- Email: [email protected]
Cancer patients' satisfaction with physicians: Princess Margaret hospital satisfaction with doctor questionnaire results
Cancer patients’ satisfaction with physicians: Princess Margaret Hospital Satisfaction with Doctor questionnaire results Charles N. Landen, Jr, MD,a Novie O. Younger, PhD,b Beth A. Collins Sharp, PhD, RN,a and Paul B. Underwood, MDa Charleston, SC OBJECTIVE: The study was performed to examine the satisfaction of a specific population of oncology patients with their physicians and to quantify its association with characteristics of their disease. STUDY DESIGN: A descriptive design was used in which patients attending a weekend cancer support retreat completed the Princess Margaret Hospital Satisfaction with Doctor Questionnaire (PMH-PSQ-MD). Demographic information was requested separately. RESULTS: Of 48 patients, 96% completed the PMH-PSQ-MD, and 67% mailed in the additional demographics. Average overall score was 3.08 (SD = 0.56, 4 being most satisfied). Patients reported a desire for more time with physicians and that their pain be better understood. Patients were most satisfied with the physician’s honesty, thoroughness, and communication. Satisfaction scores did not correlate with intensity of treatment or time since diagnosis. Scores declined as financial burden of therapy increased, but the trend was not statistically significant. CONCLUSION: Patients attending this retreat were generally satisfied with their physicians, regardless of treatment intensity. Increasing financial burden may be associated with physician dissatisfaction. (Am J Obstet Gynecol 2003;188:1177-9.)
Key words: Cancer, patient satisfaction, questionnaire
Clinicians are placing increased importance on the patient’s perception of her quality of care. In an effort to improve patient care, the patients themselves are being used as valuable resources.1 In addition to the self-evident desire for physicians to believe that patients are satisfied with their physicians, a satisfied patient is more compliant, takes a greater interest in her care, is less likely to miss appointments, and is more likely to retain information given.2-4 A reliable instrument to measure the level of a patient’s satisfaction provides several benefits. Institutional deficits in patient care can be identified and corrected, changes in policy can be evaluated for their effectiveness, and individual physicians can identify ways in which they can improve care.5 The Princess Margaret Hospital Satisfaction with Doctor Questionnaire (PMH-PSQ-MD) was developed and validated specifically for use in oncology patients in an outpatient setting.6 We gave this questionnaire to the par-
From the Departments of Obstetrics and Gynecologya and Biometry and Epidemiology,b Medical University of South Carolina. Reprint requests: Charles N. Landen, Jr, MD, Department of Obstetrics and Gynecology, 96 Jonathan Lucas Blvd, 634 CSB, Charleston, SC 29425. © 2003, Mosby, Inc. All rights reserved. 0002-9378/2003 $30.00 + 0 doi:10.1067/mob.2003.281
ticipants at Camp Bluebird, an annual weekend camp for cancer patients. Results were discussed with patients in an open forum in an attempt to measure overall patient satisfaction, identify specific areas of care on which we can improve, and find patient characteristics associated with patient dissatisfaction. Results were also shared with the physicians at the local cancer center. Methods The PMH-PSQ-MD was generously provided by their authors.6 It presents 41 statements about their physicians in the categories of information exchange, interpersonal skills, empathy, and quality of time and was validated for outpatient use with a Cronbach α score of .97. Patients responded to statements from “strongly disagree” to “strongly agree,” scored 1 to 4. For items that elicit negative responses, scores are reversed. Each patient’s score is an average of 41 equally weighted responses. The PMHPSQ-MD was developed for use after a single office visit, so the tense of questions was adjusted to refer to overall care. On arrival to Camp Bluebird, an annual weekend retreat for cancer survivors volunteering attendance in Seabrook Island, SC, campers completed the PMH-PSQMD with approval of the International Review Board of the Medical University of South Carolina. The following day, campers participated in a discussion, guided by their anonymous responses, on improving physician care. Patients gave consent to have their answers anonymously 1177
1178 Landen et al
May 2003 Am J Obstet Gynecol
Table I. Most positive and negative patient responses Question
Actual response*
Adjusted response†
1.54 3.42 3.42 3.38 1.64
3.46 3.42 3.42 3.38 3.36
2.54 2.71 2.27 2.20 2.17
2.46 2.71 2.73 2.80 2.83
Most positive top 5 18. I would not recommend this doctor to a friend 25. The doctor explained the reason why the treatment was recommended for me 23. I felt the doctor was being honest with me 3. The doctor considered my individual needs when treating my condition 24. I felt the doctor diagnosed my condition without enough information Most negative top 5 9. There are some things about my visit with the doctor that can be better 31. The doctor understands that pain is a problem for me. 26. The doctor often seems rushed 20. The doctor should tell me more about how to care for my condition 8. There usually isn’t enough time to tell the doctor everything I wanted
*Strongly agree = 4; agree = 3; disagree = 2; strongly disagree = 1. †The adjusted response for negative questions is the reverse score of actual responses (1 becomes 4, 2 becomes 3, etc), so that a patient strongly disagreeing with a negative statement gives a favorable answer, as in question No. 18. For adjusted scores, a higher score is more favorable for the physician.
