Abstract
Purpose:
Little is known about factors associated with unmet needs for symptom management in patients with cancer.
Methods:
Patients with a new diagnosis of lung and colorectal cancer from the diverse nationally representative Cancer Care Outcomes Research and Surveillance cohort completed a survey approximately 5 months after diagnosis (N = 5,422). We estimated the prevalence of unmet need for symptom management, defined as patients who report that they wanted help for at least one common symptom (pain, fatigue, depression, nausea/vomiting, cough, dyspnea, diarrhea) during the 4 weeks before the survey but did not receive it. We identified patient factors associated with unmet need by using logistic regression with random effects to account for clustering within study sites.
Results:
Overall, 15% (791 of 5,422) of patients had at least one unmet need for symptom management. Adjusting for sociodemographic and clinical factors, African American race, being uninsured or poor, having early-stage lung cancer, and the presence of moderate to severe symptoms were associated with unmet need (all P < .05). Furthermore, patients who rated their physician’s communication score < 80 (on a 0 to 100 scale) had adjusted rates of an unmet need for symptom management that were more than twice as high as patients who rated their physicians with a perfect communication score (23.1% v 10.0%; P < .001).
Conclusion:
A significant minority of patients with newly diagnosed lung and colorectal cancer report unmet needs for symptom management. Interventions to improve symptom management should consider the importance of physician communication to the patient’s experience of disease.
INTRODUCTION
Symptoms are prevalent among patients with lung and colorectal cancer, and patients with early-stage cancer have similar rates of symptoms as those with late-stage cancer.1,6-7 Recently, expert panels from the National Quality Forum and ASCO emphasized the importance of patient-reported outcomes, specifically around symptom management in routine cancer care.2 An understanding of the prevalence and presence of symptoms is important to guiding and implementing care systems to manage symptoms in cancer populations.3-5 Research has suggested that needs assessments may be more likely to identify patients who need intervention compared with quality-of-life assessments alone.4,5 Although an understanding of the presence of symptoms is a necessary first step for improving patient care, strategies should also focus on identifying patients with unmet needs for symptom management rather than presence of symptoms alone. For example, when a patient reports severe pain in the past 4 weeks, this does not give any information with regard to the receipt (or success) of pain treatments during that time period. The extent to which patients with cancer and symptoms have their needs met in the United States is largely unknown, and such data are critical to inform efforts to improve symptom management in the diverse health care settings where patients with cancer receive care. In the present study, we assessed the prevalence and factors associated with unmet needs for the management of common cancer symptoms (pain, fatigue, depression, nausea/vomiting, cough, dyspnea, diarrhea) in a population-based sample of patients with a new diagnosis of lung and colorectal cancer.
METHODS
Study Population
We used data from the Cancer Care Outcomes Research and Surveillance Consortium (CanCORS), a demographically representative national cancer study and coordinated effort from 11 primary data collection sites that evaluated the care and experiences of approximately 10,000 adult patients with lung or colorectal cancer diagnosed between 2003 and 2005.1,8 Cancer diagnosis and American Joint Committee on Cancer TNM staging system, 6th edition, stage were determined based on cancer registry data and medical records.9 Other data for this study were obtained from the CanCORS baseline patient survey conducted approximately 3 to 6 months after diagnosis. Patients were considered to have late-stage disease if they had stage IV colorectal or stage IIIb or IV lung cancer. Detailed information about study design, procedures, and cohort has been published previously.1,10,11 This study focused on the 5,422 patients who were alive and able to complete sections 8 and 9 of the CanCORS baseline survey, which addressed symptom prevalence and management. The human subjects committees at all participating institutions approved the study.
Survey Instrument
Interviewers asked patients about the presence and severity of symptoms, including pain, fatigue, depression, nausea/vomiting, cough, dyspnea, and diarrhea, during the 4 weeks before the survey by using validated tools; the survey instrument is available on the CanCORS Web site.12 As previously reported, overall symptom prevalence was estimated by using the purposefully low threshold of any report of a symptom during the 4 weeks before the survey.1 For each symptom that a patient reported, severity was assessed with validated scales, including the Brief Pain Inventory for pain; the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire for nausea/vomiting, cough, dyspnea, and diarrhea; the Short Form 36 vitality scale for fatigue; and the Center for Epidemiologic Studies Depression Scale 8 for depression. Details about criteria used to classify symptoms as moderate to severe have been previously reported.1 Patients were asked: During the last 4 weeks, did you want help for your symptom? If the answer was yes, they were also asked: During the last 4 weeks, did you get the help you wanted from the doctor? Patients who responded no to the latter question were identified as having an unmet need for that symptom. Our conceptual model for this analysis was informed by prior research.13-15 We selected independent patient sociodemographic, clinical, and care-related factors that are likely to influence symptom management for patients with cancer. Demographics (ie, age, sex, race, education), insurance status, wealth, marital status, comorbidity count (adapted from the self-administered Charlson comorbidity index and the comorbidity questions from the Prostate Cancer Outcomes Study), whether hospitalized in the year before diagnosis, enrollment in a clinical trial, length of relationship with primary care physician, and recent cancer treatment (receipt of chemotherapy, radiation, or surgery within the past 6 weeks) were also collected with the survey instrument.11,16,17 Wealth was assessed by asking: If you lost all of your current sources of income and had to live off of your savings, how long could you continue to live at your current address and standard of living? Preferred role in decisions about treatment was assessed by asking: Which statement best describes the role you would prefer to play when decisions about treatment for your cancer are made?18 Preferences with regard to care at the end of life were assessed by asking patients whether they preferred treatment that extends life as much as possible even if it means having more pain and discomfort or whether they would want treatment that focuses on relieving pain and discomfort as much as possible, even if it means not living as long. These preferences were assessed to understand how patients rated importance of symptom management in relation to other goals.
Patient reports on physician communication were calculated from five items that were based on the Consumer Assessment of Healthcare Providers and Systems family of surveys on a scale of 0 to 100.19 Responses were categorized as in a prior CanCORS analysis (0 to 79, 80 to 99, 100).13 The five items queried patients about whether their physician listened carefully, explained things in a way that was understood, gave as much information as was desired about cancer treatments, encouraged cancer-related questions, and treated patients with courtesy and respect.20 We hypothesized that this communication measure is an important factor in ensuring high-quality symptom management for patients with cancer.
Statistical Analysis
Bivariable analyses compared the prevalence of unmet need by symptom across cancer type and stage by using χ2 tests (Appendix Table A1, online only). Logistic regression with a random intercept included for each of 11 primary data collection sites was used to study patient factors associated with having at least one unmet need for symptom management on the basis of a prespecified model and included sociodemographic information and clinical information, including comorbidities, length of relationship with primary care physician, treatments received, stage and type of cancer, and the presence of moderate to severe symptoms.21 Symptoms that were only asked of specific cancer types (cough and dyspnea for lung cancer and diarrhea for colorectal cancer) were included in a variable called disease-specific symptom for purposes of the multivariable model. All variables had no or < 1% missing data except for the wealth (16% missing) and baseline care preference variables (14% missing). We used multiple imputation to handle missing data for the independent variables in the multivariable model.22 We individually tested interactions based on hypotheses about potential effect modification. (A list of tested interactions can be found in the Appendix.) One statistically significant interaction effect was observed for cancer type and stage and was included in the final model. From our final model we computed the predicted marginal probabilities for unmet needs for each level of the independent variables included in the model.23 For ease of interpretation, we used a categorical variable to describe the interaction of cancer type and stage (early-stage lung, late-stage lung, early-stage colorectal, late-stage colorectal). CanCORS data set version 1.16 was used for these analyses. Data management and descriptive analyses were performed with SAS Version 9.3 software (SAS Institute, Cary, NC). Modeling through multiply imputed data accounting for clustering effects at the study site level was conducted in Stata 12 with xtlogit (StataCorp, College Station, TX).
