Data and Statistical Analysis
The data collected were tabulated using the Microsoft Excel program. Overall summary statistics of patient baseline characteristics were calculated in terms of mean values and SDs for continuous variables and percentages and frequencies for discrete variables. The scoring system for each item in the CARE instrument is “poor” = 1, “fair” = 2, “good” = 3, “very good” = 4, and “excellent” = 5. All 10 items are then added, giving a maximum possible score of 50 and a minimum of 10. Up to two “Not Applicable” responses or missing values are allowable and are replaced with the average score for the remaining items. Questionnaires with more than two missing values or “Not Applicable” responses are removed from the analysis. In our study, no questionnaires contained more than two “Not Applicable” answers.
A bivariate analysis will be performed to compare the variables and their values among the two principal groups: control group versus anatomic model group. The choice of variables considered for the multivariate analysis was based on clinical relevance. Categoric variables compared and analyzed using the chi-square and Fisher exact tests. Mann-Whitney U tests were used to compare and analyze continuous variables. P < 0.05 was considered statistically significant for all tests.
A total of 304 patients responded to our survey (Table 1). Mean age of the respondents in the control group versus experimental group did not differ significantly (55.0 versus 53.2 years; P = 0.321). Civil status of the participating patients was most predominantly married and single in the control group with 38.4% and 33.9% of participants, respectively. In the experimental group, the married and single were also the most common civil status with 32.8% and 47.5%, respectively. No statistical difference between the groups was found with respect to civil status (P = 0.121). Medical insurance distribution analysis did not identify any statistical difference (P = 0.605). It is worthy to mention that most participants in the study had Medicaid as medical insurance (n = 185). Level of education did not differ among study groups (P = 0.401).
About half of the participants reports associated trauma for evaluation in the groups (ie, 42.4% control group and 36.8% anatomic group). No statistical difference was found between the control and experimental groups (P = 0.199) with regard to the mechanism of injury. In summary, the experimental and control groups are considered similar among the sociodemographic data analyzed. CARE questionnaire median scores (Table 2) for the control group were 44.0 ± 9.0. Scores for the experimental group were 48.0 ± 7.7. Comparison of scores between the two groups did not find any statistical difference (P = 0.482).
Figure 3 is a box plot showing the distribution of CARE questionnaire scores between the control and anatomic model groups. Lower individual scores were found in the control group, but the overall distribution of scores is similar in both groups. Average score overlapping can be visualized in the box plot.
Figure 4 is a histogram showing the difference in distribution of the time of the encounter between the two groups. The anatomic model group encounters on average lasted longer than the control group; no statistical difference was found (P < 0.0005).
Analyses of the sociodemographic characteristics did not reveal any significant difference between the control and experimental groups (Table 1). According to the reference CARE Measure normative value (Figure 2), a 50th percentile score correlates with 45.75 median score. The control group's median CARE score is 46.0. CARE Measure scores for the control group followed a similar distribution to the normative values. The experimental group's median CARE score is 48.0. When interpreting this score with the provided normative values, it correlates with a 90 to 95 percentile in satisfaction scores. Statistical analyses between median scores did not find any statistical difference, for which the differences in satisfaction percentile cannot be directly linked to the use of anatomic models.
The encounter time was significantly increased with the use of anatomic models (P < 0.005). The study does not analyze the content of the encounter and the difference of distribution of time throughout the encounter. Previous studies have suggested that most time during an encounter with a surgeon was invested in educating and helping patient makes choices about their care.3 It would be interesting to measure whether the use of the anatomic model affects the content and more specifically the allotted time in each component of the encounter. Another study reported that primary care physician patients were more satisfied with the longer encounter time but had not found the association with ambulatory orthopaedics.1 Our study confirms that the longer encounter time in the orthopaedic appointment does not mean higher empathy perception and satisfaction scores.
In this study, the control and experimental groups did not differ in the level of education between them, but it seems that our participant level of education is skewed to lesser formal education. This phenomenon may limit the generalizability of our study because more formal education may demand different information about their condition. In a recent study, it was suggested that less formal education had lower patient satisfaction scores.14 This suggestion opens the possibility of developing a study to compare the use of anatomic models that differ in educational level and how this affects patients' perceived empathy and satisfaction scores with the desired intervention. It is possible that the intervention of the anatomic model proves to be more beneficial to specific populations that share certain characteristics instead of being a measure that may help all patients.
The absence of differences between the groups on our study could be due to various variables. Although patients were not aware of the study during the encounter, participating physicians were. This phenomenon could create an observer bias, which may result in higher scores for the control group because of better performance of the physician during the encounter. Because of our study design, participating investigators had to know about the days the study had to be conducted because they were responsible to make the approach for study participation at the end of the encounter. To eliminate this bias, and possibly find more precise scores, the physician cannot be involved in the approach to participate in the questionnaire.
Validity of the CARE questionnaire could be affected because of our study population. Most of our patients are Spanish speaking. No study has been conducted to validate the translation of the measure in Spanish. Translations of documents do not necessarily translate the spirit of the document in the desired language. Interestingly, our control group scores correlated with the expected normative values of the CARE Measure, but additional studies need to be conducted to validate the questionnaire in Spanish. This phenomenon may limit the power to find any statistical difference with our proposed intervention.
