Abstract
Purpose:
Cancer survivors have unique medical care needs. “Shared care,” delivered by both oncologists and primary care providers (PCPs), may better address these needs. Little information is available on differences in outcomes among survivors receiving shared care versus oncologist-led or PCP-led care. This study compared experiences of care for survivors receiving Shared Care, Oncologist-led, PCP-led or Other care patterns.
Methods:
We used SEER-CAHPS data, including NCI’s SEER registry data, Medicare claims, and Medicare Consumer Assessment of Healthcare Providers and Systems (CAHPS) survey responses. Medicare fee-for-service beneficiaries age≥65 years in SEER-CAHPS with breast, cervical, colorectal, lung, renal, or prostate cancers or hematologic malignancies who responded to a Medicare CAHPS survey≥18 months post-diagnosis were included. CAHPS measures included ratings of overall care, personal doctor, specialist physician, health plan, prescription drug plan, and five composite scores. Survivorship care patterns were identified using proportions of oncologist, PCP, and other physician encounters. Multivariable regressions examined associations between care patterns and CAHPS outcomes.
Results:
Among 10,132 survivors, 15% received Shared Care; 10% Oncologist-led; 33% PCP-led; and 42% Other. Compared with Shared Care, we found no significant differences in experiences of care except for getting needed drugs (lower scores for PCP-led and Other care patterns). Sensitivity analyses using different patterns of care definitions similarly showed no associations between survivorship care pattern and experience of care.
Conclusions:
Within the limitations of the study dataset, survivors age 65+ receiving Shared Care reported similar experiences of care to those receiving Oncologist-led, PCP-led, and Other patterns of care.
Implications for Cancer Survivors:
Shared care may not provide survivor-perceived benefits compared with other care patterns.
Keywords: Survivorship, Oncology, Physician Practice Patterns, Patient Satisfaction, Claims Analyses, SEER Program
INTRODUCTION
In January 2019, there were approximately 16.9 million cancer survivors in the United States; this is expected to grow to almost 22.2 million by 2030.[1] Cancer survivors may have unique medical care needs related to chronic/late-occurring effects of cancer or cancer treatment and require specialized survivorship care services to ensure all survivors’ health needs are met.[2] To address survivors’ needs and improve outcomes, “shared care”, involving survivors’ medical care delivered by both oncologists and primary care providers (PCPs), may increase care coordination/communications between health care providers; decrease fragmentation of care; and improve patient experience of care. Other survivorship care models include programs led by PCPs, oncologists, and other health care providers.[3] However, little is known about outcomes for different types of survivorship care, and there is no widely agreed-upon definition of “shared care”. Furthermore, shared care may require additional time and effort from healthcare providers for communications and care coordination, which may increase provider burden.[3, 4]
Few studies have explored associations between receipt of survivorship care services and types of physicians providing survivorship care. Yu et al. reported that colorectal cancer (CRC) survivors who received at least one visit from an oncologist or gynecologist were more likely to receive mammograms than were survivors who received care only from PCPs.[5] Snyder et al. reported that CRC survivors with at least one visit with both oncologists and PCPs were more likely to receive preventive care services than those seeing only PCPs or oncologists.[6] Men seeing any PCPs in the first-year after prostate cancer diagnosis were more likely to receive certain preventive care services than those receiving care from oncologist; however, there were no significant differences five-years post-diagnosis.[7]
Research examining associations between survivors’ experience of care and their types of healthcare providers is generally limited to patient satisfaction or follow-up care ratings from survivors in limited geographic regions.[8–12] Weaver et al. found no significant difference in self-reported follow-up care ratings between survivors identifying oncologists vs. PCPs as their main physician.[8] Emery et al. reported no significant differences in distress, prostate- specific quality of life, patient satisfaction or unmet needs among prostate cancer survivors receiving usual care vs. shared care.[11] In contrast, Grunfeld et al. reported significantly greater satisfaction among breast cancer survivors in England who had follow-up care with general practice groups (i.e., PCPs) vs. hospital outpatient (specialist) clinics.[12]
SEER-CAHPS is a recently developed data resource linking patient experience information from the Medicare Consumer Assessment of Healthcare Providers and Systems (Medicare CAHPS®) Survey with clinical information from NCI’s Surveillance, Epidemiology, and End Results (SEER) cancer registry program and Medicare Fee-for-Service (FFS) claims.[13, 14] This dataset includes sociodemographic and clinical characteristics and experiences of care among a large, national sample of individuals diagnosed with cancer. Recent SEER-CAHPS studies examined experience of care among cancer survivors and individuals with cancer in the last year of life.[15, 16]
To better understand relationships between physician specialties involved in survivorship care and survivors’ experience of care, we used SEER-CAHPS to examine associations of PCP-led, Oncologist-led, or Shared Care survivorship care patterns with survivors’ experiences of care, using a data-derived classification of shared care. We hypothesized that shared care would be an enhanced collaborative process. That is, shared care would potentially facilitate communication/coordination among health care providers and between the survivor and the health care team, leading to improved experience of care among survivors.
METHODS
Study Population
The study population included individuals in SEER-CAHPS diagnosed 2000–2011 with hematologic malignancies or breast, cervical, colorectal, lung, renal, or prostate cancer. Those diagnosed with in situ, metastatic, or unknown stage disease (N=9,376) or <65 years at diagnosis (N=13,543) were excluded. Thirteen males with breast cancer were also excluded. SEER-CAHPS data are available from NCI; the process for obtaining these data is outlined at https://healthcaredelivery.cancer.gov/seer-cahps/obtain/.
As the CAHPS survey (described below) asks respondents to consider the previous six months of care, individuals included in the study were required to have responded to a Medicare CAHPS survey≥18 months after diagnosis and to have been continuously enrolled in FFS Medicare A and B at least 6 months before and 6 months after completing a CAHPS survey. This ensured that survey completion occurred during a 12-month period of survivorship care starting at least 12 months after diagnosis with stable, continuous Medicare A/B enrollment. Individuals with <2 physician encounters in this 12-month period were excluded (N=14,836), as a survivorship care pattern could not be assigned for those with only a single physician visit.
