Associations Among Cancer Survivorship Discussions, Patient and ...

VOLUME

28

䡠

NUMBER

15

䡠

MAY

20

2010

JOURNAL OF CLINICAL ONCOLOGY

O R I G I N A L

R E P O R T

Associations Among Cancer Survivorship Discussions, Patient and Physician Expectations, and Receipt of Follow-Up Care Winson Y. Cheung, Bridget A. Neville, and Craig C. Earle From the British Columbia Cancer Agency, Vancouver, British Columbia, Canada; Center for Outcomes and Policy Research, Dana-Farber Cancer Institute, Boston, MA; Institute for Clinical Evaluative Sciences, Toronto, Ontario, Canada. Submitted October 6, 2009; accepted February 24, 2010; published online ahead of print at www.jco.org on April 20, 2010. Supported by funds from the Perini Family Cancer Survivors’ Center, an award from the Canadian Association of Medical Oncologists, and a grant from the National Cancer Institute of Canada. Authors’ disclosures of potential conflicts of interest and author contributions are found at the end of this article. Corresponding author: Craig C. Earle, MD, MSc, FRCPC, Institute for Clinical Evaluative Sciences, 2075 Bayview Ave, Rm G-106, Toronto, Ontario M4N 3M5, Canada; e-mail: craig.earle@ices .on.ca. © 2010 by American Society of Clinical Oncology 0732-183X/10/2815-2577/$20.00 DOI: 10.1200/JCO.2009.26.4549

A

B

S

T

R

A

C

T

Purpose To explore the associations among cancer survivorship discussions, patient-physician expectations, and receipt of follow-up care in cancer survivors. Patients and Methods We surveyed cancer survivors about various aspects of their care, including expectations of their providers’ roles, whether discussions with a physician had occurred, and self-reported patterns of follow-up. Primary care providers (PCPs) and oncologists were also surveyed for their own perceived roles. We developed a scoring system to evaluate the level of agreement in expectations between patients and physicians and between PCPs and oncologists (where 0 ⫽ most discordant and 4 ⫽ most concordant). Regression and stratified analyses were conducted to examine the relationships among expectations, discussions, and follow-up. Results In total, 535 patients (54%) and 378 physicians (62%) responded. Survivorship care expectations were most discrepant between PCPs and oncologists (mean score, 1.78), moderate between patients and oncologists (mean score, 1.97), and most similar between patients and PCPs (mean score, 2.82). Having a conversation specifically about cancer follow-up was associated with better concordance between patients and oncologists, but not for patients and their PCPs or between physicians. Better concordance in patient-oncologist expectations also correlated with greater odds of receiving certain aspects of follow-up care, such as influenza vaccinations and physical examinations, but only if a discussion about cancer follow-up had occurred. Conclusion A discussion about cancer follow-up may affect survivorship care through its primary influence on patient-oncologist expectations. Further work is required to clarify the aspects of survivorship discussions that are important for optimal cancer survivorship care planning. J Clin Oncol 28:2577-2583. © 2010 by American Society of Clinical Oncology

INTRODUCTION

Advances in early detection and treatment strategies for various cancers, along with the aging general population, suggest that the number of cancer survivors will continue to increase steadily.1-3 Survivors face numerous difficulties, including psychosocial distress and physical ailments that may last for many years after diagnosis and treatment,4-6 highlighting the importance of developing survivorship care programs that can effectively address their short- and long-term needs. We previously reported that expectations for cancer survivorship care were highly discrepant between patients and their physicians and between primary care providers (PCPs) and oncologists, underscoring the potential complexities of delivering consistent survivorship care.7 In-

creasing time constraints among physicians and persistent disparities in health care access among patients may present additional challenges to providing comprehensive care for survivors. While this knowledge has strengthened our understanding of the different needs of survivors and improved our awareness of potential health care system barriers, few studies have formally examined possible solutions to these problems or evaluated interventions to optimize care.8-11 Communication between healthcare providers and patients is a cornerstone for fostering healthy patient-physician rapport, building mutual trust and respect, and establishing common goals and expectations for medical care. For early-stage cancers, discussions between patients and physicians have been found to be highly correlated with © 2010 by American Society of Clinical Oncology

Information downloaded from jco.ascopubs.org and provided by at VAUTRIN CENTRE on March 29, 2011 from Copyright © 2010 American Society 193.55.227.254 of Clinical Oncology. All rights reserved.