Sensitivity Analyses
For the subsample of the cohort who had medical record data (4,295 of 5,422 [79%]), we also tested whether specific visit types (medical oncology, radiation oncology, surgery, primary care) or the total number of outpatient visits in the 28 days before the interview were associated with unmet need for symptom management.
Although we were interested in reporting and describing patient factors associated with unmet needs for symptom management in the entire cohort of patients with cancer to inform health care systems, because only patients with symptoms could have an unmet need for symptom control, we repeated the analyses after limiting the cohort to patients who reported at least one symptom (5,067 of 5,422 [93.5%]).
RESULTS
Almost all (5,067 of 5,422 [93.5%]) the patients in the cohort had a least one common symptom present (detailed data previously reported).1 A majority (3,302 of 5,422 [61%]) wanted help for at least one of these reported symptoms (Appendix Table A2, online only). Among all patients, 15% (791 of 5,422) had at least one unmet need for symptom management in the 4 weeks before the interview, and 16% (791 of 5,067) who reported at least one symptom had at least one unmet need for symptom management. Unmet needs ranged from 1.4% for nausea/vomiting to 7.6% for fatigue. Fatigue, depression, cough, and dyspnea were among the most common symptoms to result in unmet need (Table 1). Among all patients with lung cancer, 19% had at least one unmet need, and among patients with colorectal cancer, 11% had at least one unmet need.
Table 1.
Prevalence of Unmet Need (did not get help) for Symptom Management in Early- and Late-Stage Lung and Colorectal Cancer (N = 5,422)
| At Least One Unmet Need for Symptoms, No. (%) | ||||
|---|---|---|---|---|
| Lung Cancer (n = 2,411) | Colorectal Cancer (n = 3,011) | |||
| Type of Unmet Need | Early Stage (n = 1,295) | Late Stage (n = 1,116) | Early Stage (n = 2,426) | Late Stage (n = 585) |
| Any unmet need | 258 (19.9)* | 189 (16.9)† | 270 (11.1)* | 74 (12.7)† |
| Pain | 49 (3.8) | 31 (2.8) | 74 (3.1) | 14 (2.4) |
| Fatigue | 114 (8.8)* | 101 (9.1) | 150 (6.2)* | 45 (7.7) |
| Depression | 65 (5.0)* | 42 (3.8) | 89 (3.7)* | 23 (3.9) |
| Nausea/vomiting | 22 (1.7) | 17 (1.5) | 33 (1.4) | 5 (0.9) |
| Cough | 72 (5.6) | 47 (4.2) | N/A | N/A |
| Dyspnea | 78 (6.0)‡ | 47 (4.2)‡ | N/A | N/A |
| Diarrhea | N/A | N/A | 36 (1.5) | 14 (2.4) |
Abbreviation: N/A, not applicable.
P < .01, comparison between early-stage lung and early-stage colorectal cancer.
P < .01, comparison between late-stage lung and late-stage colorectal cancer.
P < .01, comparison between early stage and late stage within cancer type.
In multivariable analysis that adjusted for all other factors, African American patients were more likely to have unmet needs for symptom management than their white counterparts (17.5% v 13.9%; P = .01; Table 2). Being uninsured compared with having private insurance also was associated with a higher likelihood of reporting unmet needs (19.2% v 13.7%; P = 0.02). The presence of moderate to severe pain (18.6% v 13.4%; P < .001), fatigue (21.4% v 10.0%; P < .001), depression (19.3% v 13.0%, P < .001), and disease-specific symptoms such as diarrhea, cough, and dyspnea (17.9% v 13.0%; P < .001) were all associated with a higher odds of reporting unmet needs compared with patients without these moderate to severe symptoms. However, the presence of moderate to severe nausea/vomiting compared with the presence of mild or no nausea/vomiting was not significantly associated with unmet need, although relatively few patients reported unmet needs for nausea/vomiting.
Table 2.
Predicted Marginal Probabilities for Unmet Needs for Common Symptoms in Patients With Lung and Colorectal Cancer (all patients included regardless of whether they reported symptoms; N = 5,422)
| Unmet Need for Any Symptom | ||||||
|---|---|---|---|---|---|---|
| Patient Characteristic | No. (%) | Coefficient (SE) | P | F Test P | OR (95% CI) | Predicted Marginal Probability (%) |
| Sex | .769 | |||||
| Male | 2,888 (53.3) | Reference | — | Reference | 14.3 | |
| Female | 2,534 (46.7) | 0.03 (0.09) | .769 | 1.03 (0.85 to 1.24) | 14.6 | |
| Age, years | .589 | |||||
| 21-59 | 1,840 (33.9) | 0.23 (0.19) | .234 | 1.25 (0.86 to 1.82) | 15.5 | |
| 60-69 | 1,584 (29.2) | 0.10 (0.17) | .551 | 1.11 (0.79 to 1.56) | 14.2 | |
| 70-79 | 1,424 (26.3) | 0.07 (0.17) | .699 | 1.07 (0.77 to 1.49) | 13.8 | |
| ≥ 80 | 574 (10.6) | Reference | — | Reference | 13.1 | |
| Race | .0314 | |||||
| White | 3,780 (69.7) | Reference | — | Reference | 13.9 | |
| Hispanic or Latino | 371 (6.8) | 0.25 (0.17) | .135 | 1.29 (0.92 to 1.79) | 16.6 | |
| African American | 726 (13.4) | 0.32 (0.13) | .010 | 1.38 (1.08 to 1.77) | 17.5 | |
| Other* | 545 (10.1) | −0.07 (0.15) | .658 | 0.94 (0.70 to 1.26) | 13.2 | |
| Education | .849 | |||||
| Less than high school | 919 (17.0) | −0.06 (0.15) | .692 | 0.94 (0.71 to 1.25) | 13.9 | |
| High school/some college | 3,153 (58.2) | 0.01 (0.11) | .934 | 1.01 (0.81 to 1.25) | 14.6 | |
| College or more | 1,343 (24.8) | Reference | — | Reference | 14.5 | |
| Insurance status at time of diagnosis | .143 | |||||
| VA | 634 (11.8) | 0.22 (0.22) | .304 | 1.25 (0.82 to 1.92) | 16.2 | |