Survivorship Care Patterns
Cancer survivors were classified in one of four mutually exclusive survivorship care pattern groups based on their percentages of oncologist, primary care, or other types of physician encounters: Shared Care, Oncologist-led, PCP-led, and Other (Figure 1). Total number of physician encounters and physician specialty at each encounter were determined using office/outpatient evaluation and management (E&M) visits (CPT 99201–99205, 99211–99215) from Medicare claims. Survivorship care pattern were determined based on physician specialty from all E&M encounters in Medicare claims for the 12 month study period (6 month before through 6 months after completing a CAHPS survey); the CAHPS survey therefore captured experience of care in the middle of a period of continuous Medicare Part A/B enrollment. Advanced practice provider encounters (i.e., NP or PA visits) are generally captured in these claims data under the associated physicians’ specialty. The proportion of visits by physician type and number of visits for each survivorship care model are presented in Table 1.
Figure 1:
Patterns of care definitions used in this study. Figure 1 illustrates criteria used in base case analyses and sensitivity analyze to categorize survivors as receiving Oncologist-led, PCP-led, Shared Care, or Other survivorship care patterns.
Table 1.
Characteristics of the study population by survivorship care pattern (base-case)
| Shared Care (n=1,525) | Oncologist-Led (n=1,013) | PCP-Led (n=3,314) | Other (n=4,280) | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| VARIABLE | N | Weighted % | N | Weighted % | P-Value | N | Weighted % | P-Value | N | Weighted % | P-Value |
| Female | 667 | 39.4% | 425 | 41.1% | .535 | 1,831 | 54.2% | < .001 | 1,707 | 39.7% | .895 |
| Age | .085 | < .001 | < .001 | ||||||||
| 65–69 | 66 | 3.6% | 71 | 6.1% | 88 | 2.7% | 120 | 2.3% | |||
| 70–74 | 420 | 28.9% | 307 | 29.6% | 683 | 20.8% | 909 | 20.8% | |||
| 75–79 | 459 | 29.6% | 298 | 27.7% | 956 | 28.4% | 1,273 | 30.5% | |||
| 80–84 | 340 | 21.3% | 216 | 23.1% | 802 | 23.1% | 1,089 | 24.5% | |||
| 85+ | 240 | 16.6% | 121 | 13.5% | 785 | 25.0% | 889 | 21.9% | |||
| Education | .459 | .010 | .162 | ||||||||
| <HS | 225 | 13.9% | 164 | 14.9% | 604 | 17.1% | 566 | 12.0% | |||
| HS Grad/GED | 425 | 27.7% | 299 | 28.5% | 993 | 30.7% | 1,100 | 25.5% | |||
| Some College | 345 | 23.4% | 222 | 24.2% | 757 | 23.5% | 938 | 22.2% | |||
| >=College Grad | 431 | 28.6% | 250 | 24.4% | 716 | 21.5% | 1,382 | 33.1% | |||
| Missing | 99 | 6.4% | 78 | 7.9% | 244 | 7.2% | 294 | 7.2% | |||
| Race/Ethnicity | .321 | .518 | .036 | ||||||||
| Non-Hispanic White | 1,162 | 77.3% | 766 | 77.9% | 2,494 | 76.7% | 3,457 | 81.8% | |||
| Non-Hispanic Black | 76 | 5.2% | 65 | 6.8% | 177 | 5.5% | 151 | 3.8% | |||
| Non-Hispanic Asian | 73 | 3.0% | 41 | 3.2% | 153 | 4.4% | 135 | 2.2% | |||
| Hispanic | 79 | 4.9% | 45 | 3.9% | 164 | 4.0% | 163 | 3.2% | |||
| Other | 37 | 2.6% | 20 | 1.4% | 79 | 2.2% | 90 | 2.0% | |||
| Missing | 98 | 6.9% | 76 | 6.8% | 247 | 7.2% | 284 | 7.0% | |||
| Dual enrollee | 137 | 8.4% | 106 | 10.7% | .158 | 343 | 10.0% | .201 | 275 | 5.6% | .005 |
| General health | .728 | .452 | .340 | ||||||||
| Excellent/Very Good | 465 | 32.2% | 263 | 30.0% | 1,013 | 34.3% | 1,195 | 29.7% | |||
| Good | 586 | 39.9% | 416 | 40.9% | 1,208 | 37.8% | 1,652 | 40.6% | |||
| Fair | 339 | 21.7% | 229 | 22.0% | 759 | 22.6% | 1,034 | 23.8% | |||
| Poor | 72 | 5.6% | 55 | 6.7% | 178 | 5.0% | 252 | 5.7% | |||
| Missing/Don’t Know† | 63 | 3.6% | 50 | 4.5% | 156 | 3.5% | 147 | 3.5% | |||
| Mental health | .750 | .624 | .857 | ||||||||
| Excellent/Very Good | 958 | 65.0% | 614 | 64.8% | 1,938 | 61.5% | 2,684 | 66.4% | |||
| Good | 371 | 25.4% | 267 | 25.7% | 882 | 28.4% | 1,042 | 24.3% | |||
| Fair | 115 | 8.0% | 70 | 7.8% | 286 | 8.7% | 335 | 7.8% | |||
| Poor | 21 | 1.4% | 16 | 1.6% | 57 | 1.3% | 62 | 1.3% | |||
| Missing/Don’t Know† | 60 | 3.2% | 46 | 4.3% | 151 | 3.3% | 157 | 4.1% | |||
| Self-reported comorbidities* | |||||||||||
| COPD | 142 | 14.7% | 78 | 14.4% | 896 | 383 | 18.8% | .068 | 633 | 22.9% | < .001 |
| Diabetes | 311 | 30.5% | 137 | 22.3% | .012 | 755 | 32.8% | .432 | 913 | 31.1% | .821 |
| Heart attack | 184 | 29.4% | 69 | 18.1% | .002 | 354 | 25.0% | .147 | 734 | 42.2% | < .001 |
| Stroke | 96 | 9.0% | 36 | 6.9% | .242 | 227 | 11.8% | .111 | 312 | 11.1% | .159 |