Table II. Unadjusted patient characteristics and satisfaction scores Satisfaction score Factor
Characteristic
Mean ± SD
No.
P value*
<6 mo 6-24 mo 2-5 y >5 y
3.74 ± 0.30 3.21 ± 0.85 2.91 ± 0.64 3.15 ± 0.45
2 7 6 15
.30
Low Moderate Severe
3.02 ± 0.19 3.37 ± 0.43 2.98 ± 0.82
6 13 11
.95
Minimal Moderate Significant
3.39 ± 0.47 3.25 ±0.42 2.87 ± 0.84
4 17 9
.17§
Time since diagnosis
Intensity of therapy†
Financial burden‡
*From the nonparametric Wilcoxon rank sum tests for trend in the scores as either time since diagnosis, intensity of therapy, or financial burden increased. †Assigned by the authors based on number of hospitalizations and surgeries, modalities of treatment, complications, and need for home health. ‡As assessed by the patient; actual costs to the patients are not available. §When adjusted for race, sex, age, time since diagnosis, and intensity of therapy, “minimal” and “moderate” financial burden scores were significantly lower than “significant” burden (P = .04 and .02, respectively).
shared with the audience and/or in publication. A separate survey on patient demographics and treatment course was handed out, to be completed at their convenience in the interest of time taken from camp activities. Physicians at our cancer center were given the results in an effort to improve care. Data analysis methods included summary statistics of satisfaction scores, nonparametric Wilcoxon rank sum tests for trend, and regression analysis of covariance. Results Forty-six of 48 (96%) campers agreed to complete the PMH-PSQ-MD; 31 of these 46 (67%) completed and returned the additional demographic information. Partici-
pants’ ages averaged 55 years (range 25-80 years), 87% were women, and 90% were white. Average education was 2 years of college. The most common cancer type was breast (18/31, 58%), with average time since diagnosis 6.75 years (range 4 months to 33 years). To protect confidentiality, their physicians’ names were not requested, but given the wide range of locations from which the campers came, presumably many physicians and institutions were represented. On a scale of 1 to 4, with 4 being the most satisfied, the average questionnaire score was 3.08 (± 0.56, range 1.41 to 3.95). Table I shows the questions with the five highest scores for positive and negative features of physician assessment. Limited space precludes listing scores for all 41
Landen et al 1179
Volume 188, Number 5 Am J Obstet Gynecol
questions. Tabulated values indicate that these patients were most satisfied with physician honesty, thoroughness at diagnosis and treatment, and communication. More than 90% agreed that they would recommend their physician to a friend. Patients were most dissatisfied with the short amount of time spent with their physician and the physician’s lack of understanding of their pain. Comparison of satisfaction scores with characteristics of their disease is shown in Table II. There was no significant trend in satisfaction score with increasing treatment intensity, time since diagnosis, or financial burden. Although regression analysis using scores as continuous variables showed that patients with the heaviest financial burden were significantly less satisfied than those with a minimal (P = .04) or moderate (P = .02) burden, analysis of trend is probably more appropriate and was not statistically significant. Anonymous discussion of responses with the patients and a panel of physicians the day after collection was very well received. Most acknowledged that completion of the questionnaire was a positive experience because it gave them an opportunity to discuss frustrations with medical care and showed that physicians were interested in their opinion and on improving care. Comment The patients who attend Camp Bluebird seem satisfied with their physicians. These campers represent a subset of the population who have survived treatment, are currently healthy, are interested sharing their experiences, and have access to medical care. These biases certainly make it impossible to generalize their attitudes to the entire oncologic population, but we can obtain valuable information from administration of this questionnaire. Evaluation of individual questions within the PMH-PSQMD shows that these patients are satisfied with their physicians’ competence and communication. They believe that physicians should try to gain a clearer understanding the pain and fatigue experienced by patients. Not surprisingly, patients would like more time with their physicians. In the open discussion forum, they were understanding of the time demands on physicians and the unpredictable nature of managing cancer patients in an outpatient setting, which often results in unforeseeable delays. These results are the most useful data from this study because it provides specific areas in which we can improve. Because the campers were given dedicated time to complete the questionnaire, as seen in the 96% response rate, selection bias that is often encountered in voluntary
surveys was overcome. Selection bias and recall bias do, however, apply to the 67% of patients returning the demographic questionnaire. We found that their overall satisfaction was not influenced by the degree of therapy intensity or time since diagnosis but may have an inverse association with financial burden. Patients may extrapolate expensive care to negative feelings toward their physician, or perhaps those with little satisfaction in their physicians may overestimate their financial burden. We cannot make any statements about causation from our data. Nevertheless, given the lack of correlation between satisfaction and disease severity, and the responses to questions noted in Table I, this cohort of patients appears to attribute the quality of their physician with the way in which their care is delivered, not the ultimate outcome of their disease. Although the low number of patients surveyed and biases present prohibit making statements applicable to the general oncology population, we believe the method could identify patient characteristics that are associated with dissatisfaction in a larger number of patients. The PMH-PSQ-MD can provide valuable information to the oncologist regarding the quality of care delivered, as perceived by the patient, and allow a focused effort on improving patient satisfaction. In a review of the literature, several studies evaluate patient preferences for treatment options, but few examine satisfaction with the physician. Knowing the needs of our patients would allow targeted improvement in delivery of care, and possibly improved patient outcomes. It is hoped that delivery of these responses to our own physicians, as we have done, will lead to improvements in delivery of care.
REFERENCES
1. Oberst MT. Methodology in behavioral and psychosocial cancer research: patients’ perceptions of care—measurement of quality and satisfaction. Cancer 1984;53(Suppl):2366-75. 2. Ware JE Jr, Davies AR. Behavioral consequences of consumer dissatisfaction with medical care. Eval Program Plann 1983;6:291-7. 3. Sinf J, Weiss B. Patient satisfaction with health care: intentions and change in plan. Eval Program Plann 1991;14:299-306. 4. Rubin HR, Gandek B, Rogers WH, Kosinski M, McHorney CA, Ware JE Jr. Patients’ ratings of outpatient visits in different practice settings: results from the Medical Outcomes Study. JAMA 1993;270:835-40. 5. Hopton JL, Howie JG, Porter AM. The need for another look at the patient in general practice satisfaction surveys. Fam Pract 1993;10:82-7. 6. Loblaw DA, Bezjak A, Bunston T. Development and testing of a visit-specific patient satisfaction questionnaire: the Princess Margaret Hospital Satisfaction With Doctor Questionnaire. J Clin Oncol 1999;17:1931-8.
Key words: Cancer, patient satisfaction, questionnaire
Clinicians are placing increased importance on the patient’s perception of her quality of care. In an effort to improve patient care, the patients themselves are being used as valuable resources.1 In addition to the self-evident desire for physicians to believe that patients are satisfied with their physicians, a satisfied patient is more compliant, takes a greater interest in her care, is less likely to miss appointments, and is more likely to retain information given.2-4 A reliable instrument to measure the level of a patient’s satisfaction provides several benefits. Institutional deficits in patient care can be identified and corrected, changes in policy can be evaluated for their effectiveness, and individual physicians can identify ways in which they can improve care.5 The Princess Margaret Hospital Satisfaction with Doctor Questionnaire (PMH-PSQ-MD) was developed and validated specifically for use in oncology patients in an outpatient setting.6 We gave this questionnaire to the par-
