Patient-centered care means that physicians respond to patients’ preferences including their preferences regarding involvement in treatment decisions, insofar as these preferences are reasonable and consistent with knowledge regarding the condition1. Three styles of decision-making in clinical care have been described: paternalistic, shared, and consumerist (or informed)2. According to the paternalistic model, physicians make decisions on the basis of what they believe to be in the patient's best interest, and minimal information is conveyed from the physician to the patient. According to the consumerist model, doctors provide the information that patients require to make their own decisions. Finally, according to the shared decision-making model, the physician and patient make the decision together and exchange medical and other information related to the patient's health2. A previous study has suggested that shared decision-making leads to greater patient satisfaction in medical treatment, if the physicians have an appropriate level of subject knowledge and can communicate this effectively and in an unbiased manner3.
Many studies have been undertaken on patients’ preferences regarding participation in decision-making, mostly in patients with cancer4-9, in hypothetical scenarios10-12, and in population-based surveys13,14. In orthopaedic and related fields, Deber et al. compared several patient groups, including patients with orthopaedic problems15, and Brady described the role of the patient in rheumatology care16. Furthermore, several studies have identified factors that affect perceptions regarding a willingness to undergo an orthopaedic procedure and patient participation in decision-making17-19; in addition, several authors have emphasized the importance of incorporating patients’ preferences into orthopaedic care20-24. Patients’ preoperative preferences and retrospectively perceived actual roles in decision-making in carpal tunnel syndrome have not been previously studied, to our knowledge. Since the diagnosis of carpal tunnel syndrome is generally based on clinical findings and the natural history remains unclear, questions have been raised regarding when and which treatment is most appropriate25.
We surveyed patients’ preferences preoperatively and assessed their perceived experiences postoperatively to determine their level of involvement in decision-making for carpal tunnel release and to attempt to identify the factors that affected these preferences and experiences. We hypothesized that patients with carpal tunnel syndrome would have preferences for decision-making styles associated with demographic and clinical variables such as comorbidities, a history of a surgical procedure, having a caregiver, the importance of other family members’ opinions, knowledge about carpal tunnel syndrome, education level, private insurance, whether they believed that medical costs were burdensome, and severity of symptoms. We also hypothesized that their retrospectively perceived experiences regarding decision-making style would be associated with these variables.
Subjects
From May 2008 to December 2009, we recruited seventy-eight consecutive patients with carpal tunnel syndrome who underwent surgery by a single hand surgeon at our urban tertiary referral hospital. To detect a Spearman correlation (r) of >0.03 between preferred and retrospectively perceived levels of involvement in decision-making, with a power of 80% (meaning that the chance that a true correlation would be overlooked is <20%) at a level of significance of p < 0.05, a total of seventy-four patients was required. All patients were referred by a primary care physician, a general orthopaedic surgeon, a neurologist, or a rehabilitation physician. Electrodiagnostic studies confirmed the diagnosis in all patients. There were seventy-one women and seven men, and all were ethnically Korean. The median age of the patients was fifty-seven years (range, twenty-seven to eighty-one years). The average duration of the symptoms before surgery was thirty-two months (range, six to sixty months), and the average waiting period from consultation to surgery was four weeks. We excluded those with a Workers’ Compensation issue and those with any condition that required additional surgery, such as cervical radiculopathy and cubital tunnel syndrome. However, we included those with systemic comorbidities, such as diabetes mellitus, hypothyroidism, chronic renal failure, rheumatoid arthritis, and other forms of arthritis that did not require additional surgery. An approval from our institutional review board was obtained for this study.
Consultation for Surgery
We explained the disease status, electrodiagnostic study findings, and the usual disease course in a standard fashion. We recommended carpal tunnel release when clinical symptoms of tingling, pain, or weakness did not improve after at least two months of treatment with a splint, medications, and/or corticosteroid injections. As all patients were referred by other physicians and had conservative treatment before consultation, we did not continue conservative treatment for more than two months. The operating surgeon briefly explained (1) the surgical procedure of open carpal tunnel release, which would be performed under local anesthesia on an outpatient basis; (2) the postoperative treatment with immobilization in a splint for three days and no formal physiotherapy; and (3) the likely functional outcomes, including the degree of symptom relief and possible complications. Patients with bilateral involvement were offered a choice of staged or simultaneous surgery. Patients were also provided brochures on carpal tunnel syndrome. All patients provided written informed consent before surgery.