| Cancer site | |||||||||||
| Breast/Cervical# | 382 | 23.5% | 214 | 20. 4% | .095 | 848 | 23.8% | .993 | 718 | 16.5% | < .001 |
| Colorectal | 57 | 2.9% | 60 | 5.4% | .016 | 202 | 6.2% | < .001 | 209 | 4.9% | .003 |
| Lung | 49 | 2.5% | 41 | 4.1% | .092 | 112 | 3.7% | .138 | 191 | 4.3% | .009 |
| Renal | 22 | 1.1% | 15 | 1.1% | .994 | 109 | 3.9% | < .001 | 148 | 3.7% | < .001 |
| Prostate | 684 | 49.7% | 446 | 43.4% | .027 | 974 | 29.7% | < .001 | 1,510 | 35.4% | < .001 |
| Other | 331 | 20.1% | 237 | 25.5% | .029 | 1,069 | 32.8% | < .001 | 1,504 | 35.1% | < .001 |
| Cancer stage | |||||||||||
| Local | 1,146 | 78.6% | 680 | 66.6% | < .001 | 2,568 | 77.0% | .388 | 3,301 | 76.7% | .261 |
| Regional | 312 | 17.9% | 270 | 26.8% | < .001 | 625 | 18.7% | .645 | 727 | 17.1% | .609 |
| Missing‡ | 67 | 3.5% | 63 | 6.6% | .045 | 121 | 4.3% | .370 | 252 | 6.2% | .001 |
| Years since diagnosis | .038 | < .001 | < .001 | ||||||||
| 1–2 years | 179 | 12.7% | 161 | 16.3% | 258 | 7.8% | 355 | 7.3% | |||
| 2–5 years | 783 | 48.0% | 549 | 51.2% | 1,344 | 39.2% | 1,815 | 43.2% | |||
| 5+ years | 563 | 39.3% | 303 | 32.6% | 1,712 | 53.0% | 2,110 | 49.4% | |||
| Total physician encounters, by quartile | < .001 | < .001 | .003 | ||||||||
| 2–5 Visits | 292 | 19.9% | 449 | 43.4% | 1,208 | 36.0% | 846 | 19.7% | |||
| 6–9 Visits | 480 | 31.3% | 228 | 22.2% | 952 | 30.5% | 1,131 | 26.5% | |||
| 10–14 Visits | 396 | 25.9% | 124 | 13.3% | 663 | 19.1% | 1,029 | 23.8% | |||
| 15+ Visits | 357 | 22.8% | 212 | 21.1% | 491 | 14.4% | 1,274 | 29.9% | |||
| Census region | .008 | .121 | <.001 | ||||||||
| Northeast | 279 | 18.6% | 181 | 19.7% | 559 | 21.5% | 993 | 25.3% | |||
| Midwest | 177 | 11.0% | 172 | 16.9% | 388 | 11.3% | 407 | 8.5% | |||
| South | 388 | 27.1% | 213 | 21.1% | 778 | 22.6% | 982 | 22.8% | |||
| West | 681 | 43.3% | 447 | 42.3% | 1587 | 44.6% | 1892 | 43.4% | |||
Notes:
Weighted percentages are estimates of national percentages after adjusting for sample weights and survey design.
P-values denote statistical significance of comparing the Oncologist-Led, PCP-Led, and Other groups to the Shared Care group.
Missingness includes question not answered, multiple response, and refused response.
Missingness due to unavailable data.
Cervical cancer is reported in this table combined with breast cancer to suppress presenting cells with small sample sizes (≤10). However, p-values are determined using breast cancer alone. There were no significant differences in the proportion of survivors with cervical cancer by survivorship care pattern.
Only available in CAHPS surveys 2008 and later.
Oncologists included physicians with specialties of hematology/oncology, medical oncology, surgical oncology, radiation oncology, hematology, gynecology/oncology, or urology (for individuals with prostate cancer). Physicians with more than one specialty type on E&M claims were classified as oncologists if at least one specialty code was for oncology. PCPs included family practice, general medicine, general surgery, internal medicine, geriatric medicine, obstetrics/gynecology, and preventive medicine. If they had no oncology specialty codes, physicians were classified as PCPs if at least one of their specialty codes was for primary care. Physicians with neither oncology nor PCP specialty codes for E&M visits were combined in an “other specialty” category. While survivors’ medical encounters may include care for health issues other than cancer, these patterns quantify the type(s) of physicians most frequently seen by survivors.
In determining the survivorship care pattern for each patient, claims with specialties listed as “Multispecialty Clinics” or “Group Practice” were excluded, as these may have involved multiple physician specialties. Claims from emergency department/urgent care encounters and encounters with emergency medicine physicians, anesthesiologists, pathologists, and pediatricians were excluded in determining survivorship care patterns. We also excluded visits with advanced practice providers (NPs or PAs) who were not affiliated with a physician practice, as information on the specialty of advanced practice providers was not available. Claims from physicians with specialty codes “Pain Medicine” and “Physical Medicine and Rehabilitation” were also excluded, since it was uncertain whether care provided by these physicians was cancer-related.
