INTRODUCTION

People seek musculoskeletal specialty care when a symptom becomes a concern, which is not alleviated in non-specialty care. Unhelpful thoughts regarding symptoms and feelings of worry, despair, or insecurity are common in association with musculoskeletal symptoms.1–4 People who are flexible in their thinking–ready to replace unhelpful thoughts with healthier narratives about their symptoms–experience less intense symptoms and are more capable.5,6

Patients can be coached to enter specialty care visits with an open and curious mind, prepared to move from how they currently regard their symptoms to a potentially healthier mindset.7 Previous studies found that greater patient activation (feeling agency to take action to maintain and improve one’s health)8 is correlated with greater capability, lower pain intensity, and greater satisfaction with treatment.9 A surgical question prompt list (a list of questions that patients can consider as they interact with a surgeon to consider surgery)10 has been used to help increase patient agency and increase participation in decision-making regarding discretionary surgery.11 A question list prompts patients to reflect on and ask about things that matter to them. We are interested in whether or not patients who review a question prompt list before the consultation feel greater comfort with decisions, have a greater health agency, regard their specialist as more empathetic, and are more satisfied with their care.

A Cochrane review identified 16 studies of the effectiveness of a question prompt list for an oncology consultation and found evidence that they enhance patient participation, reduce feelings of anxiety, and improve recall of information.10 One study of patients seeking specialty hand care compared a question prompt list with three simple questions and found comparable patient-perceived involvement in their care. However, it did not address whether either tool improved patient agency and experience or limited decision conflict beyond what patients experience without using either tool.12

We addressed this question: Do patients who receive a question prompt list before a musculoskeletal specialty care visit experience greater comfort with decisions, feel greater agency for their care, perceive their surgeon as more empathetic, and are they more satisfied with their care?

METHODS

Study design and setting

Our Institutional Review Board approved this study at the Office of Research at the University of Texas in Austin. New, English-literate adults (18 years or older) seeking care for musculoskeletal symptoms were invited to enroll from January 2018 to March 2020. The question prompt list was developed by surgeons and patients alike. Questions were considered important after reviewing. After verbal consent and reading a research letter, subjects were randomized 1:1 using a random number generator to be given or not given a question list before their visit.

Participants

Patients were asked by medical assistants throughout all clinical offices and voluntarily decided to participate. Participants were fluent and literate in English, and there was no concern about understanding the list of questions. Participants had an average of 5-10 minutes to review the list before the physician came in. We enrolled 146 adults (60% women) with an average age of 53 (SD=17) [Table 1]. Seventy participants (48%) received a question prompt list. More than half of the patients had non-trauma conditions (77%), and more than half had upper-extremity conditions (59%). Injuries and diagnosis include but are not limited to rotator cuff and labral tears, Dupuytren’s contracture, lumbar radiculopathy, total joint replacement, and pilon fractures.

Table 1.Demographics
N=146
Mean ± SD (Range)
Age 53 ± 17 (18-87)
Median (InterQuartile Range) Skewness
CARE measure 50 (48-50) -2.03
Decision Conflict Score 4 (0-13) 0.98
PAM score 92.31 (82.70-100) -2.5
JSPPPE Score 31 (27-35) -1.77
Gender Percentage (Number)
Male 40% (58)
Female 60% (88)
Race
White 65% (95)
Hispanic 18% (27)
Other 16% (24)
Employment
Employed 60% (88)
Retired/disabled/Student/homemaker 40% (58)
Household Income
<50K 32% (47)
50-99K 25% (36)
100k and Above 43% (63)
Education
High School Degree and Less 19% (28)
College Degree 58% (85)
Post College Graduate Degree 23% (33)
Marital Status
Married 62% (91)
Single 18% (27)
Separated/Divorced/Widowed 19% (28)
Prompt List
Received 48% (70)
Not Received 52% (76)
Diagnosis
Trauma 23% (33)
Non Trauma 77% (113)
Involved Area
Upper Extremity 59% (86)
Lower Extremity 39% (57)
Spine 2% (4)

Measurements

The list suggested questions about diagnosis, cause, treatment options, and recovery. For instance, “What is causing my symptoms?”, “Have I done anything to cause this problem?” “Are there treatments that can stop, slow, or reverse the problem?” finally, questions about the specific treatments offered, such as, “Does this treatment cure the disease?”