2577

Cheung, Neville, and Earle

patients’ satisfaction level, treatment choices, and overall quality of care.12,13 Likewise, in the palliative care setting, end-of-life discussions have been shown to be associated with earlier hospice referrals, lower health care costs, and more appropriate medical care near death.14,15 To date, research has yet to explore the impact of similar discussions on the expectations for and quality of cancer survivorship care. This type of information, however, is valuable and can help to inform survivors and their physicians and contribute to improving cancer survivorship care for the future. We conducted self-administered surveys to assess patient and physician expectations for cancer survivorship care. In this analysis of the survey, we explored the associations among cancer survivorship discussions, patient-physician expectations, and the receipt of follow-up care. Because discussions may offer survivors and their physicians the opportunity to better define their goals for the medical care they should receive, we hypothesized that expectations may be more concordant between survivors and their physicians and between physicians themselves and that follow-up care may be more appropriate when a patient-physician conversation surrounding cancer survivorship has occurred.

PATIENTS AND METHODS Study Overview This study was conducted between May 2006 and August 2007 after receiving approval from the institutional review board at Dana-Farber Cancer Institute in Boston, MA. Details of the survey design have been described previously.7 Briefly, two separate, complementary, self-administered questionnaires were developed using the modified Delphi approach; one was designed for patients and the other for physicians. The questionnaires were primarily devised to compare expectations for cancer survivorship care between patients and their physicians and between PCPs and oncologists and to determine patients’ self-reported patterns of cancer follow-up care. The patient survey was distributed to adult, English-speaking cancer survivors who were identified as having at least one component of their cancer management occur at Dana-Farber/Brigham and Women’s Cancer Center in Boston, MA, still alive at two or more years after their original cancer diagnosis, and currently cancer free and not undergoing any active cancer treatments. The survey asked cancer survivors to identify a PCP and an oncologist whom they viewed as being most responsible for the ongoing care of their cancer. The physician version of the survey was subsequently sent to these providers. All eligible patients and physicians were mailed a package containing a letter that introduced the study and invited their voluntary participation, a copy of the patient or physician questionnaire, a small monetary incentive of US$2.00, and a postage-paid envelope for returning the completed survey. A consent form was included in the patient packages, which was to be completed by patients and then returned with the questionnaire, while the receipt of a completed physician survey implied informed consent. Patient and Physician Questionnaires In addition to baseline demographics and cancer-related clinical information, the patient questionnaire asked respondents to indicate the level of responsibility that their PCP or oncologist should take with regard to four important survivorship domains: (1) follow-up of their primary cancer, (2) screening of other cancers, (3) general preventive health, and (4) management of other comorbidities. Physicians were subsequently asked in their version of the survey to indicate their own perceived roles in the same four key survivorship areas. Responses to these questions on both the patient and physician questionnaires were evaluated on a 5-point Likert scale that reflected the perceived degree of physician responsibility (“none,” “a little,” “some,” “a lot,” and “full”). In the patient survey only, participants were also asked whether a discussion about cancer follow-up (eg, “Did you and your doctor ever have a 2578