Our approach to identifying survivorship patterns of care classified survivors as receiving Shared Care, Oncologist-led, PCP-led, or Other physician patterns based on their proportion of E&M visits with oncologists, PCPs, and other types of physicians over the 12-month study period. Survivors were classified in the Oncologist-led pattern if ≥60% of their physician encounters were with oncologists and in the PCP-led pattern if ≥60% of their physician encounters were with PCPs. We defined the Shared Care group as survivors not in Oncologist-led or PCP-led patterns who had more than the median proportion of both oncologist and PCP visits (i.e., ≥17% of encounters with oncologists and ≥33% of encounters with PCPs). This is a more data-driven definition of Shared Care than in some prior studies that defined Shared Care as one or more visits with both PCPs and oncologists.[5, 6, 17] All survivors not classified in Oncologist-led, PCP-led, or Shared Care patterns were classified in the Other pattern. Details regarding the proportion of visits by physician type and the number of physician visits for each survivorship care model are presented in Table 1.
We also conducted sensitivity analyses to explore whether our findings were robust to a different definition of Shared Care. For this alternative definition, we classified survivors in the Shared Care pattern if: (1) their pattern of physician visits did not meet criteria for either Oncologist-led or PCP-led care; (2) their proportion of oncologist encounters plus PCP encounters was >50% of all physician encounters; (3) the proportion of oncologist visits was greater than the proportion of other physician encounters (non-oncologist and non-PCP); and (4) their proportion of PCP encounters was greater than the proportion of other physician encounters. This alternative definition reduced the percentage of survivors in Shared Care by approximately 6%, re-categorizing these survivors to the Other survivorship care pattern.
Outcome Measures
The CAHPS survey queries respondents about care received within the last 6-months. Study analyses include ten standard CAHPS measures as dependent variables: five global ratings (overall care, personal doctor, specialist physician, Medicare FFS, and prescription drug plan) scored 0 (worst possible rating) to 10 (best); and five composite measure scores (Doctor Communication, Care Coordination, Getting Needed Care, Getting Care Quickly, and Getting Needed Drugs) transformed to a 0–100 scale. The Care Coordination composite was in CAHPS only since 2012; analyses of this measured therefore included a smaller sample. Self-reported comorbidities, general health status, and mental health status were also obtained from the CAHPS survey. Individuals with missing responses for a CAHPS rating or composite were excluded from analysis of that measure. Technical specifications of the Medicare CAHPS are available at https://www.ma-pdpcahps.org/en/.
Statistical Analyses
To account for survey design, we used SAS 9.4 to produce survey-weighted frequencies and used SUDAAN for SAS to conduct chi-squared analyses. SUDAAN’s PROC CROSSTAB and PROC DESCRIPT allowed us to adjust models for CAHPS survey weights and FIPS state sampling units. Weights provided by NCI with these data were used to generate nationally-representative estimates. T-tests were used to compare means; Wilcoxon rank-sum tests were used to compare medians. We ran multivariable linear regression models for each of the ten CAHPS outcome measures using SUDAAN’s PROC REGRESS including a control variable for cancer type, and separate analyses for three subgroups (breast cancer, prostate cancer, and all other cancer survivors). To control for other patient characteristics potentially affecting CAHPS ratings/scores, models adjusted for sex, age, education, race, Medicare/Medicaid dual-enrollee status, self-reported general-health status, self-reported mental-health status, cancer site, cancer stage, and years since diagnosis (1–2 years, 2–5 years, 5+ years). We also adjusted for total number of physician encounters (in quartiles) during the study period. Self-reported comorbidity information was available for the SEER-CAHPS population only from 2008 onward; we included this information in Table 1 but not in regression analyses. As each CAHPS measure involved three different comparisons (Shared Care vs. each of Oncologist-led, PCP-led, and Other), the threshold for statistical significance was reduced by a factor of three to 0.017 (0.05 divided by 3).
RESULTS
Study Population Characteristics
Table 1 presents characteristics of the study population; 10,132 survivors met study inclusion criteria. Approximately 15% were classified as receiving care in a Shared Care pattern; 10% in Oncologist-led; 33% PCP-led; and 42% Other care patterns. Compared to survivors in Shared Care, those in the PCP-led and Other patterns tended to be older and diagnosed >5 years before survey participation. The PCP-led group tended to have lower education, while a greater proportion of survivors in the Other group were non-Hispanic White. Compared to survivors in Shared Care, those in Oncologist-led patterns were less likely to report diabetes or heart attack, while those in Other patterns were more likely to report Chronic Obstructive Pulmonary Disease (COPD) or heart attack. The mean number of comorbidities among survivors in Oncologist-led patterns (0.36) was significantly lower than that among survivors in Share Care (0.55) while the mean among survivors in Other patters (0.71) was significantly higher (data not shown). Among survivors in the Other pattern, the most frequent non-oncology and non-PCP specialties seen were cardiology, dermatology, urology (excluding those with prostate cancer), ophthalmology, and orthopedic surgery (data not shown). The most frequent cancer types in the study population were breast/cervical and prostate; however, the patterns of survivorship care patterns were similar across different cancer types. Compared with Share Care, the Oncologist-Led survivorship care pattern was more frequently observed in the Midwest census region and less frequently observed in the South. The Other care patterns was more frequently observed in the Northeast and less frequently observed in the Midwest and South.
Table 2 present details regarding physician evaluation and management (E&M) visits by care pattern. As expected, survivors in Shared Care had similar proportions and median numbers of oncologist and PCP visits. For survivors in Oncologist-led, PCP-led, and Other patterns, the majority of visits were with oncologists, PCPs, and other types of physicians, respectively.
Table 2.