| Private | 1,910 (35.5) | Reference | — | Reference | 13.7 | |
| Medicare + supplemental | 1,996 (37.1) | −0.01 (0.13) | .952 | 0.99 (0.76 to 1.29) | 13.6 | |
| Public | 622 (11.6) | 0.15 (0.15) | .334 | 1.16 (0.86 to 1.56) | 15.3 | |
| None | 214 (4.0) | 0.47 (0.20) | .018 | 1.60 (1.08 to 2.37) | 19.2 | |
| Wealth (how much money saved to live at current cost of living) | .239 | |||||
| < 1 month | 1,034 (22.7) | 0.24 (0.12) | .047 | 1.27 (1.00 to 1.60) | 15.6 | |
| 1-2 months | 568 (12.5) | 0.18 (0.15) | .244 | 1.19 (0.89 to 1.61) | 14.9 | |
| 3-6 months | 574 (12.6) | 0.18 (0.14) | .221 | 1.19 (0.90 to 1.58) | 14.9 | |
| 7-12 months | 359 (7.9) | 0.31 (0.17) | .072 | 1.37 (0.97 to 1.92) | 16.4 | |
| > 1 year | 2,019 (44.3) | Reference | — | Reference | 13.1 | |
| Marital status | .923 | |||||
| Not married | 2,259 (41.7) | −0.01 (0.09) | .923 | 0.99 (0.83 to 1.82) | 14.4 | |
| Married | 3,160 (58.3) | Reference | — | Reference | 14.5 | |
| Surgery in past 6 weeks | .336 | |||||
| Not received | 5,204 (96.0) | Reference | — | Reference | 14.6 | |
| Received | 218 (4.0) | −0.21 (0.22) | .336 | 0.81 (0.53 to 1.25) | 12.5 | |
| Radiation in past 6 weeks | .256 | |||||
| Not received | 4,808 (88.7) | Reference | — | Reference | 14.7 | |
| Received | 614 (11.3) | −0.15 (0.13) | .256 | 0.86 (0.67 to 1.11) | 13.2 | |
| Chemotherapy in past 6 weeks | .719 | |||||
| Not received | 3,085 (56.9) | Reference | — | Reference | 14.6 | |
| Received | 2,337 (43.1) | −0.03 (0.09) | .719 | 0.97 (0.80 to 1.16) | 14.3 | |
| Comorbidities | .885 | |||||
| 0-1 | 3,286 (60.7) | Reference | — | Reference | 14.5 | |
| ≥ 2 | 2,125 (39.3) | −0.01 (0.09) | .885 | 0.99 (0.82 to 1.18) | 14.4 | |
| In the year before diagnosis, hospitalized for any other medical condition | .908 | |||||
| No | 4,312 (79.9) | Reference | — | Reference | 14.5 | |
| Yes | 1,086 (20.1) | −0.01 (0.10) | .908 | 0.99 (0.81 to 1.21) | 14.4 | |
| Clinical trial | .230 | |||||
| No | 5,207 (96.0) | Reference | — | Reference | 14.6 | |
| Yes | 215 (4.0) | −0.28 (0.24) | .230 | 0.75 (0.48 to 1.20) | 11.8 | |
| Preferred role in decisions about treatment | .560 | |||||
| Patient controlled | 1,941 (36.1) | Reference | — | Reference | 13.8 | |
| Shared control | 3,115 (57.9) | 0.09 (0.09) | .296 | 1.10 (0.92 to 1.31) | 14.8 | |
| Physician controlled | 322 (6.0) | 0.11 (0.18) | .559 | 1.11 (0.78 to 1.59) | 15.0 | |
| Length of relationship with PCP | .262 | |||||
| No PCP/DK, < 5 years | 3,143 (58.1) | Reference | — | Reference | 14.9 | |
| Long term, > 5 years | 2,263 (41.9) | −0.10 (0.09) | .262 | 0.90 (0.75 to 1.08) | 13.8 | |
| Prefer treatment that | .635 | |||||
| Extends life as much as possible | 2,282 (48.7) | Reference | — | Reference | 14.7 | |
| Relieves pain or discomfort as much as possible | 2,401 (51.3) | −0.05 (0.09) | .635 | 0.95 (0.79 to 1.16) | 14.2 | |
| Moderate to severe symptoms | < .001 | |||||
| Pain | ||||||
| No | 4,607 (85.0) | Reference | — | Reference | 13.4 | |
| Yes | 815 (15.0) | 0.45 (0.11) | < .001 | 1.57 (1.28 to 1.93) | 18.6 | |
| Fatigue | < .001 | |||||
| No | 3,695 (68.1) | Reference | — | Reference | 10.0 | |
| Yes | 1,727 (31.9) | 0.98 (0.10) | < .001 | 2.66 (2.20 to 3.21) | 21.4 | |
| Depression | < .001 | |||||
| No | 4,598 (84.8) | Reference | — | Reference | 13.0 | |
| Yes | 824 (15.2) | 0.54 (0.10) | < .001 | 1.72 (1.40 to 2.10) | 19.3 | |
| Nausea/vomiting | .170 | |||||
| No | 5,097 (94.0) | Reference | — | Reference | 14.3 | |
| Yes | 325 (6.0) | 0.20 (0.15) | .170 | 1.23 (0.92 to 1.64) | 16.6 | |
| Cough, dyspnea, diarrhea | < .001 | |||||
| No | 4,164 (76.8) | Reference | — | Reference | 13.0 | |
| Yes | 1,258 (23.2) | 0.43 (0.12) | < .001 | 1.54 (1.23 to 1.94) | 17.9 | |
| Physician communication (higher scores indicate better communication) | < .001 | |||||
| 0-79 | 975 (18.0) | 1.12 (0.11) | < .001 | 3.06 (2.49 to 3.76) | 23.1 | |
| 80-99 | 1,447 (26.7) | 0.62 (0.10) | < .001 | 1.85 (1.52 to 2.26) | 16.2 | |
| 100 | 2,994 (55.3) | Reference | — | Reference | 10.0 | |
| Cancer stage | .918 | |||||
| Early | 3,721 (68.6) | Reference | — | — | — | |
| Late | 1,701 (31.4) | −0.02 (0.15) | .918 | — | — | |
| Cancer type | ||||||
| Lung | 2,411 (44.5) | 0.40 (0.13) | .002 | — | — | |
| Colorectal | 3,011 (55.5) | Reference | — | — | — | |
| Cancer stage/cancer type interaction | −0.42 (0.19) | .027 | .0275 | |||
| Early stage | ||||||
| Lung | 1,295 (23.9) | — | — | 1.49 (1.16-1.92) | 17.9 | |
| Colorectal | 2,426 (44.7) | — | Reference | 13.4 | ||
| Late stage | ||||||
| Lung | 1,116 (20.6) | — | — | 0.84 (0.55-1.27) | 13.1 | |
| Colorectal | 585 (10.8) | — | — | 0.85 (0.63-1.15) | 13.3 | |
Abbreviations: DK, don’t know; OR, odds ratio; PCP, primary care provider; SE, standard error; VA, Veterans Affairs.
Other race category includes Asian (n = 249), American Indian/Alaskan Native (n = 45), Native Hawaiian (n = 8), Pacific Islander (n = 15), more than one race (n = 163), and other (n = 65).
Patients who rated their physicians most highly (100 on a scale of 1 to 100) in quality of communication were less likely to report unmet needs for symptom management (adjusted proportion with unmet needs, 10.0%) compared with patients who rated their physicians lower (16.2% for ratings 80 to 99 [P < .001] and 23.0% for ratings < 80 [P < .001]).