© 2010 by American Society of Clinical Oncology

discussion about who would follow you for your most recent cancer?”) or general medical care (“Did you and your doctor ever have a discussion about who would handle your other medical problems besides your cancer, such as diabetes, high blood pressure, or arthritis?”) had occurred with a physician. Furthermore, we inquired about the pattern of follow-up care that patients received in the past 12 months, including the receipt of influenza vaccinations, physical examinations, routine blood tests, and routine imaging studies. Statistical Considerations Descriptive statistics were used to summarize baseline demographic data and clinical characteristics of respondents. Surveys from patients were matched with those of their self-identified PCPs and oncologists to form patient-PCP and patient-oncologist pairs, respectively. Patients without a corresponding physician survey were not included in the matched analyses. Similarly, among patients with both a PCP and an oncologist survey, responses from PCPs were matched with those of the corresponding oncologists to form PCP-oncologist pairs. Concordance/discordance in expectations for the four survivorship areas was assessed for each group of matched survey pairs. For patient-PCP and patient-oncologist pairs, concordance was defined as complete agreement in Likert response categories, whereas discordance was signified by any disagreement in response categories between respondents of a matched survey pair. For PCP-oncologist pairs, however, responses that represented a potential duplication of medical services (ie, both physicians indicating “substantial” responsibility for a task) or omission of medical services (ie, both physicians claiming “minimal” responsibility for a task) were considered to be discordant. If both physicians responded that they had “some” degree of responsibility, this was not considered discordant since survivorship care tasks can be shared. For this analysis specifically, we developed a composite scoring system that would allow for an overall assessment of cancer survivorship care expectations. For each of the four survivorship domains, a score of 0 was assigned when expectations were discordant and a score of 1 was given when expectations were concordant. A composite concordance score (CCS) was derived by summing the individual scores from each of the four survivorship areas, thus yielding a possible total CCS that ranged from a minimum of 0 (most discordant) to a maximum of 4 (most concordant). Considering the narrow range of the CCS, we conducted two-sample t tests to compare means in order to assess (1) the relationship between patient-physician discussion status (occurred v did not occur) and CCS and (2) the association between CCS and receipt of follow-up care. Multivariable logistic regression models, with CCS as the dependent variable, were subsequently constructed, which included all available patient factors and physician characteristics, regardless of significance on univariate testing. We developed additional multivariable logistic models, with receipt of follow-up care (yes/no) as the dependent variable, that were stratified by patient-physician discussion status to determine whether the effect of CCS on receipt of follow-up care was modified by discussions about cancer survivorship. All tests were two-sided and a P value ⬍ .05 was considered statistically significant. SAS 9.1.3 (SAS Institute, Cary, NC) was used for all statistical analyses.

RESULTS

Characteristics of Respondents Table 1 summarizes the baseline demographics and clinical characteristics of the cancer survivor and physician cohorts. In total, 535 patients and 378 physicians completed the survey (response rates were 54% and 62%, respectively). Among patients, 104 (19%) indicated they were either currently not cancer free or still receiving active cancer treatments. Therefore, these cases were excluded, leaving 431 patient surveys (81%) eligible for analysis. Within this patient cohort, females, breast cancer survivors, white patients, high-income earners, and those who were married, obtained some postsecondary level education, and reported good health were more highly represented. For physicians, 255 (67%) were PCPs and 123 (33%) were oncologists. JOURNAL OF CLINICAL ONCOLOGY


Role of Cancer Survivorship Discussions

Table 1. Baseline Clinical Characteristics of Patient and Physician Respondents and Self-Reported Patterns of Cancer Survivorship Care Survey Respondents Characteristic Patient variables (n ⫽ 431) Age, years Median Range Sex Male Female Cancer site Breast Lung Gastrointestinal Genitourinary Gynecologic Hematologic Head and neck Other Unreported Marital status Single Married Separated/divorced Education Less than 8th grade Some high school High school graduation Some university University graduation Graduate school Unreported Race Black Asian White Multiracial Other Income, $ ⬍ 20,000 20,000-39,000 40,000-59,000 60,000-79,000 ⬎ 80,000 Unreported Ever smoker Yes No Unreported General health Poor Fair Good Very good Excellent Unreported (continued in next column)

www.jco.org

No.

Table 1. Baseline Clinical Characteristics of Patient and Physician Respondents and Self-Reported Patterns of Cancer Survivorship Care (continued) Survey Respondents

% Characteristic

57 16-91 117 314

27 73

223 17 28 37 18 40 31 16 21

52 4 6 9 4 9 7 4 5

46 320 65

11 74 15

4 11 60 82 106 166 2

1 3 14 19 25 39 0.5

10 7 406 2 6

2 2 94 0.5 1

23 39 41 49 199 80

6 9 10 12 48 15

200 229 2

47 52 1

5 24 112 187 99 4

1 6 26 44 23 1

Physician variables (n ⫽ 378) Age, years Median Range Sex Male Female Practice Academic Community Specialty Primary care Oncology Measures of cancer survivorship care (n ⫽ 431) Discussion about cancer follow-upⴱ With oncologist With PCP Both Neither Unreported Discussion about general medical careⴱ With oncologist With PCP Both Neither Unreported Flu vaccination† Yes No Unreported Routine physical examinations† Yes No Routine blood tests† Yes No Routine imaging studies† Yes No

No.

%

50 31-79 239 139

63 37

156 222

41 59

255 123

67 33

135 15 145 74 62

31 4 34 17 14

33 88 101 145 64

8 20 23 34 15

288 142 1

33 67 0.5

382 49

89 11

258 173

60 40

282 149

65 35

Abbreviation: PCP, primary care provider. ⴱ Discussion between patient and physician. †In the preceding year.