Mean proportion and median number of oncologist, primary care, and other physician specialty E&M visits by survivorship care pattern
| Physician type | Shared Care | Oncologist-Led | PCP-Led 4 | Other Care | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Proportion of Oncologist, Primary Care, and Other Physician Specialty Visits | |||||||||||
| Mean | Std. Error | Mean | Std. Error | P-Value | Mean | Std. Error | P-Value | Mean | Std. Error | P-Value | |
| Oncologist | 34.20% | 0.44% | 79.77% | 0.64% | <.001 | 7.83% | 0.39% | <.001 | 14.25% | 0.43% | <.001 |
| Primary care | 44.79% | 0.26% | 11.05% | 0.57% | <.001 | 78.97% | 0.40% | <.001 | 25.88% | 0.44% | <.001 |
| Other physician specialty | 21.01% | 0.53% | 9.18% | 0.51% | <.001 | 13.21% | 0.44% | <.001 | 59.87% | 0.50% | <.001 |
| Median Number of Oncologist, Primary Care, and Other Physician Specialty Visits | |||||||||||
| Median | Interquartile Range | Median | Interquartile Range | P-Value | Median | Interquartile Range | P-Value | Median | Interquartile Range | P-Value | |
| Oncologist | 3 | 2 | 5 | 6 | <.001 | 0 | 1 | <.001 | 1 | 2 | <.001 |
| Primary care | 4 | 3 | 0 | 2 | <.001 | 5 | 5 | <.001 | 2 | 4 | <.001 |
| Other physician specialty | 2 | 3 | 0 | 2 | <.001 | 1 | 2 | <.001 | 6 | 6 | <.001 |
Notes:
Weighted percentages are estimates of national percentages after adjusting for sample weights and survey design.
P-values denote the statistical significance when comparing distributions within survivorship care patterns (Oncologist-Led, PCP-Led, and Other) to distribution within the Shared Care pattern.
Association of Survivorship Care Pattern and Experience of Care
Regression results examining associations of survivorship care pattern with experience of care are presented in Table 3. Each cell in Table 3 provides results from three separate CAHPS rating comparisons (Shared Care vs. Oncologist-led, PCP-led, and Other care patterns); each column therefore includes results from 30 regression models.
Table 3.
Association (regression model estimates, adjusted CI) of survivor-reported experiences of care for PCP-led, Oncologist-led, or Other survivorship care patterns vs. Shared Care (reference)#
| Study Population | |||||
|---|---|---|---|---|---|
| (1) All survivors | (2) Breast cancer survivors only | (3) Prostate cancer survivors only | (4) All survivors except breast and prostate cancers | ||
| OUTCOME MEASURE | Comparison Survivorship Care Pattern | Beta (98.3% CI) | Beta (98.3% CI) | Beta (98.3% CI) | Beta (98.3% CI) |
| Global Measures | |||||
| Rate Health Care | Oncologist-led | 0.15 (−0.07,0.37) | 0.47 (0.06,0.87) | −0.17 (−0.50,0.17) | 0.37 (0.02,0.72) |
| PCP-led | 0.09 (−0.08,0.26) | 0.25 (−0.05,0.56) | 0.04 (−0.22,0.29) | 0.02 (−0.31,0.34) | |
| Other | −0.05 (−0.21,0.11) | 0.03 (−0.29,0.36) | −0.06 (−0.29,0.18) | −0.11 (−0.41,0.20) | |
| Rate Personal Doctor | Oncologist-led | 0.00 (−0.22,0.22) | 0.17 (−0.24,0.57) | −0.08 (−0.41,0.25) | 0.05 (−0.28,0.39) |
| PCP-led | 0.03 (−0.12,0.18) | 0.07 (−0.19,0.33) | −0.03 (−0.26,0.21) | 0.02 (−0.24,0.28) | |
| Other | −0.09 (−0.23,0.06) | −0.04 (−0.32,0.24) | −0.07 (−0.29,0.15) | −0.11 (−0.36,0.13) | |
| Rate Specialist | Oncologist-led | 0.02 (−0.19,0.24) | 0.45 (0.09,0.80) | −0.28 (−0.62,0.05) | 0.21 (−0.12,0.54) |
| PCP-led | −0.03 (−0.20,0.14) | 0.21 (−0.10,0.52) | −0.18 (−0.47,0.10) | −0.01 (−0.30,0.29) | |
| Other | −0.03 (−0.18,0.12) | 0.08 (−0.25,0.41) | −0.23 (−0.45,−0.00) | 0.11 (−0.16,0.37) | |
| Rate Health Plan | Oncologist-led | 0.21 (−0.04,0.46) | 0.07 (−0.35,0.49) | 0.22 (−0.18,0.63) | 0.29 (−0.09,0.68) |
| PCP-led | 0.11 (−0.09,0.30) | −0.03 (−0.34,0.27) | 0.10 (−0.25,0.46) | 0.18 (−0.09,0.46) | |
| Other | −0.01 (−0.20,0.18) | −0.13 (−0.45,0.19) | 0.05 (−0.29,0.38) | 0.02 (−0.25,0.30) | |
| Rate Prescription Drug Plan | Oncologist-led | −0.04 (−0.42,0.34) | 0.68 (−0.15,1.52) | −0.32 (−0.96,0.31) | −0.09 (−0.67,0.49) |
| PCP-led | −0.05 (−0.44,0.35) | 0.54 (−0.15,1.24) | −0.31 (−1.08,0.46) | −0.07 (−0.55,0.41) | |
| Other | −0.18 (−0.54,0.18) | 0.09 (−0.59,0.78) | −0.12 (−0.61,0.38) | −0.27 (−0.74,0.21) | |
| Composite Measures | |||||
| Doctor Communication | Oncologist-led | −0.13 (−2.46,2.21) | 1.63 (−2.66,5.91) | −1.50 (−5.14,2.14) | 1.53 (−2.23,5.28) |
| PCP-led | −0.05 (−1.90,1.81) | 0.23 (−2.81,3.28) | −1.01 (−4.20,2.19) | 0.95 (−1.91,3.81) | |
| Other | −1.54 (−3.33,0.24) | −1.57 (−4.85,1.71) | −1.85 (−4.63,0.93) | −0.59 (−3.38,2.20) | |
| Getting Care Quickly | Oncologist-led | −0.04 (−3.50,3.42) | 4.58 (−2.76,11.92) | −2.10 (−7.11,2.92) | 0.16 (−5.81,6.13) |
| PCP-led | 0.55 (−2.02,3.12) | 3.63 (−1.48,8.74) | −1.68 (−5.78,2.41) | 0.57 (−3.73,4.87) | |
| Other | −0.88 (−3.29,1.53) | 2.54 (−2.20,7.27) | −2.16 (−5.91,1.59) | −1.41 (−5.54,2.73) | |