From the Departments of Obstetrics and Gynecologya and Biometry and Epidemiology,b Medical University of South Carolina. Reprint requests: Charles N. Landen, Jr, MD, Department of Obstetrics and Gynecology, 96 Jonathan Lucas Blvd, 634 CSB, Charleston, SC 29425. © 2003, Mosby, Inc. All rights reserved. 0002-9378/2003 $30.00 + 0 doi:10.1067/mob.2003.281
ticipants at Camp Bluebird, an annual weekend camp for cancer patients. Results were discussed with patients in an open forum in an attempt to measure overall patient satisfaction, identify specific areas of care on which we can improve, and find patient characteristics associated with patient dissatisfaction. Results were also shared with the physicians at the local cancer center. Methods The PMH-PSQ-MD was generously provided by their authors.6 It presents 41 statements about their physicians in the categories of information exchange, interpersonal skills, empathy, and quality of time and was validated for outpatient use with a Cronbach α score of .97. Patients responded to statements from “strongly disagree” to “strongly agree,” scored 1 to 4. For items that elicit negative responses, scores are reversed. Each patient’s score is an average of 41 equally weighted responses. The PMHPSQ-MD was developed for use after a single office visit, so the tense of questions was adjusted to refer to overall care. On arrival to Camp Bluebird, an annual weekend retreat for cancer survivors volunteering attendance in Seabrook Island, SC, campers completed the PMH-PSQMD with approval of the International Review Board of the Medical University of South Carolina. The following day, campers participated in a discussion, guided by their anonymous responses, on improving physician care. Patients gave consent to have their answers anonymously 1177
1178 Landen et al
May 2003 Am J Obstet Gynecol
Table I. Most positive and negative patient responses Question
Actual response*
Adjusted response†
1.54 3.42 3.42 3.38 1.64
3.46 3.42 3.42 3.38 3.36
2.54 2.71 2.27 2.20 2.17
2.46 2.71 2.73 2.80 2.83
Most positive top 5 18. I would not recommend this doctor to a friend 25. The doctor explained the reason why the treatment was recommended for me 23. I felt the doctor was being honest with me 3. The doctor considered my individual needs when treating my condition 24. I felt the doctor diagnosed my condition without enough information Most negative top 5 9. There are some things about my visit with the doctor that can be better 31. The doctor understands that pain is a problem for me. 26. The doctor often seems rushed 20. The doctor should tell me more about how to care for my condition 8. There usually isn’t enough time to tell the doctor everything I wanted
*Strongly agree = 4; agree = 3; disagree = 2; strongly disagree = 1. †The adjusted response for negative questions is the reverse score of actual responses (1 becomes 4, 2 becomes 3, etc), so that a patient strongly disagreeing with a negative statement gives a favorable answer, as in question No. 18. For adjusted scores, a higher score is more favorable for the physician.
Table II. Unadjusted patient characteristics and satisfaction scores Satisfaction score Factor
Characteristic
Mean ± SD
No.
P value*
<6 mo 6-24 mo 2-5 y >5 y
3.74 ± 0.30 3.21 ± 0.85 2.91 ± 0.64 3.15 ± 0.45
2 7 6 15
.30
Low Moderate Severe
3.02 ± 0.19 3.37 ± 0.43 2.98 ± 0.82
6 13 11
.95
Minimal Moderate Significant
3.39 ± 0.47 3.25 ±0.42 2.87 ± 0.84
4 17 9
.17§
Time since diagnosis
Intensity of therapy†
Financial burden‡
*From the nonparametric Wilcoxon rank sum tests for trend in the scores as either time since diagnosis, intensity of therapy, or financial burden increased. †Assigned by the authors based on number of hospitalizations and surgeries, modalities of treatment, complications, and need for home health. ‡As assessed by the patient; actual costs to the patients are not available. §When adjusted for race, sex, age, time since diagnosis, and intensity of therapy, “minimal” and “moderate” financial burden scores were significantly lower than “significant” burden (P = .04 and .02, respectively).