Among them, there were more community than academic practitioners. Overall, there were 409 matched patient-oncologist survey pairs, 233 patient-PCP pairs, and 232 PCP-oncologist pairs. Nonrespondents who formally opted out of survey participation did not differ significantly from respondents with respect to all demographic features. Survivorship Care Patterns The mean CCS between patients and PCPs and between patients and oncologists were 2.82 and 1.97, respectively, consistent with our prior report showing that expectations of patient-PCP pairs for each of the four key survivorship care domains were generally more concordant than those of patient-oncologist pairs.7 The observed agreement © 2010 by American Society of Clinical Oncology


2579


Table 2. Association Between Mean Concordance Scores and Survivorship Care Discussions Discussion About Cancer Follow-Up Occurred

Discussion About General Medical Care

Did Not Occur

Occurred

Did Not Occur

Survey Pair

Mean

SD

Mean

SD

Pⴱ

Mean

SD

Mean

SD

Pⴱ

Patient/oncologist (n ⫽ 409) Patient/PCP (n ⫽ 233) PCP/oncologist (n ⫽ 232)

2.05 2.81 1.71

1.01 0.92 1.12

1.78 2.84 1.93

0.94 1.13 1.08

.01 .82 .17

1.99 2.77 1.73

0.99 1.07 1.08

1.95 2.88 1.84

1.00 0.90 1.14

.67 .40 .46

Abbreviations: SD, standard deviation; PCP, primary care physician. ⴱ P value based on two-sample t tests.

between PCPs and oncologists was lowest, with a mean CCS of only 1.78. Patients’ self-reported patterns of cancer survivorship care are highlighted in Table 1. In total, 17% and 34% indicated that they failed to talk with at least one of their doctors about their cancer follow-up and general medical care, respectively. For the majority of patients who described having a survivorship discussion, cancer follow-up conversations occurred more frequently with oncologists while conversations about general medical care tended to involve PCPs. Most patients reported undergoing routine physical examinations, blood tests, and imaging studies in the past 12 months. However, most did not report receiving an influenza vaccination within the same time period. The mean number of visits to PCPs was 2.01 (standard deviation, 1.53) and to oncologists was 2.43 (standard deviation, 1.49) during the preceding year. Discussion and Follow-Up and Their Association With Survivorship Expectations Bivariate analyses of the relationships between survivorship discussions and concordance of expectations and the association between these expectations and follow-up care are described in Tables 2 and 3, respectively. Compared with no discussion, having a conversation about cancer follow-up was associated with a statistically significant higher mean CCS between patients and oncologists (2.05 v 1.78; P ⫽ .01). No additional relationships between discussion of different aspects of care and CCS were significant. In general, a higher CCS correlated with receipt of follow-up care. This was seen for patientoncologist expectations where better concordance was significantly associated with receiving influenza vaccinations (P ⫽ .01), physical examinations (P ⫽ .01), and imaging studies (P ⫽ .03). Of note, a higher CCS between PCPs and oncologists was associated with the lack of routine blood tests (P ⫽ .01). Importantly, we did not observe any clear, consistent associations to support the notion that discussions were associated with follow-up care.

Our multivariable analysis continued to demonstrate similar relationships, which are illustrated in Tables 4 and 5. Individuals reporting a discussion about cancer follow-up showed increasing odds of concordant patient-oncologist expectations. Likewise, a high CCS between patients and oncologists remained predictive of influenza vaccinations and routine physical examinations. We also explored possible patient and physician determinants for the presence of a survivorship discussion as well as for the receipt of follow-up care. While there was some suggestion that patients who were older, nonwhite, and poorer were less likely to have a survivorship discussion and less inclined to receive follow-up care, none achieved consistent statistical significance. In addition, we examined concordance of expectations for each of the four survivorship domains separately and their individual relationships with discussion and follow-up, but we were unable to identify any clear, consistent correlations. Finally, in an exploratory analysis of the patient-oncologist pairs only (Table 6), we stratified our model by discussion status. It appears that the impact of concordant patient-oncologist expectations on influenza vaccinations and physical examinations was observed only for those in whom a discussion about cancer follow-up had occurred. DISCUSSION