| Getting Needed Care | Oncologist-led | 1.25 (−1.11,3.61) | 3.34 (−0.98,7.66) | −0.86 (−4.57,2.84) | 3.40 (−0.64,7.44) |
| PCP-led | 0.63 (−1.34,2.61) | 2.26 (−1.26,5.77) | −0.89 (−4.17,2.39) | 1.11 (−2.32,4.53) | |
| Other | −1.18 (−3.06,0.69) | −0.49 (−4.18,3.20) | −2.11 (−5.00,0.78) | −0.53 (−3.85,2.78) | |
| Getting Needed Drugs | Oncologist-led | −1.95 (−4.07,0.17) | −0.34 (−5.01,4.33) | −3.63 (−6.82,−0.44) | −0.91 (−4.39,2.56) |
| PCP-led | −2.15 (−4.13,−0.18) | 0.80 (−2.52,4.12) | −4.18 (−8.14,−0.22) | −2.28 (−5.20,0.65) | |
| Other | −1.98 (−3.43,−0.53) | 0.52 (−2.95,3.98) | −2.70 (−4.67,−0.73) | −2.44 (−4.91,0.04) | |
| Care Coordination | Oncologist-led | −0.76 (−4.67,3.15) | 2.16 (−7.14,11.46) | 1.34 (−3.86,6.53) | −3.03 (−10.36,4.29) |
| PCP-led | 1.01 (−1.97,3.98) | 0.01 (−5.92,5.94) | 1.26 (−3.20,5.73) | 0.66 (−4.79,6.12) | |
| Other | −1.00 (−3.85,1.85) | −6.02 (−12.16,0.13) | 1.07 (−3.01,5.14) | −1.25 (−6.64,4.15) | |
Notes:
Negative regression coefficients (betas) indicate significantly lower ratings from survivors in the specific care pattern vs. survivors in Shared Care; positive betas indicate higher rating from survivors in the specific care pattern vs. those in Shared Care.
In each model-specific column, regression parameter estimates (betas) and 98.3% confidence intervals (CIs, adjusted for p-value of 0.017) for survivors in PCP-led, Oncologist-led (ONC-led), or Other survivorship care patterns compared with Shared Care for each experience of care outcome measure are presented. Statistically significant (with significance set at P < .017 to correct for multiple comparisons on each measure) regression results are bolded.
All models adjusted for age, race/ethnicity, educational attainment category, Medicare/Medicaid dual enrollment status, self-reported general and mental health status, cancer stage, years since diagnosis category, and number of physician encounters (by quartile, where a higher quartile indicates a greater number of encounters). The all survivors model (model no. 1) also adjusts for sex and cancer site. The model excluding breast and prostate cancer survivors (model no. 4) adjusts for sex but not cancer site.
Survivors in Shared Care had similar experience of care ratings to those in Oncologist-led, PCP-led, and Other patterns. We observed few significant differences in experience of care between groups. Compared with survivors in Shared Care, those in the Other pattern had significantly lower ratings for Getting Needed Drugs (regression coefficient [beta] −1.97, 95% CI [−3.17, −0.79]). Survivors in the PCP-led pattern also had significantly lower ratings for Getting Needed Drugs (−2.15 [−3.78, −0.53]).
Separate analyses were performed for the two largest subgroups of survivors: men with prostate cancer and women with breast cancer (Table 3). Among prostate cancer survivors, those in Other patterns had lower ratings for specialist care (−0.23 [−0.41, −0.04]). Prostate cancer survivors in non-Shared Care patterns all had significantly lower ratings for Getting Needed Drugs (Oncologist-led: −3.63 [−6.25, −1.01]; PCP-led: −4.18 [−7.44, −0.93]; Other: −2.70 [−4.31, −1.08]) than did those in Shared Care. Among breast cancer survivors, those in the Oncologist-led pattern had significantly higher ratings for overall care (0.47 [0.13, 0.80]) and specialist care (0.45 [0.16, 0.74]). Among survivors of all other cancer types (except breast and prostate cancers) combined, the only significant difference was those in the Oncologist-led pattern had higher ratings for overall care (0.37 [0.08, 0.66]).
Association of Survivorship Care Pattern and Experience of Care by Disease Stage and among Recently Diagnosed Survivors
For risk-stratified survivorship care, there may be differences by stage at diagnosis for survivors in Shared Care vs. other survivorship care patterns. While stage was controlled for in regression analyses (Table 3), regression analyses were also performed examining the association of survivorship care pattern and experience of care for patients with local-stage disease at diagnosis and separately those with regional-stage disease at diagnosis. Results are presented in Appendix Tables 1 and 2, respectively. Similar to results presented in Table 3 (with the combined population of patients with local and regional-stage disease, and regression analyses controlling for stage at diagnosis), there were few significant associations between survivorship care pattern and experience of care. Among the overall survivor populations with local-stage disease (Appendix Table 1) or regional-stage disease (Appendix Table 2), there were no significant associations between survivorship care pattern and experience of care.