Survey Design
Degner et al. developed the Control Preferences Scale to assess the role that patients want to play during treatment decision-making26. The Control Preferences Scale consists of five items that portray different roles in decision-making. These range from a fully active to a fully passive role (see Appendix). The five items rather than three were chosen because studies showed that decisional preferences have a rank order or a unidimensional scale27-29. The Control Preferences Scale has been tested in various populations and has been proven to be a clinically relevant, reliable, valid instrument for the measurement of decisional control preferences27-29. Degner et al.27, in a cross-sectional survey of women with breast cancer, used the Control Preferences Scale to assess the preferred and perceived roles of the patients in the decision-making for the initial treatment. Kremer et al.29 also reported a cross-sectional survey of patients with human immunodeficiency virus-acquired immune deficiency syndrome (HIV-AIDS) regarding the patients’ recent decision-making for antiretroviral treatment.
Before consultation for surgery, patients who agreed to participate in the study were requested to indicate their preferred decision-making role related to treatment of the carpal tunnel syndrome. The physician was blind to the patients’ preferred decision-making role. Although most patients made a decision about surgery on the day of the consultation, a few elected to take time to think about or discuss it with their family members. Two days postoperatively at the first clinic visit after surgery, they were asked, with use of the Control Preferences Scale, to identify the roles that they actually had during decision-making about surgery. Patients were not reminded of their prior answers. Although in most previous studies patients had been asked about the preferred and perceived roles at the same time, we surveyed the preferred role preoperatively and the perceived role postoperatively. We considered that minimizing the time between surgery and the retrospective survey would better reflect the patients’ memory of their decision, and could reduce the bias from different outcomes.
Patients’ demographic characteristics were also obtained during postoperative visits. The confounding variables examined were comorbidities, such as diabetes mellitus, thyroid disease, hypertension, and rheumatoid arthritis; a history of a surgical procedure; the existence of a caregiver; the importance of other family members’ opinions; knowledge about carpal tunnel syndrome; education level; private insurance; and whether they thought that medical costs were burdensome (Table I). These variables were chosen from previous studies that investigated factors associated with patients’ preferred style of medical decision-making11,15,16,19. The existence of comorbidities, a caregiver, and private insurance was directly measured as yes or no. (In South Korea, all citizens are required to enroll in the Korean National Health Insurance Program. Patients pay, on the average, 30% of all medical costs, and the government pays 70% of the cost. Approximately 97% of the population is covered by this system, and the remaining 3% includes those who are either covered by a medical aid program or are temporary or illegal residents. Therefore, all patients in the current study were covered by the National Health Insurance Program. Private insurance refers to extra insurance programs that reimburse the 30% of the medical costs that patients pay to hospitals.) The importance of other family members’ opinions, knowledge about carpal tunnel syndrome, and whether the patient thought that medical costs were burdensome were rated by the patient on a 5-point Likert-type scale; responses of “strongly agree” and “agree” were interpreted as indicating yes, and “neutral,” “disagree,” and “strongly disagree” were interpreted as indicating no.
Study subjects were also requested preoperatively to complete the Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire30,31 as a part of routine preoperative assessment. The validated DASH questionnaire quantifies general disabilities related to the upper extremity. The questionnaire contains thirty items; twenty-one concern difficulties with specific tasks, five evaluate symptoms, and the other four address social function, work, sleep, and confidence. DASH scores range from 0 to 100, with higher scores indicating greater disability of the upper extremity. DASH scores were used to assess subjective disability and its association with roles in decision-making.
Data Analysis
Basic descriptive statistics were calculated for patients’ preferred and perceived roles and demographic characteristics. Congruency between preferred and retrospectively perceived (experienced) roles was assessed with use of Spearman correlation and kappa coefficients.
The patients’ preferred and perceived roles were each considered as dependent variables, and the confounding variables were analyzed as independent variables in this study. The Spearman correlation test was used to determine the association between confounding variables and the patients’ preferred and retrospectively perceived roles in decision-making. Variables found to have a significant association by univariate analysis were subjected to multivariate analysis to determine the independence of the associations. The Kruskal-Wallis test and post hoc tests were used to compare DASH scores. Significance was accepted for p values of <0.05.