It is estimated that there are over 10 million cancer survivors in the United States, and this segment of the population is expected to continue to grow.16 A landmark report published by the Institute of Medicine17 highlighted the diverse array of physical and psychosocial issues associated with cancer survivorship and underscored its importance as a research priority. However, there are concerns that many of these survivorship issues are frequently overlooked by the health care system.18,19 We previously reported that expectations for survivorship care differed between patients and their physicians, and that a lack of

Table 3. Association Between Mean Concordance Scores and Receipt of Follow-Up Care Flu Vaccination

Physical Examination

Laboratory Tests

Imaging Studies

Survey Pair

Yes

No

Pⴱ

Yes

No

Pⴱ

Yes

No

Pⴱ

Yes

No

Pⴱ

Patient/oncologist (n ⫽ 409) Patient/PCP (n ⫽ 233) PCP/oncologist (n ⫽ 232)

2.14 2.83 1.79

1.89 2.79 1.75

.01 .78 .84

2.02 2.83 1.80

1.54 2.69 1.53

.01 .52 .30

1.99 2.76 1.52

1.94 2.91 2.13

.61 .25 .01

2.05 2.86 1.94

1.82 2.75 1.70

.03 .40 .11

Abbreviation: PCP, primary care physician. ⴱ P value based on two-sample t tests.

2580


JOURNAL OF CLINICAL ONCOLOGY



Table 4. Multivariable Analysis of CCS and Survivorship Care Discussions Discussion About Cancer Follow-Up CCS Patient/oncologist 0 1 2 3 4 Patient/PCP 0 1 2 3 4 PCP/oncologist 0 1 2 3 4

OR

95% CI

Pⴱ

1.00 1.40 1.95 2.73 3.81

1.04 to 1.88 1.08 to 3.52 1.13 to 6.60 1.18 to 12.4

.02

1.00 0.96 0.92 0.89 0.85

0.64 to 1.28 0.41 to 2.07 0.27 to 2.98 0.17 to 4.29

.84

1.00 0.87 0.75 0.65 0.56

0.66 to 1.13 0.44 to 1.28 0.29 to 1.45 0.19 to 1.64

.29

Discussion About General Medical Care OR

95% CI

Pⴱ

1.00 1.09 1.20 1.31 1.43

0.85 to 1.41 0.72 to 1.98 0.61 to 2.80 0.52 to 3.94

.49

1.00 0.87 0.76 0.67 0.58

0.61 to 1.25 0.37 to 1.56 0.23 to 1.95 0.14 to 2.44

.46

1.00 0.93 0.87 0.81 0.75

0.73 to 1.19 0.53 to 1.42 0.38 to 1.70 0.28 to 2.02

.57

Abbreviations: CCS, composite concordance score; OR, odds ratio; PCP, primary care physician. ⴱ P value based on multivariable logistic regression, adjusted for patient and physician characteristics listed in Table 1.

clarity surrounding their respective roles may contribute to suboptimal levels of care.7 A few studies have described interventions to improve future survivorship care planning, but none have explored the role of patient-physician communication.8-11 We analyzed surveys from cancer survivors and their providers and found that patientphysician discussions were associated with greater agreement in some

survivorship care expectations and that concordance in these expectations was in turn correlated with a higher propensity for certain aspects of follow-up. While we originally hypothesized that patients whose views on survivorship care are best aligned with those of their physicians are more likely to receive appropriate care, our stratified model confirmed this association to be true only for patients who reported having a conversation about cancer survivorship with their physician. In the absence of such a discussion, concordance in expectations failed to correlate with follow-up. In fact, there was a nonsignificant trend for possibly less care (eg, fewer influenza vaccinations). One interpretation of this finding is that similarities in patient-physician expectations by themselves are insufficient for ensuring survivorship care. Our multivariable and stratified analyses in Tables 5 and 6, respectively, suggest that a patient-physician discussion may contribute to survivorship care planning by acting indirectly through its primary influence on patient-physician expectations. Discussions about cancer follow-up appear to modify the effect of concordant expectations on survivorship care, which is consistent with prior reports demonstrating a positive impact of patient-physician communication on clinical decision making and coordination of care across various clinical settings in oncology.12-15 Importantly, discussions mainly posed a positive effect on patient-oncologist expectations and not on patient-PCP or PCPoncologist expectations. The mean CCS between patients and their oncologists was moderate when compared with those of the other matched pairs, proposing that relationships in which expectations were already mostly discordant or mostly concordant may be less susceptible to the effect of a patient-physician discussion. Another reason for this difference is that survivors may come to view their oncologists as their primary provider of healthcare and medical advice, considering the unique patient-physician rapport that often