When defining empathy, the opinion of the patient seems to be more important than the physician's perspective of self.13 Cultural context has to be an important aspect to improve empathy scores with our patients. In a recent study, the second most common issue for report development was lack of proper communication.18 It is therefore important, as suggested by a study, to recognize that patient preferences as per the amount of information desired for their diagnosis may vary.9 Every physician should be familiarized with the population they serve and general characteristics that define them. If clinicians do not accurately assess preferences for information, it can result in a conflict between the ideal and actual care received.10
Satisfied patients do not necessarily have better outcomes but are more likely to comply with treatments, keep office appointments, and not file complaints or lawsuits.16 Pamphlets and informational brochures can supplement, but not replace, effective communication.4 Empathy is dependent on a cognitive response, which is the capacity to identify and understand another.7 Research suggests that empathy is a skill that can be taught, and surgical curricula in residency should include empathy training.6 A provider who is not content with his or her own patient satisfaction scores needs to question himself what measures need to be done to improve.17 A previous study specifically demonstrates that surgeons demonstrate weakness in the ability to assess the patient understanding of the situation, discussing risks and uncertainties associated with their care.3 Given the multiple variable that seem to affect patient satisfaction scoring, it seems that physicians will need to be actively studying the characteristics of their population to identify strategies to improve satisfaction scores.17
To our knowledge, this is the first study that directly addresses the use of anatomic models during the medical encounter and how these affect the empathy perception and patient satisfaction scores in the orthopaedic population. Anatomic models are just one of the many tools that may improve communication, empathy, and patient satisfaction. Although there may be various nonmodifiable patient factors, such as age and geographic location,11 there are also modifiable variables that can be addressed to improve patient satisfaction. Previous research has reported that only 53% of the patients treated by orthopaedic surgeons disclosed real concerns about surgery.8 Consciousness about the topic may help the orthopaedic surgeon to improve his interaction with patients. Practical solutions such as the use of question list made pre-encounter to remind patients to ask important questions may make significant differences in the communication.12 All aspects of the encounter need to be addressed to optimize patient satisfaction.
The use of anatomic models did not confer a statistical advantage to improve the perceived empathy and patient satisfaction during an initial appointment. The encounter time was increased with the use of anatomic models. As orthopaedic surgeons, we are responsible for finding strategies to improve the communication skills with our patients. Dedicating time to understand the variables that affect patient satisfaction may help improve patient-physician interaction. Further investigation should be considered to use other strategies to provide better care for our patients.
1. Parrish RC, Menendez ME, Mudgal CS, Jupiter JB, Chen NC, Ring D: Patient satisfaction and its relation to perceived visit duration with a hand surgeon. J Hand Surg Am 2016;41:257–62.e1-4.
2. Steinhausen S, Ommen O, Antoine SL, Koehler T, Pfaff H, Neugebauer E: Short and long term subjective medical treatment outcome of trauma surgery patients: The importance of physician empathy. Patient Prefer Adherence 2014;8:1239–1253.
3. Levinson W, Hudak P, Tricco AC: A systematic review of surgeon-patient communication: Strength and opportunities for improvement. Patient Educ Couns 2013;93:3–17.
4. Tongue J, Epps HR, Forese LL: Communication skills for patient-centered care. Instr Course Lect 2005;54:3–9.
5. Levinson W, Roter DL, Mullooly JP, Dull VT, Frankel RM: Physician-patient communication: The relationship with malpractice claims among primary care physicians and surgeons. JAMA 1997;277:553–559.
6. Han JL, Pappas TN: A review of empathy, its importance, and its teaching in surgical training. J Surg Educ 2018;75:88–94.
7. Mercer SW, Reynolds WJ: Empathy and quality of care. BR J Gen Pract 2002;52:S9–S12.
8. Hudak PL, Armstrong K, Braddock C III, Frankel RM, Levinson W: Older patients unexpressed concerns about orthopedic surgery. J Bone Joint Surg Am 2008;90:1427–1435.
9. Grady A, Carey M, Bryant J, Sanson-Fisher R, Hobden B: A systematic review of patient practioner communication interventions involving treatment decisions. Patient Educ Couns 2017;100:199–211.
10. Elkin EB, Kim SH, Casper ES, Kissane DW, Schrag D: Desire for information and involvement in treatment decisions: Elderly cancer patient's preferences and their physician's perceptions. J Clin Oncol 2007;25:5275–5280.
11. Abtahi AM, Presson AP, Zhang C, Saltzman CL, Tyser AR: Association between orthopedic outpatient satisfaction and non modifiable patient factors. J Bone Joint Surg Am 2015;97:1041–1048.
12. Lim L, Chow P, Wong CY, et al: Doctor—patient communication, knowledge, and question prompt lists in reducing preoperative anxiety—a randomized control study. Asian J Surg 2011;34:175–180.
13. Dehning S, Reiß E, Krause D, et al: Empathy in high-tech and high-touch medicine. Patient Educ Couns 2014;95:259–264.
14. Bible J, Kay HF, Shau DN, O'Neill KR, Segebarth PB, Devin CJ: What patient characteristics could potentially affect patient satisfaction scores during a spine clinic? Spine (Phila Pa 1976) 2015;40:1039–1044.
15. Mercer SW, Maxwell M, Heaney D, Watt GC: The consultation and relational empathy measure: Development and preliminary validation and reliability of an empathy—based consultation process measure. Fam Pract 2004;21:699–705.
16. Menendez ME, Chen NC, Mudgal CS, Jupiter JB, Ring D: Physician empathy as a driver of hand surgery patient satisfaction. J Hand Surg Am 2015;40:1860–1865.e2.
17. Smith MJ, Choma TJ: Patient satisfaction in musculoskeletal medicine. Curr Rev Musculoskelet Med 2017;10:207–211.
Copyright © 2018 The Authors. Published by Wolters Kluwer Health, Inc. on behalf of the American Academy of Orthopaedic Surgeons
18. King JD, van Dijk P, Overbeek CL, Hageman M, Ring D: Patient complaints emphasize non technical aspects of care at tertiary referral hospital. Arch Bone Jt Surg 2017;5:74–81.