Medicare CAHPS does not specifically request information on experience with cancer care. While analyses presented in Table 3 control for years since diagnosis, we separately examined associations of survivorship care patterns and experience of care for more recently diagnosed survivors (those diagnosed within two years of survey completion), who may focus more on cancer-related care in providing ratings. Results of this subgroup analysis are presented in Appendix Table 3. As with the main results, the majority of associations between survivorship care plan and experience of care ratings are not statistically significant. However, among all survivors who completed a CAHPS survey within 2 years of diagnosis, those receiving PCP-led care or Other care reported significantly worse ratings for Getting Needed Drugs than did those in Shared Care; there was no significant difference between survivors in Shared Care vs. those in Oncologist-Led care. Among breast cancer survivors diagnosed within 2 years of survey completion, only those in PCP-Led care reported significantly worse ratings for Getting Needed Drugs than did those in Shared Care. Among recently diagnosed prostate cancer survivors, those in Oncologist-Led care or Other care patterns reported worse ratings for Getting Needed Drugs than did those in Shared Care.
Sensitivity Analysis
As discussed in the Methods section, we performed sensitivity analyses by varying threshold definitions for categorizing survivorship care patterns. The sensitivity analyses used a different threshold of oncologist and primary care physician visits for classification of individuals as receiving Shared Care; this alternate definition of Shared Care resulted in approximately 6% of survivors being re-categorized from Shared Care to the Other survivorship care pattern. In models of all survivors using this alternative definition, there were no significant associations between survivorship care pattern and survivors’ experience of care ratings (data not shown).
DISCUSSION
To our knowledge, this is the first study using national data to examine associations between multiple domains of cancer survivors’ experience of care and type of physicians providing their care (i.e., survivorship care patterns). We found few significant associations between experience of care and survivorship care pattern in base-case analyses and no significant associations in sensitivity analyses. The lack of significant differences in patient experience of care associated with survivorship care patterns is surprising based on prior studies of patient preference for health care providers to lead survivorship care. Previous studies suggest that cancer survivors in the U.S. may prefer survivorship care led by oncologists rather than PCPs.[18–20] In our study, breast cancer survivors in Oncologist-led care patterns did report significantly higher ratings for their specialist physicians, and all other cancer survivors (excluding breast and prostate cancer survivors) in Oncologist-led care reported higher ratings for their overall health care compared with those in Shared Care (Table 3). However, the magnitude of these difference is small. The general lack of differences in experience of care ratings for survivors in Oncologist-led versus PCP-led versus Shared Care patterns suggests that survivors’ preferences may be flexible regarding type(s) of physician(s) leading their follow-up care.
The presence of multiple physician specialties in survivorship shared care could create challenges for care coordination.[17, 8] However, in the present study, ratings for care coordination among survivors in Shared Care did not significantly differ from those in Oncologist-led or PCP-led patterns. Sisler et al. similarly reported that ratings for continuity of care were similar among CRC survivors whether or not an oncologist was included in follow-up care.[21] Hudson et al., in evaluating five survivor-reported measures of care coordination, found that PCPs scored higher than oncologists in keeping track of survivor’s health care, whereas oncologists scored higher for communicating with other health care professionals; for three other measures, there were no significant differences between PCP and oncologists.[22]
Our findings may have implications regarding risk-stratified care patterns for cancer survivors.[23–25] However, few significant associations between experience of care and survivorship care pattern were observed in separate analyses of those with local vs. regional stage disease (Appendix Tables 1 and 2). More research is needed to examine whether survivors’ treatment histories and risk factors influence association between care pattern and care experience.
An unexpected finding presented in this study is the high proportion of survivorship care that is provided by physician specialists other than oncologists or primary care physicians. As presented in Table 1, 42% of survivors are in the Other care pattern. That does not mean that these survivors received no care from primary care physicians or oncologists, but that physicians of other specialties provided a majority of their care. Comprehensive and patient-centered survivorship care will need to go beyond the oncology-primary care collaboration that are often the focus of shared care and include a multi-disciplinary team. More research is needed to examine the role of other health care providers in survivorship care.
This study has several limitations. First, as with any study using retrospective, cross-sectional data, we were limited to variables in the data sets and could not explore changes in individuals’ experience of care over time. The study population was limited to individuals living in states participating in SEER who completed a Medicare FFS CAHPS survey; results may not be generalizable to all cancer survivors. In determining survivorship care pattern, we excluded visits with physician specialty listed as “Multispecialty Clinics” or “Group Practice”; it is possible these visits represented shared care. We also excluded visits with advanced practice providers (NPs or PAs) who were not affiliated with a physician practice, as information on the specialty of these advanced practice providers was not available. Although the Shared Care group included both PCP and oncologist visits, we do not have data on coordination/communication between providers, i.e., whether cancer care was actually shared or what was discussed during a visit. In addition, survivors may reflect on their care in general and may not provide ratings specific to their survivorship care.
Analyses included all physician visits to classify survivorship care patterns. We did not attempt to classify which visits specifically involved “survivorship care”, as any health care encounters with cancer survivors may be considered (at least in part) to include survivorship care. Data on comorbidities were available only for individuals completing the SEER-CAHPS survey in 2008 or later and were therefore not included in regression analyses.
While few significant results were observed in base-case analyses, the absence of any significant findings in sensitivity analyses suggests that results are influenced by how care patterns are defined. Future studies should explore definitions of survivorship care patterns to identify the most evidence-based ways of characterizing these patterns. In addition, survivors were not randomly assigned to survivorship care patterns. Specialties of physicians seen by survivors may reflect individual preferences, demographic/clinical characteristics, or resources in their care setting, and thus may not be unbiased. However, by controlling for survivors’ demographic and clinical characteristics, we were able to mitigate for such potential confounders in multivariable analyses.
Despite these limitations, our findings provide important information regarding shared care for cancer survivors. A variety of stakeholders has called for increased use of survivorship shared care for certain survivor populations.[26–29, 2] However, there is little evidence that shared care improves outcomes. Our study, using a data-driven definition of shared care, found few differences in experience of care for older Medicare cancer survivors receiving shared care versus PCP-led or oncologist-led care. This suggests that shared care may not provide patient-perceived benefits to survivors compared with care led by only PCPs or oncologists. Shared care may also increase provider burden and costs. Without stronger supporting evidence, efforts to facilitate or mandate shared care for survivors may result in limited benefits.