Table 5. Multivariable Analysis of CCS and Likelihood of Receiving Survivorship Care Flu Vaccination CCS

OR

Patient/oncologist 0 1 2 3 4 Patient/PCP 0 1 2 3 4 PCP/oncologist 0 1 2 3 4

95% CI

Physical Examination P

ⴱ

OR

95% CI

.01 1.00 1.38 1.90 2.62 3.61

0.81 to 1.39 0.66 to 1.93 0.53 to 2.68 0.43 to 3.73

P

0.76 to 1.86 0.58 to 3.46 0.44 to 6.43 0.33 to 11.9

OR

95% CI

0.92 to 1.55 0.95 to 2.39 0.97 to 3.71 0.99 to 5.73

.49

.58 1.00 0.89 0.79 0.70 0.62

0.77 to 1.75 0.59 to 3.06 0.45 to 5.36 0.34 to 9.39

0.58 to 1.36 0.34 to 1.84 0.20 to 2.49 0.11 to 3.37

⬍ .01 1.00 0.57 0.33 0.19 0.11

0.43 to 0.76 0.19 to 0.57 0.08 to 0.43 0.04 to 0.33

Pⴱ .13

1.00 1.26 1.58 1.99 2.50

0.82 to 1.49 0.67 to 2.23 0.55 to 3.32 0.45 to 4.96

.45 1.00 1.19 1.41 1.68 2.00

Imaging Studies ⴱ

.52

1.00 1.16 1.34 1.55 1.79

0.58 to 1.73 0.34 to 2.98 0.20 to 5.14 0.12 to 8.87

.66 1.00 1.06 1.13 1.20 1.27

95% CI

.68 1.00 1.15 1.33 1.53 1.76

0.82 to 1.92 0.67 to 3.69 0.55 to 7.08 0.45 to 13.6

OR 1.00 1.11 1.22 1.35 1.49

1.17 to 2.37 1.36 to 5.62 1.59 to 13.3 1.85 to 31.6

.29 1.00 1.26 1.58 1.98 2.49

P

.03 1.00 1.66 2.77 4.60 7.65

1.03 to 1.84 1.07 to 3.39 1.10 to 6.24 1.14 to 11.5

Laboratory Tests ⴱ

.14 1.00 0.82 0.67 0.55 0.45

0.63 to 1.06 0.40 to 1.13 0.25 to 1.21 0.16 to 1.28

Abbreviations: CCS, composite concordance score; OR, odds ratio; PCP, primary care physician. ⴱ P value based on multivariable logistic regression, adjusted for patient and physician characteristics listed in Table 1.

www.jco.org



2581


Table 6. Effect of Discussion About Follow-Up on the Association Between CCS and Survivorship Care Flu Vaccination Occurred CCS

OR

95% CI

Physical Examination

Did Not Occur Pⴱ

OR

95% CI

Occurred Pⴱ

OR

95% CI

Laboratory Tests

Did Not Occur P ⴱ OR

Occurred P ⴱ OR

95% CI

95% CI

Imaging Studies


95% CI

Occurred P ⴱ OR

95% CI


95% CI

Pⴱ

Patient/ ⬍ .01

oncologist

.86

.02

.11

.90

.33

.14

.13

0

1.0

1.0

1.0

1.0

1.0

1.0

1.0

1.0

1

1.68 1.16 to 1.91

0.97 0.65 to 1.43

1.86 1.11 to 3.09

1.49 0.91 to 2.44

0.98 0.68 to 1.42

1.21 0.83 to 1.76

1.20 0.94 to 1.54

1.35 0.92 to 2.00

2

2.83 1.35 to 3.66

0.93 0.43 to 2.05

3.44 1.24 to 9.57

2.23 0.83 to 5.97

0.95 0.46 to 2.00

1.45 0.68 to 3.09

1.45 0.88 to 2.37

1.83 0.84 to 4.01

3

4.77 1.57 to 7.00

0.90 0.28 to 2.93

6.39 1.38 to 29.6

3.33 0.76 to 14.6

0.93 0.31 to 2.83

1.75 0.57 to 5.43

1.74 0.83 to 3.66

2.48 0.77 to 8.02

4

8.03 1.82 to 13.4

0.87 0.18 to 4.19

4.98 0.70 to 35.6

0.91 0.21 to 4.01

2.11 0.47 to 9.55

2.10 0.78 to 5.64

3.35 0.70 to 16.0

11.9

1.53 to 91.6

Abbreviations: CCS, composite concordance score; OR, odds ratio. ⴱ P value based on multivariable logistic regression, adjusted for patient and physician characteristics listed in Table 1.