The interaction of stage and type of cancer was statistically significant such that patients with early-stage lung cancer were more likely to report unmet needs for symptom management compared with patients with early-stage colorectal cancer, late-stage lung cancer, or late-stage colorectal cancer (17.9% v 13.4% v 13.1% v 13.3%; P = .03 for interaction). As aforementioned, no other interactions tested were statistically significant. In sensitivity analyses restricted to patients for whom we had medical record data, neither visits with specialty physicians (medical oncology, radiation oncology, surgery, or primary care) nor total number of outpatient visits in the 28 days before the interview was associated with an unmet need for symptom management. Furthermore, results were similar in analyses that restricted the cohort to the 5,067 who reported having at least one symptom (Appendix Table A3, online only).
DISCUSSION
Symptoms are prevalent among patients with cancer as shown previously in studies that used data from CanCORS and other sources.1,6,7 Untreated symptoms are likely to lead to poor quality of life and may lead to nonadherence to cancer treatment regimens. Many current measurement strategies focus on the assessment of symptom prevalence rather than on unmet needs.2,3,6 However, we found that although patients with lung and colorectal cancer in the current study reported a high prevalence of common symptoms, including moderate to severe symptoms, rates of unmet need for symptom management were relatively low. Nonetheless, 15% of the population and 16% of patients who reported any symptoms reported at least one unmet need for common symptom management during the 4-week period, a number that is not insignificant. A measure of patient-reported unmet need for symptom management may be more reflective of deficits in care and more actionable than symptom prevalence alone. Research has demonstrated that nurse and telephone-based interventions can lead to improved symptom control for patients with cancer.24-26 Cost-effectiveness of such interventions could be optimized if they were tailored to the patients with unmet needs.
The quality of physician communication as rated by patients was strongly associated with unmet need for symptom management. Patients who rated their physician’s communication most highly had adjusted rates of unmet needs for symptom management that were less than one half of those with the lowest ratings. Although the current findings do not allow for a demonstration of causality, the data suggest that communication skills, such as those captured in the modified Consumer Assessment of Healthcare Providers and Systems measure (empathy, listening, and attentiveness to patient needs) are associated with high-quality symptom management. These findings are consistent with previous reports that patients rate communication highly when physicians are responsive to their needs.27-31 The notable exception by Weeks et al13 showed that highly rated patient-physician communication scores are associated with patients who do not understand that their disease is incurable. This patient-reported quality metric and its divergent association with two important aspects of palliative care (symptom management and discussion of prognosis and goals of care) illustrate the need for a better understanding of the benefits and limitations of this measure for quality improvement efforts.
We also found that African-American patients were more likely than white patients to report not having their symptom management needs met, even after adjustment for financial barriers such as wealth and insurance status. This finding is consistent with a prior study that showed that ethnicity is associated with unmet needs for symptom management among patients with breast cancer.15 The authors suggested that this was most likely due to inadequate physician-patient communication related to cultural barriers or lack of cultural sensitivity. In the current study, we were able to adjust for patient-rated physician communication quality, but a difference in unmet needs for symptom management was evident, which suggests that reasons other than communication quality may contribute to this finding or that our measure of communication quality may not be sensitive to more-nuanced communication behaviors. A qualitative study by Song et al32 reported that communication behaviors of health care practitioners directly affect the cancer care experience of African-American patients. Further research is needed to assess whether attention to providing high-quality culturally sensitive health care may help to overcome communication barriers and unmet symptom needs.
Uninsured patients were also more likely to report unmet needs for symptom management, which suggests that access to health care may be a concern even among patients with lung or colorectal cancer. Insurance barriers may manifest in terms of limited access to particular treatments, such as newer, less burdensome chemotherapeutic agents; supportive medications, such as advanced medicines and/or delivery systems (eg, pain medication patch, pump), antiemetics, and antidepressants; and other forms of supportive care, such as mental health counseling, caregiving support, transportation, and postsurgical occupational or physical therapy. Lack of health insurance may also be associated with health literacy and awareness of and access to other cancer-related support services.33
Patients with early-stage lung cancer reported more unmet needs than patients with early- and late-stage colorectal cancer or late-stage lung cancer. This unexpected finding may reflect a different approach to symptom management in early-stage disease compared with late-stage disease and merits further study. Of note, there was more unmet need for disease-specific lung symptoms (cough and dyspnea) than for disease-specific colorectal symptoms. This is likely associated with the difficulty in treating lung cancer–specific symptoms (dyspnea, cough) compared with colorectal cancer–specific symptoms (diarrhea).34 Dyspnea also is known to be a symptom that has a large negative impact on quality of life.35-37 Patients with moderate to severe symptoms were also more likely to report an unmet need for symptom management, consistent with prior research.15
Although we know the rates of unmet needs based on self-report, we do not have descriptive data on what these unmet needs are. Some reports of unmet needs may be related to lack of clinician attention to assessment and treatment of symptoms; however, others may represent symptoms that have proven refractory to available treatments, which would suggest the need for additional research and development for novel therapies for symptom management. Another limitation is that this is a cross-sectional, observational study. Whether lack of symptom management occurs consistently for similar patients over time or whether deficits are intermittent is unknown. Furthermore, a patient was considered to have unmet need if he or she reported an unmet need for at least one symptom. The burden of unmet need may vary from patient to patient, but supportive care provided in the oncology setting should address all symptoms, and all needs would ideally be met, hence the choice of this outcome.
Additional limitations are that we could not include patients who were unable to respond to the detailed symptom questions. Therefore, we may have underestimated unmet need in later-stage disease. Finally, limitations in power for this analysis led us to combine several race categories (Asian, American Indian or Alaskan Native, Native Hawaiian and Pacific Islander, more than one race, and other) into one category. This topic merits future study.
In a large, nationally representative cohort of patients with incident lung and colorectal cancer, we studied symptom management as an important step toward understanding patient factors associated with unmet symptom needs. This work adds important data about unmet needs for the treatment of these symptoms in patients with two of the most prevalent cancers in the United States. The findings represent an important foundation upon which to build current quality measurement strategies as well as a greater understanding of areas to which novel supportive care interventions should be targeted.1,38-41
Acknowledgment
Supported by American Cancer Society grant number 119663-RSG-10-176-01-PCSM. The CanCORS Consortium was supported by grants from the National Cancer Institute (NCI) to the Statistical Coordinating Center at Dana-Farber Cancer Institute (U01 CA093344), the primary data collection and research centers at Harvard Medical School and Northern California Cancer Center (U01 CA093324); Dana-Farber Cancer Institute and Cancer Research Network (U01 CA093332); RAND Corporation and University of California, Los Angeles (U01 CA093348); University of Alabama at Birmingham (U01 CA093329); University of Iowa (U01 CA093339); and University of North Carolina (U01 CA093326). Also supported by a Department of Veterans Affairs grant to the Durham VA Medical Center (U01 CDA093344 [MOU] and HARQ 03-438MO-03).
A.M.W. was supported by National Institutes of Health (NIH) National Center for Advancing Translational Science UCLA Clinical and Translational Science Institute grant number (UL1 TR000124) and the NIH loan repayment program. N.L.K. was funded by NIH (R011CA164021-01A1 and K24CA181510). The content is solely the responsibility of the authors and does not necessarily represent the official views of the American Cancer Society, NCI, Patient-Centered Outcomes Research Institute, or NIH.