After the visit concluded, including a specific treatment plan, the physician left the room, and the research assistant administered the surveys. Patients in both groups completed a demographic survey (age, gender, ethnicity, etc.), the Decisional Conflict Scale (measures personal perceptions of uncertainty in choosing options; modifiable factors contributing to uncertainty and unsupported in decision making, and effective decision making), a measure of perceived clinician empathy (CARE measure), a measure of patient activation (PAM score: assesses an individual´s knowledge, skill and confidence for managing their health and health care), and an 11-point ordinal measure of satisfaction with the visit. The Decisional Conflict Scale was used as our primary outcome variable. It is a validated measure of patient uncertainty in making decisions about their care, the factors contributing to their uncertainty, and whether or not the patient felt involved in their care through shared decision-making.11,13

Statistical Analysis

Descriptive statistics were performed on all patients. Shapiro-Wilk tests were conducted to test the normality of continuous variables. The mean and standard deviation were reported for normally distributed variables; the median and interquartile range were reported for all nonparametric variables. To seek factors associated with receiving a QPL, we performed bivariate analyses using Mann-Whitney U tests for nonparametric response variables. Spearman rank-order correlations were calculated for all continuous explanatory variables. All variables with P < 0.10 were moved to multivariable analysis. Multivariable linear regression analysis was used to identify variables associated with CARE score. Alpha was set at 0.05.

A priori power analysis using G power determined that a sample size of 128 subjects would have 80 percent power to detect a difference of effect size 0.5 with alpha set at 0.05. Anticipating the potential for 10% incomplete surveys, we aimed to enroll more than 141 patients.

RESULTS

In bivariate analysis, there were no significant differences in (Median [IQR]) in decision comfort (4 [0-13], 5 [0-14]), agency (92.3 [80.8-100], 92.3 [83.7-100]), perceived clinician empathy, (31 [26-34], 31 [27.5-35]) and satisfaction with care (50 [48-50], 50 [46.5-50]) between two groups whether they received a question prompt list (QPL) prior to a visit or not [Tables 2-6].

Table 2.Bivariate Analysis of Factors Associated with Receiving A Prompt List
Variable Not Received Received P-value
Mean ± SD Mean ± SD
Age 55 ± 17 51 ± 17 0.22
Median (IQR) Median (IQR)
CARE Measure 50 (46.5-50) 50 (48-50) 0.69
Decision Conflict Score 5 (0-14) 4 (0-13) 0.86
PAM score 92.3(83.7-100) 92.3(80.8-100) 0.49
JSPPPE Score 31 (27.5-35) 31 (26-34) 0.48
Gender Percentage (Number) Percentage (Number)
Male 19% (28) 21%(30) 0.45
Female 33% (48) 27% (40)
Race
White 34% (49) 32% (46) 0.62
Hispanic 11% (16) 7% (11)
Other 7% (11) 9% (13)
Employment
Employed 32% (46) 29% (42) 0.94
Retired/disabled/Student/homemaker 20% (30) 19% (28)
Household Income
<50K 18% (26) 14% (21) 0.43
50-99K 14% (21) 10% (15)
100k and Above 20% (29) 23% (34)
Education
High School Degree and Less 9% (14) 9% (14) 0.38
College Degree 33% (48) 25% (37)
Post College Graduate Degree 9% (14) 13% (19)
Marital Status
Married 34% (50) 28% (41) 0.42
Single 7% (11) 11% (16)
Separated/Divorced/Widowed 10% (15) 9% (13)
Diagnosis
Trauma 40% (59) 37% (54) 0.94
Non Trauma 12% (17) 11% (16)
Involved Area
Upper Extremity 29% (42) 30% (44) 0.61
Lower Extremity 22% (32) 17% (25)
Spine 1% (2) 0.5% (1)

SD: Standard deviation, IQR: interquartile range, CARE: consultation and relational empathy, PAM: patient activation measure, JSPPPE: Jefferson scale of patient perceptions of physician empathy