Sources of Funding
This study was supported in part by the institution's research fund (11-2009-004).
Preferred and Retrospectively Perceived (Experienced) Roles in Decision-Making
Of the seventy-eight patients, five chose a fully active role as their preferred role in decision-making before consultation for surgery; eleven, semiactive; twenty, collaborative; twenty-eight, semipassive; and fourteen, fully passive. When asked postoperatively what role they actually played in decision-making, four chose a fully active role; twenty-two, semiactive; twenty-five, collaborative; nineteen, semipassive; and eight, fully passive (see Appendix).
When the semiactive and semipassive roles were considered as shared decision-making, fifty-nine patients (76%) preferred to share decision-making with the physician, five (6%) preferred consumerism, and fourteen (18%) wanted the physician to decide before consultation. Postoperatively, sixty-six patients (85%) stated that they experienced shared decision-making, four (5%) responded that they made the decision on their own, and eight (10%) stated that the doctor made decisions on their behalf.
Congruence Between Preferred and Experienced Decision-Making Roles
The correlation between preferred and experienced roles was significant (r = 0.525, p < 0.001), and a moderate degree of agreement was found (kappa = 0.350, p < 0.001). In total, 50% (thirty-nine) of the patients experienced their preferred style of medical decision-making. Demographic factors assessed, such as comorbidity, previous operation, existence of a caregiver, placing importance on family members’ opinions, having knowledge about the condition, level of education, benefit from private insurance, and fear of the burden imposed by the medical cost, were not found to be associated with patients’ experiencing their preferred style of decision-making.
Factors Associated with Preferred and Experienced Roles in Decision-Making
In general, patients with a history of a surgical procedure and those who placed importance on the opinions of family members preferred a more active role, according to univariate analysis. However, multivariate analysis showed that only a history of a surgical procedure was independently associated with a preference for a more active role in medical decision-making, after controlling for age and sex (Table II). Patients with a caregiver and private insurance were more likely to play a more active role in medical decision-making, according to univariate and multivariate analyses (Table III). Factors such as comorbidities, having knowledge of the condition, education level, and a fear of the burden imposed by medical costs were not found to be associated with preferred or experienced roles.
When DASH scores were compared with use of the Kruskal-Wallis test, they were found to be significantly different for patients with different preferences (p = 0.034) (Table IV). Post hoc analysis with use of the Mann-Whitney test indicated that patients who preferred a collaborative role had lower DASH scores than those who preferred a fully active role (p = 0.002) or those who preferred a fully passive role (p = 0.009). These results indicate that patients who prefer shared decision-making are likely to have less severe symptoms. Similarly, when we compared patients with different experiences of decision-making, those who had a collaborative role had a lower mean DASH score than those who experienced a fully active or a fully passive role. However, the difference was not significant (p = 0.284) (Table IV).
The traditional clinician-centered and disease-focused approaches to illness are believed to be less effective than patient-centered, biopsychosocial approaches, especially for patients with chronic pain23,32. Patient-centered care means that physicians respond to patients’ preferences, including their preferences for involvement in treatment decision-making1. In the present study, we surveyed patients to determine their preferred and experienced roles in decision-making for carpal tunnel release and attempted to identify the factors that influenced their choices.
This study demonstrates that for carpal tunnel syndrome, which raises issues regarding quality of life rather than those related to a life-threatening condition, the majority of patients (76%) preferred to share decision-making with their physicians and most (85%) experienced shared decision-making regarding carpal tunnel release. Although it is not easy to compare our data with previously published findings, our findings indicate that patients with carpal tunnel syndrome have a greater preference for shared decision-making before surgery than do patients with a life-threatening condition. In a study of 1012 women with breast cancer, only 44% were found to want to select their treatment collaboratively and 34% wanted to delegate the responsibility for decision-making to their physicians27. In another study, 51% of 436 patients who had been newly diagnosed as having cancer wanted to share treatment decision-making33. Other studies have described high preferences for a passive role in decision-making in patients with cancer34,35. Deber et al. compared preferences and found that 64% and 67% of patients with breast or prostate cancer, respectively, preferred to share the decision, whereas 79% and 83% of those with orthopaedic and rheumatologic conditions preferred to do so15.