N.K.A.’s work on this project was completed while he was affiliated with NCI.
Presented at the Annual Assembly of the American Academy of Hospice and Palliative Medicine, Denver, CO, March 7-10, 2012; the ASCO Palliative Care Symposium, Boston, MA, October 24-25, 2014; and the Annual Assembly of the American Academy of Hospice and Palliative Medicine, Philadelphia, PA, February 25-28, 2015.
Appendix
Interactions Tested
Based on hypotheses about potential effect modification, the following interactions were individually tested: moderate to severe pain and depression, moderate to severe fatigue and depression, moderate to severe disease-specific symptom and depression, race and wealth, receipt of radiation treatment and stage, receipt of radiation treatment and receipt of surgery, receipt of radiation and receipt of chemotherapy, decision role and education, decision role and age, decision role and wealth, cancer type and stage, and cancer type and insurance.
Table A1.
Bivariable Analyses for Patient Characteristics by Unmet Need (N = 5,422)
| No Unmet Need | At Least One Unmet Need | |||
|---|---|---|---|---|
| No. | % | No. | % | |
| Sex* | ||||
| Male | 2,485 | 86.05 | 403 | 13.95 |
| Female | 2,146 | 84.69 | 388 | 15.31 |
| Age*, years | ||||
| 0-59 | 1,514 | 82.28 | 326 | 17.72 |
| 60-69 | 1,354 | 85.48 | 230 | 14.52 |
| 70-79 | 1,249 | 87.71 | 175 | 12.29 |
| ≥ 80 | 514 | 89.55 | 60 | 10.45 |
| Race* | ||||
| White | 3,264 | 86.35 | 516 | 13.65 |
| Hispanic or Latino | 299 | 80.59 | 72 | 19.41 |
| African American | 599 | 82.51 | 127 | 17.49 |
| Other | 469 | 86.06 | 76 | 13.94 |
| Education* | ||||
| Less than high school | 762 | 82.92 | 157 | 17.08 |
| High school/some college | 2,682 | 85.06 | 471 | 14.94 |
| College degree or more | 1,181 | 87.94 | 162 | 12.06 |
| Insurance* | ||||
| VA | 508 | 80.13 | 126 | 19.87 |
| Private | 1,646 | 86.18 | 264 | 13.82 |
| Medicare + supplemental | 1,771 | 88.73 | 225 | 11.27 |
| Public | 511 | 82.15 | 111 | 17.85 |
| None | 161 | 75.23 | 53 | 24.77 |
| If lost all income, how long could you continue to live at your current address and standard of living?* | ||||
| < 1 month | 820 | 79.30 | 214 | 20.70 |
| 1-2 months | 465 | 81.87 | 103 | 18.13 |
| 3-6 months | 486 | 84.67 | 88 | 15.33 |
| 7-12 months | 307 | 85.52 | 52 | 14.48 |
| > 1 year | 1,818 | 90.04 | 201 | 9.96 |
| Marital status* | ||||
| Married | 2,744 | 86.84 | 416 | 13.16 |
| Other | 1,884 | 83.40 | 375 | 16.60 |
| Surgery in past 6 weeks | ||||
| Not received | 4,443 | 85.38 | 761 | 14.62 |
| Received | 188 | 86.24 | 30 | 13.76 |
| Radiation in past 6 weeks* | ||||
| Not received | 4,131 | 85.92 | 677 | 14.08 |
| Received | 500 | 81.43 | 114 | 18.57 |
| Chemotherapy in past 6 weeks | ||||
| Not received | 2,656 | 86.09 | 429 | 13.91 |
| Received | 1,975 | 84.51 | 362 | 15.49 |
| No. of comorbidities (self-reported, ≥ 2)* | ||||
| 0-1 | 2,861 | 87.07 | 425 | 12.93 |
| ≥ 2 | 1,761 | 82.87 | 364 | 17.13 |
| In the year just before diagnosis, hospitalized for any other medical condition?* | ||||
| Yes | 900 | 82.87 | 186 | 17.13 |
| No | 3,710 | 86.04 | 602 | 13.96 |
| Clinical trial | ||||
| Yes | 190 | 88.37 | 25 | 11.63 |
| No | 4,441 | 85.29 | 766 | 14.71 |
| Which statement best describes the role you would prefer to play when decisions about treatment of your cancer are made? | ||||
| Patient controlled | 1,670 | 86.04 | 271 | 13.96 |
| Shared control | 2,655 | 85.23 | 460 | 14.77 |
| Physician controlled | 269 | 83.54 | 53 | 16.46 |
| Length of relationship with PCP* | ||||
| No PCP/DK, < 5 years | 2,636 | 83.87 | 507 | 16.13 |
| Long term, > 5 years | 1,981 | 87.54 | 282 | 12.46 |
| Prefer treatment that | ||||
| Extends life as much as possible | 1,937 | 84.88 | 345 | 15.12 |
| Relieves pain or discomfort as much as possible | 2,050 | 85.38 | 351 | 14.62 |
| Moderate to severe symptoms | ||||
| Pain* | ||||
| Yes | 574 | 70.43 | 241 | 29.57 |
| No | 4,058 | 88.09 | 549 | 11.91 |
| Fatigue* | ||||
| Yes | 1,247 | 72.21 | 480 | 27.79 |
| No | 3,383 | 91.58 | 311 | 8.42 |
| Depression* | ||||
| Yes | 550 | 66.75 | 274 | 33.25 |
| No | 3,910 | 89.05 | 481 | 10.95 |
| Nausea/vomiting* | ||||
| Yes | 231 | 71.08 | 94 | 28.92 |
| No | 4,397 | 86.32 | 697 | 13.68 |
| Cough, dyspnea, diarrhea* | ||||
| Yes | 947 | 75.28 | 311 | 24.72 |
| No | 3,566 | 88.75 | 452 | 11.25 |
| Physician communication score (higher scores indicate better communication)* | ||||
| 0-79 | 704 | 72.21 | 271 | 27.79 |
| 80-99 | 1,203 | 83.14 | 244 | 16.86 |
| 100 | 2,719 | 90.81 | 275 | 9.19 |
| Cancer type* | ||||
| Lung | 1,964 | 81.46 | 447 | 18.54 |
| Colorectal | 2,667 | 88.58 | 344 | 11.42 |
| Cancer stage | ||||
| Early | 3,193 | 85.81 | 528 | 14.19 |
| Late | 1,438 | 84.54 | 263 | 15.46 |
Abbreviation: DK, don’t know; PCP, primary care physician; VA, Veterans Affairs.
Univariable logistic regression of any unmet need on patient characteristic (at least one characteristic level with P < .05).
Table A2.