Table 3.Bivariate Analysis of Factors Associated with Care Measure
Variable Spearman Rho P-value
Age 0.002 0.98
Decision Conflict Score -0.29 <0.001
PAM score 0.24 0.002
JSPPPE Score 0.46 <0.001
Gender Median (IQR)
Male 49.5 (41-50) 0.04
Female 50 (48-50)
Race
White 50 (46-50) 0.23
Hispanic 50 (49-50)
Other 50 (44.5-50)
Employment
Employed 50 (48-50) 0.67
Retired/disabled/Student/homemaker 50 (42-50)
Household Income
<50K 50 (46-50) 0.48
50-99K 50 (48-50)
100k and Above 50 (47-50)
Education
High School Degree and Less 50 (43-50) 0.84
College Degree 50 (48-50)
Post College Graduate Degree 50 (48-50)
Marital Status
Married 50 (43-50) 0.61
Single 50 (48-50)
Separated/Divorced/Widowed 50 (48-50)
Diagnosis
Trauma 50 (48-50) 0.39
Non Trauma 50 (46-50)
Involved Area
Upper Extremity 50 (48-50) 0.56
Lower Extremity 50 (46-50)
Spine 50 (48-50)

SD: Standard deviation, IQR: interquartile range, CARE: consultation and relational empathy, PAM: patient activation measure, JSPPPE: Jefferson scale of patient perceptions of physician empathy

Table 4.Bivariate Analysis of Factors Associated with PAM Score
Variable Spearman Rho P-value
Age 0.008 0.92
Decision Conflict Score -0.57 <0.001
JSPPPE Score 0.28 <0.001
Gender Median (IQR)
Male 91.34 (80.76-100) 0.38
Female 94.23 (82.69-100)
Race
White 90.38 (80.76-100) 0.22
Hispanic 94.23 (82.69-100)
Other 95.19 ( 88.46-100)
Employment
Employed 92.30 (82.69-100) 0.72
Retired/disabled/Student/homemaker 93.26 ( 80.76-100)
Household Income
<50K 92.30 (80.76-100) 0.62
50-99K 92.30 (82.69-99.03)
100k and Above 92.30 ( 82.69-100)
Education
High School Degree and Less 92.30 ( 83.65-100) 0.71
College Degree 92.30 (82.69-100)
Post College Graduate Degree 92.30 (82.69-100)
Marital Status
Married 92.30 (80.76-100) 0.65
Single 90.38 (84.61-100)
Separated/Divorced/Widowed 96.15 ( 86.53-100)
Diagnosis
Trauma 92.30 ( 82.69-100) 0.33
Non Trauma 94.23 (84.61-100)
Involved Area
Upper Extremity 92.30 (82.69-100) 0.68
Lower Extremity 94.23 (82.69-100)
Spine 92.30 (75 -92.31)

SD: Standard deviation, IQR: interquartile range, CARE: consultation and relational empathy, PAM: patient activation measure, JSPPPE: Jefferson scale of patient perceptions of physician empathy

Table 5.Bivariate Analysis of Factors Associated with Descional Conflict
Variable Spearman Rho P-value
Age -0.02 0.8
JSPPPE Score -0.32 <0.001
Gender Median (IQR)
Male 5 (0-14) 0.69
Female 4 (0-13)
Race
White 4 (0-14) 0.68
Hispanic 4 (0-9)
Other 4 (0-13)
Employment
Employed 5 (0-15) 0.33
Retired/disabled/Student/homemaker 4 (0-11)
Household Income
<50K 5 (0-13) 0.82
50-99K 3.5 (0-12.5)
100k and Above 5 (0-15)
Education
High School Degree and Less 5 (0-14.5) 0.86
College Degree 4 (0-13)
Post College Graduate Degree 5 (0-14)
Marital Status
Married 5 (0-14) 0.88
Single 4 (0-11)
Separated/Divorced/Widowed 4 (0-12.5)
Diagnosis
Trauma 4 (0-14) 0.29
Non Trauma 4 (0-7)
Involved Area
Upper Extremity 4 (0-13) 0.79
Lower Extremity 5 (0-14)
Spine 16 (0 -16)

SD: Standard deviation, IQR: interquartile range, CARE: consultation and relational empathy, PAM: patient activation measure, JSPPPE: Jefferson scale of patient perceptions of physician empathy