shared with the audience and/or in publication. A separate survey on patient demographics and treatment course was handed out, to be completed at their convenience in the interest of time taken from camp activities. Physicians at our cancer center were given the results in an effort to improve care. Data analysis methods included summary statistics of satisfaction scores, nonparametric Wilcoxon rank sum tests for trend, and regression analysis of covariance. Results Forty-six of 48 (96%) campers agreed to complete the PMH-PSQ-MD; 31 of these 46 (67%) completed and returned the additional demographic information. Partici-
pants’ ages averaged 55 years (range 25-80 years), 87% were women, and 90% were white. Average education was 2 years of college. The most common cancer type was breast (18/31, 58%), with average time since diagnosis 6.75 years (range 4 months to 33 years). To protect confidentiality, their physicians’ names were not requested, but given the wide range of locations from which the campers came, presumably many physicians and institutions were represented. On a scale of 1 to 4, with 4 being the most satisfied, the average questionnaire score was 3.08 (± 0.56, range 1.41 to 3.95). Table I shows the questions with the five highest scores for positive and negative features of physician assessment. Limited space precludes listing scores for all 41
Landen et al 1179
Volume 188, Number 5 Am J Obstet Gynecol
questions. Tabulated values indicate that these patients were most satisfied with physician honesty, thoroughness at diagnosis and treatment, and communication. More than 90% agreed that they would recommend their physician to a friend. Patients were most dissatisfied with the short amount of time spent with their physician and the physician’s lack of understanding of their pain. Comparison of satisfaction scores with characteristics of their disease is shown in Table II. There was no significant trend in satisfaction score with increasing treatment intensity, time since diagnosis, or financial burden. Although regression analysis using scores as continuous variables showed that patients with the heaviest financial burden were significantly less satisfied than those with a minimal (P = .04) or moderate (P = .02) burden, analysis of trend is probably more appropriate and was not statistically significant. Anonymous discussion of responses with the patients and a panel of physicians the day after collection was very well received. Most acknowledged that completion of the questionnaire was a positive experience because it gave them an opportunity to discuss frustrations with medical care and showed that physicians were interested in their opinion and on improving care. Comment The patients who attend Camp Bluebird seem satisfied with their physicians. These campers represent a subset of the population who have survived treatment, are currently healthy, are interested sharing their experiences, and have access to medical care. These biases certainly make it impossible to generalize their attitudes to the entire oncologic population, but we can obtain valuable information from administration of this questionnaire. Evaluation of individual questions within the PMH-PSQMD shows that these patients are satisfied with their physicians’ competence and communication. They believe that physicians should try to gain a clearer understanding the pain and fatigue experienced by patients. Not surprisingly, patients would like more time with their physicians. In the open discussion forum, they were understanding of the time demands on physicians and the unpredictable nature of managing cancer patients in an outpatient setting, which often results in unforeseeable delays. These results are the most useful data from this study because it provides specific areas in which we can improve. Because the campers were given dedicated time to complete the questionnaire, as seen in the 96% response rate, selection bias that is often encountered in voluntary
surveys was overcome. Selection bias and recall bias do, however, apply to the 67% of patients returning the demographic questionnaire. We found that their overall satisfaction was not influenced by the degree of therapy intensity or time since diagnosis but may have an inverse association with financial burden. Patients may extrapolate expensive care to negative feelings toward their physician, or perhaps those with little satisfaction in their physicians may overestimate their financial burden. We cannot make any statements about causation from our data. Nevertheless, given the lack of correlation between satisfaction and disease severity, and the responses to questions noted in Table I, this cohort of patients appears to attribute the quality of their physician with the way in which their care is delivered, not the ultimate outcome of their disease. Although the low number of patients surveyed and biases present prohibit making statements applicable to the general oncology population, we believe the method could identify patient characteristics that are associated with dissatisfaction in a larger number of patients. The PMH-PSQ-MD can provide valuable information to the oncologist regarding the quality of care delivered, as perceived by the patient, and allow a focused effort on improving patient satisfaction. In a review of the literature, several studies evaluate patient preferences for treatment options, but few examine satisfaction with the physician. Knowing the needs of our patients would allow targeted improvement in delivery of care, and possibly improved patient outcomes. It is hoped that delivery of these responses to our own physicians, as we have done, will lead to improvements in delivery of care.
REFERENCES
1. Oberst MT. Methodology in behavioral and psychosocial cancer research: patients’ perceptions of care—measurement of quality and satisfaction. Cancer 1984;53(Suppl):2366-75. 2. Ware JE Jr, Davies AR. Behavioral consequences of consumer dissatisfaction with medical care. Eval Program Plann 1983;6:291-7. 3. Sinf J, Weiss B. Patient satisfaction with health care: intentions and change in plan. Eval Program Plann 1991;14:299-306. 4. Rubin HR, Gandek B, Rogers WH, Kosinski M, McHorney CA, Ware JE Jr. Patients’ ratings of outpatient visits in different practice settings: results from the Medical Outcomes Study. JAMA 1993;270:835-40. 5. Hopton JL, Howie JG, Porter AM. The need for another look at the patient in general practice satisfaction surveys. Fam Pract 1993;10:82-7. 6. Loblaw DA, Bezjak A, Bunston T. Development and testing of a visit-specific patient satisfaction questionnaire: the Princess Margaret Hospital Satisfaction With Doctor Questionnaire. J Clin Oncol 1999;17:1931-8.