In the present study, we found that patients who preferred a collaborative role had lower DASH scores, indicating less severe symptoms, than did those who preferred a fully active or a fully passive role. These results suggest that patients with less severe symptoms are more likely to prefer shared decision-making. A previous study regarding the willingness to undergo carpal tunnel release concluded that the most important reason for choosing surgery was symptom severity, suggesting that more severe symptoms can affect patients’ willingness to have an operation18. Patients with more severe symptoms may either actively decide to have surgery or be willing to accept the recommendation of surgery by their surgeons.
In the present study, 50% (thirty-nine) of the seventy-eight patients reported that they experienced their preferred decision-making style, but no significant association was found between this finding and demographic factors. A small number of studies have examined the congruence between preferred and experienced roles. Ford et al. reported that 61% of patients achieved preferred roles with general practitioners36. Murray et al. reported that 70% of the respondents to a survey of a large U.S. population with a variety of diseases had experienced their preferred styles and that a high socioeconomic status and a good relationship with a regular doctor were associated with the achievement of a preferred level of involvement14. The patients in the present study were usually referred for consultation for surgery and had no long-standing doctor-patient relationship with the surgeon. It may be that the reason for the lower congruence (50%) between preferred and perceived style in this study is that we used the five-level Control Preferences Scale, whereas the above two studies used three-level scales (active, collaborative, and passive). In a study by Degner et al., in which women with breast cancer were surveyed with use of the same five-level Control Preferences Scale, only 42% believed that they had achieved their preferred level of involvement in decision-making27.
We found that a history of a surgical procedure was associated with a more active role in decision-making and that having a caregiver and private insurance were strongly associated with experiencing a more active role. It is likely that the experience of a previous operation clarified uncertainties or decisional conflicts regarding surgery and that the presence of a caregiver allowed patients greater freedom during decision-making. Interestingly, although the total cost of bilateral carpal tunnel release is relatively low (less than US$1000) and the national health-care system provides 70% of the cost in the authors’ health-care system, patients with additional private insurance were more likely to experience an active role in decision-making. Previous studies have shown that patients’ desires for an active role are positively related to a feeling of being informed about the condition10,29,37. On the other hand, it has also been shown that a low income, an African-American ethnicity14, and old age15,26 are associated with a preference for paternalism rather than shared decision-making. However, Degner et al. concluded that sociodemographic variables account for only 15% of variances in preferences, which makes it difficult to predict the role that a given patient is likely to prefer26. Therefore, the only way a physician can gain insight into an individual patient's preference is by direct inquiry, which can be regarded as a basic clinical skill for physicians10.
A number of study limitations warrant consideration. First, this is an observational study, and associations found between variables and patients’ preferences and experiences of decision-making may not be causal. A longitudinal study of patients with carpal tunnel syndrome is required to generalize the findings of this study. Second, our cohort was mainly composed of women, and the level of control desired may differ by sex, although sex was not associated with patients’ favoring any decisional style in our literature review. Furthermore, this study was conducted at a tertiary hospital, and our findings may not represent patients with carpal tunnel syndrome in general. Third, this study was performed on patients treated by one physician, and the doctor-patient relationship was limited to consultations regarding carpal tunnel syndrome. Accordingly, the use of different indications for surgery, methods of consultation, and a long-lasting doctor-patient relationship could have resulted in different patient preferences and experiences of decision-making. Fourth, we did the retrospective survey two days after surgery to minimize the time between surgery and the retrospective survey to better reflect the patients’ memory of their decision. Kremer et al.29 also asked their patients with HIV to respond to the Control Preferences Scale by thinking about the most recent decision they had made. However, the early postoperative period with pain and the degree of symptomatic improvement may affect a patient's perception of involvement in the decision. Finally, this study was done in ethnically Korean patients, and cultural and health-care system differences may also affect patient preferences. Future studies are required to determine whether there are differences in patients’ preferences between surgical and medical decision-making, and whether meeting patients’ preferences actually results in better outcomes in carpal tunnel syndrome. If it does, decisional aids should be devised to reduce conflicts and enhance patient participation in the decision-making process38,39.
In conclusion, the majority of patients with carpal tunnel syndrome preferred to share surgical decision-making with the physician, and those preferring a collaborative role had less severe symptoms than those preferring a fully active or a fully passive role in decision-making. A history of a surgical procedure and having a caregiver and private insurance were associated with a more active role. It is hoped that this information may help to promote patient-centered consultation in patients with carpal tunnel syndrome.