Table 6.Bivariate Analysis of Factors Associated with JSPPPE
Variable Spearman Rho P-value
Age 0.09 0.8
Gender Median (IQR)
Male 30 (25-35) 0.18
Female 32 (27.5-34)
Race
White 31 (27-34) 0.47
Hispanic 31.5 (25-34.5)
Other 32 (27-35)
Employment
Employed 31 (27-34) 0.93
Retired/disabled/Student/homemaker 31 (26-35)
Household Income
<50K 32 (27-35) 0.76
50-99K 29.5 (26-33)
100k and Above 31 (28-35)
Education
High School Degree and Less 31.5 ( 26.5- 33.5) 0.65
College Degree 31 ( 27- 35)
Post College Graduate Degree 31 ( 28- 34)
Marital Status
Married 31 ( 26- 34) 0.66
Single 31 ( 26- 35)
Separated/Divorced/Widowed 32 (27.5-35)
Diagnosis
Trauma 31 (26-34) 0.25
Non Trauma 31 (28-35)
Involved Area
Upper Extremity 31 (26-34) 0.65
Lower Extremity 31 (27-35)
Spine 33 (30 -35)

SD: Standard deviation, IQR: interquartile range, CARE: consultation and relational empathy, PAM: patient activation measure, JSPPPE: Jefferson scale of patient perceptions of physician empathy

DISCUSSION

Question lists encouraged discussion and improved patient knowledge in prior studies of people considering surgery.10,14 We tested the influence of a question list on patient experience after an initial musculoskeletal specialty care visit. We did not find a difference in comfort with decisions or other aspects of patient experience when people received a question list prior to their visit.

Our study has limitations. First, it was performed at several musculoskeletal specialty care offices in a relatively wealthy, educated, White urban center in the United States [Table 1]. There is evidence that Black patients generally have shorter visits, ask fewer questions, and participate less than their counterparts.14,15 Patients with lower incomes have been shown to ask fewer questions.11 Therefore, it is possible that these populations could benefit more from a question prompt list than the people in our sample. Although the question list was not reviewed by the physician prior to each interaction, nor were they made aware of who received the list, the questions asked by patients may have alerted clinicians to patients who received the prompt list. A study that intentionally made clinicians aware of whether or not patients received a question list might alter clinician behavior and result in different outcomes—a hypothesis that merits testing. The current study addressed how experience with specialty care might be altered when patients were more prepared and inquisitive.

The observation that there was no measured benefit to receiving a question list suggests that these tools may not be sufficient to improve patient-clinician communication and development of health strategies or that the tools we used to measure a difference were inadequate. A list that prompts patients to ask about things that matter to them may or may not improve their connection with the surgeon and their perception of the clinician’s empathy. Furthermore, experience measures (empathy and satisfaction) have known shortcomings as research tools, such as powerful ceiling effects and non-normal distributions.16 Other measures, such as anxiety regarding symptoms and trust in the clinician, might register a difference.14,16 A systematic review of studies of question prompt lists for patients receiving cancer care has demonstrated associations with increased question asking, decreased anxiety over time, and improved recall as subsequent visits,17 none of which were measured in this study. On the other hand, evidence suggests that a greater difference can be made by adjusting clinician tactics than by providing patients with tools. For instance, one trial in oncology randomized clinicians to train how to engage patients in care decisions or whether patients should receive a communication aid. They found that training clinicians had a large positive effect on independent observer ratings of the involvement of patients in decisions and a positive impact on patient’s perception of their involvement. The patient communication aid did not improve feelings of involvement. Neither intervention affected satisfaction with the consultation or decisional conflict.18 There are important differences between cancer care, where treatment is necessary, and the potential harms of treatment are greater, and upper extremity specialty care, where most treatments are discretionary, which makes even small potential harms potentially more influential.

We are interested in developing tools and interventions to help patients explore what matters most to them (their values), correct any misconceptions about what tests and treatments offer, and allow patients to choose treatment and testing options most consistent with their values and priorities. Clinician variation in biases and priorities with regard to tests and treatment can be tempered through shared decision-making with patient values. In future studies, we aim to target injuries based on acuity as we have acknowledged that patient preparation regarding questions they may ask and their anxiety around the clinical encounter will differ among patients. We found that—in a population of relatively socially advantaged people–a question list was not associated with lower decision conflict or greater perceived empathy, satisfaction, or agency. Future research can address the use of specific tools in less advantaged patients. Still, we think the current evidence may point to a need for tools and strategies that alter clinician behavior, perhaps by emphasizing guiding rather than directing patient approaches in a musculoskeletal specialty care visit.


Declaration of conflict of interest

The authors do not have any potential conflicts of interest pertinent to this manuscript.

Declaration of funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Declaration of ethical approval for study

Approved by UT Austin IRB. IRB number: 2017-02-0122

There is no information (names, initials, hospital identification numbers, or photographs) in the submitted manuscript that can be used to identify patients.