Prevalence of Patients Wanting Help for Symptoms in Early- and Late-Stage Lung and Colorectal Cancer (N = 5,422)
| Prevalence of Wanting Help for Symptoms, No. (%) | ||||
|---|---|---|---|---|
| Lung Cancer (n = 2,411) | Colorectal Cancer (n = 3,011) | |||
| Symptom | Early Stage (n = 1,295) | Late Stage (n = 1,116) | Early Stage (n = 2,426) | Late Stage (n = 585) |
| Any symptom | 884 (68.3)*† | 815 (73.0)*‡ | 1239 (51.1)*† | 364 (62.2)*‡ |
| Pain | 464 (35.8)† | 421 (37.7)‡ | 555 (22.9)*† | 165 (28.2)*‡ |
| Fatigue | 404 (31.2)*† | 426 (38.2)* | 568 (23.4)*† | 189 (32.3)* |
| Depression | 278 (21.5)† | 240 (21.5) | 369 (15.2)† | 113 (19.3) |
| Nausea/vomiting | 231 (17.8)* | 311 (27.9)* | 442 (18.2)* | 181 (30.9)* |
| Cough | 351 (27.1) | 352 (31.5) | N/A | N/A |
| Dyspnea | 416 (32.1) | 402 (36.0) | N/A | N/A |
| Diarrhea | N/A | N/A | 499 (20.6)* | 153 (26.2)* |
NOTE. The majority of patients who reported wanting help for any symptom had moderate to severe symptoms (68%), but the rest of the patients who wanted help for their symptoms reported mild symptoms. Abbreviation: N/A, not applicable.
P < .01, comparison between early stage and late stage within cancer type.
P < .01, comparison between early-stage lung and early-stage colorectal cancer.
P < .01, comparison between late-stage lung and late-stage colorectal cancer.
Table A3.
Predicted Marginal Probabilities for Unmet Needs for Common Symptoms in Patients With Lung and Colorectal Cancer (which retains only patients who reported symptoms, n = 5,067)
| Unmet Need for Any Symptom | ||||||
|---|---|---|---|---|---|---|
| Patient Characteristic | No. (%) | Coefficient (SE) | P | F test P | OR (95% CI) | Predicted Marginal Probability (%) |
| Sex | .838 | |||||
| Male | 2,673 (52.8) | Reference | Reference | 15.3 | ||
| Female | 2,374 (47.3) | 0.02 (0.09) | .838 | 1.02 (0.85 to 1.23) | 15.6 | |
| Age, years | .626 | |||||
| 21-59 | 1,760 (34.7) | 0.23 (0.19) | .233 | 1.26 (0.86 to 1.82) | 16.5 | |
| 60-69 | 1,475 (29.1) | 0.11 (0.18) | .507 | 1.12 (0.80 to 1.59) | 15.2 | |
| 70-79 | 1,302 (25.7) | 0.08 (0.17) | .625 | 1.09 (0.78 to 1.52) | 14.9 | |
| ≥ 80 | 530 (10.5) | Reference | Reference | 14.0 | ||
| Race | .035 | |||||
| White | 3,525 (69.6) | Reference | Reference | 14.8 | ||
| Hispanic or Latino | 347 (6.9) | 0.25 (0.17) | .137 | 1.28 (0.92 to 1.79) | 17.8 | |
| African American | 688 (13.6) | 0.32 (0.13) | .012 | 1.37 (1.07 to 1.76) | 18.6 | |
| Other | 507 (10.0) | −0.07 (0.15) | .646 | 0.93 (0.69 to 1.25) | 14.1 | |
| Education | .854 | |||||
| Less than high school | 868 (17.2) | −0.06 (0.15) | .663 | 0.94 (0.71 to 1.25) | 14.9 | |
| High school/some college | 2,971 (58.7) | 0.001 (0.11) | .990 | 1.00 (0.81 to 1.24) | 15.6 | |
| College degree or more | 1,222 (24.2) | Reference | Reference | 15.6 | ||
| Insurance status at time of diagnosis | .149 | |||||
| VA | 597 (11.9) | 0.24 (0.21) | .268 | 1.27 (0.83 to 1.95) | 17.5 | |
| Private | 1,799 (35.8) | Reference | Reference | 14.6 | ||
| Medicare + supplemental | 1,839 (36.6) | −0.01 (0.14) | .968 | 0.99 (0.76 to 1.30) | 14.6 | |
| Public | 579 (11.5) | 0.16 (0.15) | .284 | 1.18 (0.87 to 1.59) | 16.5 | |
| None | 209 (4.2) | 0.45 (0.20) | .023 | 1.57 (1.07 to 2.32) | 20.2 | |
| Wealth (how much money saved to live at current cost of living) | .332 | |||||
| < 1 month | 997 (23.4) | 0.22 (0.12) | .063 | 1.25 (0.99 to 1.57) | 16.6 | |
| 1-2 months | 547 (12.8) | 0.15 (0.15) | .303 | 1.17 (0.87 to 1.57) | 15.8 | |
| 3-6 months | 546 (12.8) | 0.15 (0.14) | .283 | 1.17 (0.88 to 1.55) | 15.8 | |
| 7-12 months | 343 (8.0) | 0.28 (0.17) | .107 | 1.32 (0.94 to 1.86) | 17.3 | |
| > 1 year | 1,832 (43.0) | Reference | Reference | 14.1 | ||
| Marital status | .865 | |||||
| Not married | 2,140 (42.3) | −0.02 (0.09) | .865 | 0.98 (0.83 to 1.17) | 15.3 | |
| Married | 2,925 (57.8) | Reference | Reference | 15.6 | ||
| Surgery in past 6 weeks | .290 | |||||
| Not received | 4,860 (95.9) | Reference | Reference | 15.6 | ||
| Received | 207 (4.1) | −0.23 (0.21) | .290 | 0.79 (0.52 to 1.22) | 13.1 | |
| Radiation in past 6 weeks | .249 | |||||
| Not received | 4,461 (88.0) | Reference | Reference | 15.7 | ||
| Received | 606 (12.0) | −0.15 (0.13) | .249 | 0.86 (0.67 to 1.11) | 14.1 | |
| Chemotherapy in past 6 weeks | .432 | |||||
| Not received | 2,790 (55.1) | Reference | Reference | 15.9 | ||
| Received | 2,277 (44.9) | −0.07 (0.09) | .432 | 0.93 (0.77 to 1.12) | 15.0 | |
| Comorbidities | .714 | |||||
| 0-1 | 3,013 (59.6) | Reference | Reference | 15.7 | ||
| ≥ 2 | 2,044 (40.4) | −0.03 (0.09) | .714 | 0.97 (0.81 to 1.16) | 15.3 | |
| In the year before diagnosis, hospitalized for any other medical condition | .879 | |||||
| No | 4,009 (79.5) | Reference | Reference | 15.5 | ||
| Yes | 1,036 (20.5) | −0.02 (0.10) | .879 | 0.98 (0.81 to 1.21) | 15.4 | |
| Clinical trial | .234 | |||||
| No | 4,859 (95.9) | Reference | Reference | 15.6 | ||
| Yes | 208 (4.1) | −0.28 (0.23) | .224 | 0.75 (0.47 to 1.19) | 12.6 | |
| Preferred role in decisions about treatment | .596 | |||||
| Patient controlled | 1,809 (36.0) | Reference | Reference | 14.8 | ||
| Shared control | 2,920 (58.1) | 0.09 (0.09) | .334 | 1.09 (0.91 to 1.30) | 15.8 | |
| Physician controlled | 299 (6.0) | 0.11 (0.18) | .543 | 1.12 (0.75 to 1.59) | 16.1 | |
| Length of relationship with PCP | .260 | |||||
| No PCP/DK, < 5 years | 2,954 (58.5) | Reference | Reference | 16.0 | ||
| Long term, > 5 years | 2,097 (41.5) | −0.10 (0.09) | .260 | 0.90 (0.75 to 1.07) | 14.8 | |
| Prefer treatment that | .667 | |||||
| Extends life as much as possible | 2,155 (49.2) | Reference | Reference | 15.7 | ||
| Relieves pain or discomfort as much as possible | 2,229 (50.8) | −0.04 (0.10) | .668 | 0.96 (0.79 to 1.16) | 15.3 | |
| Moderate to severe symptoms | ||||||
| Pain | < .001 | |||||
