is contrary to the usual observation that nonresponders in clinical studies have poorer health than responders. However, in this study, nonresponders are different from those in other studies because they were patients that did not answer the expectation questions rather than patients that did not participate at all in the study. Data from the current study may suggest that patients with better than average health-related quality of life do not have fully developed expectations of their surgery and thus did not respond to the expectation questions. Perhaps these patients may perceive less capacity for the surgery to improve their health-related quality of life as their preoperative level is better than average. Future work is needed to identify whether current health-related quality of life affects patients' expectations of surgery. Finally, the questionnaires were developed in the English language and translated by each participating center; however, the questions were not back translated and thus may have had a different meaning in the translated language. Study replication incorporating back translation of questions would therefore be useful.
In summary, THR primarily targets impairment; however, the current study findings suggest that few patients expect such outcomes following surgery. In contrast, patients were found to expect and want improvements in disability-related outcomes. This discrepancy may have important clinical implications for patient satisfaction. Moreover, future work is needed to address the predictive utility of particular types of expectations (impairment, activity limitations, or participation restriction expectations). This work has the potential to inform clinical practice by identifying those patients most likely to have a poor outcome from surgery.
All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Hobbs had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design. Dreinhöfer, Günther, Dieppe.
Acquisition of data. Dreinhöfer, Günther, Dieppe.
Analysis and interpretation of data. Hobbs, Dixon, Rasmussen, Judge.
ROLE OF THE STUDY SPONSOR
Centrepulse Orthopaedic (Sulzer Medical) had no involvement in study design, data collection, data analysis, or in the production of the manuscript, and we are not required to seek their approval for the content of the manuscript. Their only role was in the facilitation of the study group.
The authors wish to thank the many surgical teams and individuals who contributed to the success of this project. We would also like to thank the EUROHIP advisor committee comprising Hermann Brenner, Maxime Dougados, Klaus Hug, and Heiner Raspe. The EUROHIP study group consists of the following members who have contributed to this project: Martin Krismer and Bernd Stoeckl, University of Orthopedic Surgery, Innsbruck, Austria; Karl Knahr and Oswald Pinggera, Orthopedic Spital Wien-Speising, Vienna, Austria; Pekka Ylinen, Orton Orthopaedic Hospitala, Helsinki, Finland; Moussa Hammadouche, Groupe Hospitalier Cochin, Paris, France; Christian Delaunay, Clinique De L'yette, Longjumeau, France; Philippe Chiron, Centre Hospitalier Ranguell, Toulouse, France; Wolfhart Puhl, Karsten Dreinhoefer, Markus Floeren, Sabrina Baumann, and Dagmar Groeber-Graetz, University of Ulm, Ulm, Germany; Klaus-Peter Guenther and Stefan Fickert, Carl-Gustav Carus University, Dresden, Germany; Joachim Löhr, Alexander Katzer, and Dietrich Klüber, ENDO-Clinic, Hamburg, Germany; Volker Ewerbeck, Peter Aldinger, and Dominik Parsch, University of Heidelberg, Heidelberg, Germany; Wolfram Neumann, Ingmar Meinecke, and Thomas Bittner, Otto von Guericke University, Magdeburg, Germany; Wilfried von Eiff and Conrad Middendorf, Center for Hospital Management, Munster, Germany; Hans-Peter Scharf, Peter Schraeder, and Sabine Schmitt, University Clinic Mannheim, Mannheim, Germany; David Rowley, Ninewells Hospital and Medical School, Dundee, UK; Ian Learmonth, Avon Orthopaedic Centre, Bristol, UK; Paul Dieppe, Victoria Cavendish, and Susan Williams, HSRC, University of Bristol, Bristol, UK; Peter Kellermann and Ildiko Fistzer, University of Szeged, Szeged, Hungary; Thorvaldur Ingvarsson, Akureyri University Hospital, Akureyri, Iceland; Paolo Gallinaro and Alessandro Masse, Universita degli Studi di Torino, Turin, Italy; Andrzej Gorecki and Maciek AmbroAmbroziak, Medical University of Warsaw, Warsaw, Poland; Eduardo Garcia-Cimbrelo, Hospital La Paz, Madrid, Spain; Anna Nilsdotter and Urban Benger, Helsingborg Hospital, Skane, Sweden; Christian Hellerfelt and Christer Olson, Lasarett Karlshamm, Sweden; Joerg Huber and Ivan Broger, Kantonalspital Aarau, Switzerland; Robert Theiler, Kurt Uehlinger, and Angela Hett, Stadtspital Triemli, Zurich, Switzerland; and Til Stuermer, Harvard Medical School, Boston, Massachusetts.