develops during and after cancer treatment. In support of this hypothesis, we observed that it was only discussions about cancer follow-up, which occurred more frequently between patients and oncologists, that resulted in more congruent expectations and increased receipt of certain follow-up care. Because this was a cross-sectional survey, we were unable to delineate the directionality of events. Therefore, we cannot exclude the possibility that individuals with lower CCS at the outset were the ones in whom personal motivations were also the strongest to seek formal discussions with their doctors. We are also uncertain of the reasons why some aspects of follow-up were unaltered by concordant expectations or the presence of patientphysician discussions. The performance of blood tests and imaging studies typically requires more coordination between physicians and other healthcare workers (eg, laboratory and radiology technicians), and thus may be less amenable to behavioral interventions, such as survivorship discussions. Alternately, these aspects of care may necessitate multidisciplinary interventions in order to alter their patterns of use. There are several limitations to our study. First, we surveyed a geographically localized group of patients and physicians, and we analyzed only successfully returned questionnaires. Additionally, our sample consisted largely of white, well-educated, female breast cancer patients who may have had expectations for survivorship care that varied considerably from those in different demographic groups. For these reasons, our results may not be completely generalizable to other centers, nonrespondents, minorities, or survivors across different cancer types. Second, survivorship care patterns were based on patients’ self-report only and were susceptible to recall bias. These data were neither corroborated by their providers nor verified by medical record review. Furthermore, we did not collect information regarding the context in which follow-up interventions were given, so we could not determine whether procedures, such as vaccinations, were administered to survivors for the appropriate clinical indications. Moreover, other important survivorship care areas, such as psychosocial and vocational issues, were not considered in the CCS. Third, each physician completed the questionnaire only once, and their questions were not directed to the care of a particular patient, but to any of their patients who might be a cancer survivor. Therefore, it is possible that some responses might have been different if they were answering about a specific patient. Fourth, we dichotomized survivorship care discussion into whether one was reported to have 2582


occurred or not. Survivorship care discussions between patients and physicians are likely quite complex and dynamic, and future work should focus on aspects of these discussions, such as content, format, and delivery, which may impact their overall effectiveness on survivorship care planning. Finally, we focused our analyses on evaluating statistically significant differences in CCS between matched survey pairs without considering clinically meaningful differences. Symptom and quality-of-life analyses in oncology have typically used a minimum difference of 5% between matched pairs as a threshold for clinical meaningfulness.20-22 Because the CCS was created specifically for this analysis and it has not been used previously, whether this same definition can be applied to survivorship research warrants further study. In summary, cancer survivorship care is likely driven by multiple factors. Our study findings provide preliminary support to suggest that survivorship care discussions may increase some aspects of follow-up care through a primary influence on patient-oncologist expectations. Which elements within these discussions are most important and whether they foster truly appropriate follow-up care will require more in-depth evaluations in prospective, longitudinal studies. Further research should work toward developing models for facilitating patient-physician discussions that would best promote and optimize care for future cancer survivors.

AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST The author(s) indicated no potential conflicts of interest.

AUTHOR CONTRIBUTIONS Conception and design: Winson Y. Cheung, Craig C. Earle Financial support: Craig C. Earle Administrative support: Craig C. Earle Provision of study materials or patients: Craig C. Earle Collection and assembly of data: Winson Y. Cheung, Bridget A. Neville Data analysis and interpretation: Winson Y. Cheung, Bridget A. Neville Manuscript writing: Winson Y. Cheung, Craig C. Earle Final approval of manuscript: Winson Y. Cheung, Bridget A. Neville, Craig C. Earle JOURNAL OF CLINICAL ONCOLOGY