| No | 4,252 (83.9) | Reference | Reference | 14.4 | ||
| Yes | 815 (16.1) | 0.43 (0.10) | < .001 | 1.54 (1.25 to 1.88) | 20.0 | |
| Fatigue | < .001 | |||||
| No | 3,340 (65.9) | Reference | Reference | 10.9 | ||
| Yes | 1,727 (34.1) | 0.93 (0.09) | < .001 | 2.53 (2.10 to 3.05) | 22.3 | |
| Depression | < .001 | |||||
| No | 4,243 (83.7) | Reference | Reference | 13.9 | ||
| Yes | 824 (16.3) | 0.53 (0.10) | < .001 | 1.69 (1.38 to 2.07) | 20.5 | |
| Nausea/vomiting | .162 | |||||
| No | 4,742 (93.6) | Reference | Reference | 15.3 | ||
| Yes | 325 (6.4) | 0.21 (0.15) | .162 | 1.23 (0.92 to 1.64) | 17.7 | |
| Cough, dyspnea, diarrhea | < .001 | |||||
| No | 3,809 (75.2) | Reference | Reference | 13.9 | ||
| Yes | 1,258 (24.8) | 0.43 (0.12) | < .001 | 1.54 (1.22 to 1.93) | 19.1 | |
| Physician communication (higher scores indicate better communication) | < .001 | |||||
| 0-79 | 947 (18.7) | 1.09 (0.11) | < .001 | 2.98 (2.42 to 3.66) | 24.3 | |
| 80-99 | 1,380 (27.3) | 0.60 (0.10) | < .001 | 1.82 (1.49 to 2.23) | 17.2 | |
| 100 (reference) | 2,735 (54.0) | Reference | Reference | 10.8 | ||
| Cancer stage | .821 | |||||
| Early | 3,423 (67.6) | Reference | ||||
| Late | 1,644 (32.4) | −0.03 (0.15) | .821 | |||
| Cancer type | .009 | |||||
| Lung | 2,372 (46.8) | 0.34 (0.13) | .009 | |||
| Colorectal | 2,695 (53.2) | Reference | ||||
| Cancer stage/type interaction | −0.38 (0.19) | .044 | .044 | |||
| Early stage | ||||||
| Lung | 1,272 (25.1) | 1.40 (0.72 to 2.73) | 18.7 | |||
| Colorectal | 2,151 (42.5) | Reference | 14.7 | |||
| Late stage | ||||||
| Lung | 1,100 (21.7) | 0.93 (0.44 to 1.94) | 13.8 | |||
| Colorectal | 544 (10.7) | 0.97 (0.72 to 1.30) | 14.3 | |||
Abbreviations: DK, don’t know; OR, odds ratio; PCP, primary care physician; SE, standard error; VA, Veterans Affairs.
AUTHOR CONTRIBUTIONS
Conception and design: Anne M. Walling, Nancy L. Keating, Katherine L. Kahn, Sydney Dy, Jennifer W. Mack, Jennifer Malin, Neeraj K. Arora, John L. Adams, Diana Tisnado
Financial support: Jennifer Malin
Collection and assembly of data: Nancy L. Keating, Katherine L. Kahn, Jennifer Malin
Data analysis and interpretation: All authors
Manuscript writing: All authors
Final approval of manuscript: All authors
AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
Lower Patient Ratings of Physician Communication Are Associated With Unmet Needs for Symptom Management in Patients With Lung and Colorectal Cancer
The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or jop.ascopubs.org/site/misc/ifc.xhtml.
Anne M. Walling
No relationship to disclose
Nancy L. Keating
No relationship to disclose
Katherine L. Kahn
No relationship to disclose
Sydney Dy
No relationship to disclose
Jennifer W. Mack
No relationship to disclose
Jennifer Malin
Employment: Anthem
Stock or Other Ownership: Anthem
Neeraj K. Arora
No relationship to disclose
John L. Adams
No relationship to disclose
Anna Liza M. Antonio
No relationship to disclose
Diana Tisnado
Employment: Pricespective (I)
References
- 1. Walling AM, Weeks JC, Kahn KL, et al: Symptom prevalence in lung and colorectal cancer patients. J Pain Symptom Mange 49:192-202, 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Basch E, Snyder C, McNiff K, et al. Patient-reported outcome performance measures in oncology. J Oncol Pract. 2014;10:209–211. doi: 10.1200/JOP.2014.001423. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. doi: 10.1016/j.jpainsymman.2014.07.010. Etkind SN, Daveson BA, Kwok W, et al: Capture, transfer, and feedback of patient-centered outcomes data in palliative care populations: Does it make a difference? A systematic review. J Pain Symptom Manage 49:611-624, 2015. [DOI] [PubMed] [Google Scholar]
- 4.Snyder CF, Blackford AL, Brahmer JR, et al. Needs assessments can identify scores on HRQOL questionnaires that represent problems for patients: An illustration with the Supportive Care Needs Survey and the QLQ-C30. Qual Life Res. 2010;19:837–845. doi: 10.1007/s11136-010-9636-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Snyder CV, Dy SM, Hendricks DE, et al. Asking the right questions: Investigating needs assessments and Health-related quality of life questionnaires for use in oncology clinical practice. Support Care Cancer. 2007;15:1075–1085. doi: 10.1007/s00520-007-0223-1. [DOI] [PubMed] [Google Scholar]
- 6.Barbera L, Seow H, Howell D, et al. Symptom burden and performance status in a population-based cohort of ambulatory cancer patients. Cancer. 2010;116:5767–5776. doi: 10.1002/cncr.25681. [DOI] [PubMed] [Google Scholar]
- 7.Cleeland CS, Zhao F, Chang VT, et al. The symptom burden of cancer: Evidence for a core set of cancer-related and treatment-related symptoms from the Eastern Cooperative Oncology Group Symptom Outcomes and Practice Patterns study. Cancer. 2013;119:4333–4340. doi: 10.1002/cncr.28376. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. National Cancer Institute Division of Cancer Control & Population Sciences: Cancer Care Outcomes Research & Surveillance Consortium (CanCORS). http://healthservices.cancer.gov/cancors.
- 9.American Joint Committee on Cancer . AJCC Cancer Staging Atlas. Chicago, IL: Spingerlink; 2006. [Google Scholar]
- 10.Ayanian JZ, Chrischilles EA, Fletcher RH, et al. Understanding cancer treatment and outcomes: The Cancer Care Outcomes Research and Surveillance Consortium. J Clin Oncol. 2004;22:2992–2996. doi: 10.1200/JCO.2004.06.020. [DOI] [PubMed] [Google Scholar]
- 11.Malin JL, Ko C, Ayanian JZ, et al. Understanding cancer patients’ experience and outcomes: Development and pilot study of the Cancer Care Outcomes Research and Surveillance patient survey. Support Care Cancer. 2006;14:837–848. doi: 10.1007/s00520-005-0902-8. [DOI] [PubMed] [Google Scholar]
- 12.Cancer Care Outcomes Research and Surveillance Consortium http://www.cancors.org/public.
- 13.Weeks JC, Catalano PJ, Cronin A, et al. Patients’ expectations about effects of chemotherapy for advanced cancer. N Engl J Med. 2012;367:1616–1625. doi: 10.1056/NEJMoa1204410. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Mack JW, Walling A, Dy S, et al. Patient beliefs that chemotherapy may be curative and care received at the end of life among patients with metastatic lung and colorectal cancer. Cancer. 2015;121:1891–1897. doi: 10.1002/cncr.29250. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Yoon J, Malin JL, Tisnado DM, et al. Symptom management after breast cancer treatment: Is it influenced by patient characteristics. Breast Cancer Res Treat. 2008;108:69–77. doi: 10.1007/s10549-007-9580-1. [DOI] [PubMed] [Google Scholar]
- 16. Reference deleted.
- 17.Katz JN, Chang LC, Sangha O, et al. Can comorbidity be measured by questionnaire rather than medical record review. Med Care. 1996;34:73–84. doi: 10.1097/00005650-199601000-00006. [DOI] [PubMed] [Google Scholar]
- 18.Keating NL, Beth Landrum M, Arora NK, et al. Cancer patients’ roles in treatment decisions: Do characteristics of the decision influence roles. J Clin Oncol. 2010;28:4364–4370. doi: 10.1200/JCO.2009.26.8870. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Ayanian JZ, Zaslavsky AM, Arora NK, et al. Patients’ experiences with care for lung cancer and colorectal cancer: Findings from the Cancer Care Outcomes Research and Surveillance Consortium. J Clin Oncol. 2010;28:4154–4161. doi: 10.1200/JCO.2009.27.3268. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. doi: 10.1097/00005650-199903001-00003. Hays RD, Shaul JA, Williams VS, et al: Psychometric properties of the CAHPS 1.0 survey measures. Consumer Assessment of Health Plans study. Med Care 37:MS22-MS31, 1999 (suppl 3) [DOI] [PubMed] [Google Scholar]
- 21.Gelman A, Hill J. Data Analysis Using Regression and Multilevel/Hierarchical Models. Cambridge, UK: Cambridge University Press; 2006. doi: 10.1017/CBO9780511790942. [Google Scholar]
- 22. He Y, Zaslavsky AM, Harrington DP, et al: Imputation in a Multiformat and Multiwave Survey of Cancer Care. Proceedings in Health Policy Statistics. Alexandria, VA, American Statistical Association, 2007, pp 1541-1549. [Google Scholar]
- 23.Little RJA. Direct standardization: A tool for teaching linear models for unbalanced data. Am Stat. 1982;36:38–43. [Google Scholar]
- 24.Rueda JR, Solà I, Pascual A, et al. Non-invasive interventions for improving well-being and quality of life in patients with lung cancer. Cochrane Database Syst Rev. 2011;7:CD004282. doi: 10.1002/14651858.CD004282.pub3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Sikorskii A, Given CW, Siddiqi AE, et al: Testing the differential effects of symptom management interventions in cancer. Psychoonology 24:25-32, 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Sikorskii A, Given CW, Given B, et al. Symptom management for cancer patients: A trial comparing two multimodal interventions. J Pain Symptom Manage. 2007;34:253–264. doi: 10.1016/j.jpainsymman.2006.11.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Quigley DD, Martino SC, Brown JA, et al. Evaluating the content of the communication items in the CAHPS(®) clinician and group survey and supplemental items with what high-performing physicians say they do. Patient. 2013;6:169–177. doi: 10.1007/s40271-013-0016-1. [DOI] [PubMed] [Google Scholar]
- 28.Browne K, Roseman D, Shaller D, et al. Analysis & commentary. Measuring patient experience as a strategy for improving primary care. Health Aff (Millwood) 2010;29:921–925. doi: 10.1377/hlthaff.2010.0238. [DOI] [PubMed] [Google Scholar]
- 29. doi: 10.1097/MLR.0b013e31826cbc0d. Dyer N, Sorra JS, Smith SA, et al: Psychometric properties of the Consumer Assessment of Healthcare Providers and Systems (CAHPS®) Clinician and Group Adult Visit Survey. Med Care 50:S28-S34, 2012 (suppl) [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Hargraves JL, Hays RD, Cleary PD. Psychometric properties of the Consumer Assessment of Health Plans Study (CAHPS) 2.0 adult core survey. Health Serv Res. 2003;38:1509–1527. doi: 10.1111/j.1475-6773.2003.00190.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Tallman K, Janisse T, Frankel RM, et al. Communication practices of physicians with high patient-satisfaction ratings. Perm J. 2007;11:19–29. doi: 10.7812/tpp/06-106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. doi: 10.1037/a0025334. Song L, Hamilton JB, Moore AD: Patient-healthcare provider communication: Perspectives of African American cancer patients. Health Psychol 31:539-547, 2012. [DOI] [PubMed] [Google Scholar]
- 33.John DA, Kawachi I, Lathan CS, et al. Disparities in perceived unmet need for supportive services among patients with lung cancer in the Cancer Care Outcomes Research and Surveillance Consortium. Cancer. 2014;120:3178–3191. doi: 10.1002/cncr.28801. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Kamal AH, Maguire JM, Wheeler JL, et al. Dyspnea review for the palliative care professional: Treatment goals and therapeutic options. J Palliat Med. 2012;15:106–114. doi: 10.1089/jpm.2011.0110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Ho SF, O’Mahony MS, Steward JA, et al. Dyspnoea and quality of life in older people at home. Age Ageing. 2001;30:155–159. doi: 10.1093/ageing/30.2.155. [DOI] [PubMed] [Google Scholar]
- 36. doi: 10.1016/j.ahj.2009.03.021. Arnold SV, Spertus JA, Jones PG, et al: The impact of dyspnea on health-related quality of life in patients with coronary artery disease: Results from the PREMIER registry. Am Heart J 157:1042-1049, 2009. [DOI] [PubMed] [Google Scholar]
- 37.Smith EL, Hann DM, Ahles TA, et al. Dyspnea, anxiety, body consciousness, and quality of life in patients with lung cancer. J Pain Symptom Manage. 2001;21:323–329. doi: 10.1016/s0885-3924(01)00255-x. [DOI] [PubMed] [Google Scholar]
- 38.Lorenz KA, Dy SM, Naeim A, et al. Quality measures for supportive cancer care: The Cancer Quality-ASSIST Project. J Pain Symptom Manage. 2009;37:943–964. doi: 10.1016/j.jpainsymman.2008.05.018. [DOI] [PubMed] [Google Scholar]
- 39. ASCO Institute for Quality: Quality Oncology Practice Initiative (QOPI®) and the QOPI Certification Program (QCP™). http://www.instituteforquality.org/qopi-qcp.
- 40. National Quality Forum: http://www.qualityforum.org.
- 41. Institute of Medicine: Improving Palliative Care for Cancer: Summary and Recommendations (2001). Washington, DC, Institute of Medicine, 2001.