REFERENCES 1. Jemal A, Siegel R, Ward E, et al: Cancer statistics, 2009. CA Cancer J Clin 59:225-249, 2009 2. Jemal A, Thun MJ, Ries LA, et al: Annual report to the nation on the status of cancer, 19752005, featuring trends in lung cancer, tobacco use, and tobacco control. J Natl Cancer Inst 100:16721694, 2008 3. Kinsey T, Jemal A, Liff J, et al: Secular trends in mortality from common cancers in the United States by educational attainment, 1993-2001. J Natl Cancer Inst 100:1003-1012, 2008 4. Foster C, Wright D, Hill H, et al: Psychosocial implications of living 5 years or more following a cancer diagnosis: A systematic review of the research evidence. Eur J Cancer Care (Engl) 18:223247, 2009 5. Meyerowitz BE, Kurita K, D’Orazio LM: The psychological and emotional fallout of cancer and its treatment. Cancer J 14:410-413, 2008 6. Ferrell BR, Dow KH, Leigh S, et al: Quality of life in long-term cancer survivors. Oncol Nurs Forum 22:915-922, 1995 7. Cheung WY, Neville BA, Cameron DB, et al: Comparisons of patient and physician expectations for cancer survivorship care. J Clin Oncol 27:24892495, 2009 8. Morey MC, Snyder DC, Sloane R, et al: Effects of home-based diet and exercise on functional

outcomes among older, overweight long-term cancer survivors: RENEW—A randomized controlled trial. JAMA 301:1883-1891, 2009 9. Korstjens I, May AM, van Weert E, et al: Quality of life after self-management cancer rehabilitation: A randomized controlled trial comparing physical and cognitive-behavioral training versus physical training. Psychosom Med 70:422-429, 2008 10. Osborn RL, Demoncada AC, Feuerstein M: Psychosocial interventions for depression, anxiety, and quality of life in cancer survivors: Metaanalyses. Int J Psychiatry Med 36:13-34, 2006 11. Demark-Wahnefried W, Clipp EC, Morey MC, et al: Lifestyle intervention development study to improve physical function in older adults with cancer: Outcomes from Project LEAD. J Clin Oncol 24:3465-3473, 2006 12. Chen JY, Tao ML, Tisnado D, et al: Impact of physician-patient discussions on patient satisfaction. Med Care 46:1157-1162, 2008 13. Liang W, Burnett CB, Rowland JH, et al: Communication between physicians and older women with localized breast cancer: Implications for treatment and patient satisfaction. J Clin Oncol 20:1008-1016, 2002 14. Zhang B, Wright AA, Huskamp HA, et al: Health care costs in the last week of life: Associations with end-of-life conversations. Arch Intern Med 169:480-488, 2009 15. Wright AA, Zhang B, Ray A, et al: Associations between end-of-life discussions, patient mental

health, medical care near death, and caregiver bereavement adjustment. JAMA 300:1665-1673, 2008 16. National Cancer Institute. Cancer survivorship research: About cancer survivorship—US prevalence estimates. http://dccps.nci.nih.gov/ocs/aboutsurvivorship.html 17. Hewitt M, Greenfield S, Stovall E (eds): From Cancer Patient to Cancer Survivor: Lost in Transition. Washington, DC: Committee on Cancer Survivorship, Improving Care and Quality of Life, National Cancer Policy Board, Institute of Medicine, and National Research Council, National Academies Press, 2006 18. Earle CC, Neville BA: Under use of necessary care among cancer survivors. Cancer 101:17121719, 2004 19. Earle CC, Burstein HJ, Winer EP, et al: Quality of non-breast cancer health maintenance among elderly breast cancer survivors. J Clin Oncol 21: 1447-1451, 2003 20. Ringash J, O’Sullivan B, Bezjak A, et al: Interpreting clinically significant changes in patient reported outcomes. Cancer 110:196-202, 2007 21. Osoba D, Rodrigues G, Myles J, et al: Interpreting the significance of changes in health-related quality-of-life scores. J Clin Oncol 16:139-144, 1998 22. Barrett B, Brown D, Mundt M, et al: Sufficiently important difference: Expanding the framework of clinical significance. Med Decis Making 25:250-261, 2005

■ ■ ■

Let JOP Online Work for You Access all content from the first to the current issue of the Journal of Oncology Pracce (JOP); consult ASCO Clinical Guideline summaries; link to citaons from 1,100+ HighWirehosted journals. Receive regular e-mail alerts. Receive RSS feeds of the most-read arcles from the current issue and the last two issues. Get the latest cancer policy news. Access pracce resources and tools for enhancing quality of care. Find a job through the online career center. Download content to your PDA. Download figures into preformaed PowerPoint slides. Visit jop.ascopubs.org to see what JOP online has to offer.

www.jco.org



2583