Supplementary Material
Acknowledgement:
This study was supported by funding from the National Cancer Institute (contract number HHSN261201000166U and HHSN261201500132U). Analyses presented in this manuscript were initiated when MTH was an employee of RTI International and EEK was an employee of ICF, Inc.
Footnotes
Publisher's Disclaimer: This Author Accepted Manuscript is a PDF file of a an unedited peer-reviewed manuscript that has been accepted for publication but has not been copyedited or corrected. The official version of record that is published in the journal is kept up to date and so may therefore differ from this version.
The authors indicate no conflicts of interest. Disclaimers: The article was prepared as part of some of the authors’ (MTH, MAM, EEK) official duties as employees of the US Federal Government. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the National Cancer Institute.
References
- 1.American Cancer Society. Cancer Treatment & Survivorship Facts & Figures 2019–2021. Atlanta: American Cancer Society, 2019. [Google Scholar]
- 2.Hewitt M, Greenfield S, Stovall E, editors. From cancer patient to cancer survivor: lost in transition. Washington, DC: National Academies Press; 2005. [Google Scholar]
- 3.Viswanathan M, Halpern M, Swinson Evans T, Birken SA, Mayer DK, Basch E. AHRQ Comparative Effectiveness Technical Briefs. Models of Cancer Survivorship Care Rockville (MD): Agency for Healthcare Research and Quality (US); 2014. [PubMed] [Google Scholar]
- 4.Institute of Medicine. Implementing Cancer Survivorship Care Planning: Workshop Summary. Washington, DC: The National Academies Press; 2007. [Google Scholar]
- 5.Yu X, McBean AM, Virnig BA. Physician visits, patient comorbidities, and mammography use among elderly colorectal cancer survivors. Journal of cancer survivorship : research and practice. 2007;1(4):275–82. doi: 10.1007/s11764-007-0037-7. [DOI] [PubMed] [Google Scholar]
- 6.Snyder CF, Earle CC, Herbert RJ, Neville BA, Blackford AL, Frick KD. Trends in follow-up and preventive care for colorectal cancer survivors. Journal of general internal medicine. 2008;23(3):254–9. doi: 10.1007/s11606-007-0497-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Snyder CF, Frick KD, Herbert RJ, Blackford AL, Neville BA, Carducci MA et al. Preventive care in prostate cancer patients: following diagnosis and for five-year survivors. Journal of cancer survivorship : research and practice. 2011;5(3):283–91. doi: 10.1007/s11764-011-0181-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Weaver KE, Aziz NM, Arora NK, Forsythe LP, Hamilton AS, Oakley-Girvan I et al. Follow-Up Care Experiences and Perceived Quality of Care Among Long-Term Survivors of Breast, Prostate, Colorectal, and Gynecologic Cancers. Journal of Oncology Practice. 2014;10(4):e231–e9. doi: 10.1200/JOP.2013.001175. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Wattchow DA, Weller DP, Esterman A, Pilotto LS, McGorm K, Hammett Z et al. General practice vs surgical-based follow-up for patients with colon cancer: randomised controlled trial. British Journal of Cancer. 2006;94(8):1116–21. doi: 10.1038/sj.bjc.6603052. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Cannon AJ, Darrington DL, McIlvain HE, Bauer LK, Vose JM, Armitage JO et al. Association of number of follow-up providers with outcomes in survivors of hematologic malignancies. Leukemia & lymphoma. 2010;51(10):1862–9. doi: 10.3109/10428194.2010.510896. [DOI] [PubMed] [Google Scholar]
- 11.Emery JD, Jefford M, King M, Hayne D, Martin A, Doorey J et al. ProCare Trial: a phase II randomized controlled trial of shared care for follow-up of men with prostate cancer. BJU international. 2017;119(3):381–9. doi: 10.1111/bju.13593. [DOI] [PubMed] [Google Scholar]
- 12.Grunfeld E, Fitzpatrick R, Mant D, Yudkin P, Adewuyi-Dalton R, Stewart J et al. Comparison of breast cancer patient satisfaction with follow-up in primary care versus specialist care: results from a randomized controlled trial. The British journal of general practice : the journal of the Royal College of General Practitioners. 1999;49(446):705–10. [PMC free article] [PubMed] [Google Scholar]
- 13.Chawla N, Urato M, Ambs A, Schussler N, Hays RD, Clauser SB et al. Unveiling SEER-CAHPS(R): a new data resource for quality of care research. Journal of general internal medicine. 2015;30(5):641–50. doi: 10.1007/s11606-014-3162-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.National Cancer Institute. SEER-CAHPS Linked Data Resource. 2017. https://healthcaredelivery.cancer.gov/seer-cahps/. Accessed January 8, 2020.
- 15.Halpern MT, Urato MP, Kent EE. The health care experience of patients with cancer during the last year of life: Analysis of the SEER-CAHPS data set. Cancer. 2017;123(2):336–44. doi: 10.1002/cncr.30319. [DOI] [PubMed] [Google Scholar]
- 16.Halpern MT, Urato MP, Lines LM, Cohen JB, Arora NK, Kent EE. Healthcare experience among older cancer survivors: Analysis of the SEER-CAHPS dataset. Journal of geriatric oncology. 2018;9(3):194–203. doi: 10.1016/j.jgo.2017.11.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Neuman HB, Schumacher JR, Schneider DF, Winslow ER, Busch RA, Tucholka JL et al. Variation in the Types of Providers Participating in Breast Cancer Follow-Up Care: A SEER-Medicare Analysis. Annals of surgical oncology. 2017;24(3):683–91. doi: 10.1245/s10434-016-5611-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Mao JJ, Bowman MA, Stricker CT, DeMichele A, Jacobs L, Chan D et al. Delivery of survivorship care by primary care physicians: the perspective of breast cancer patients. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2009;27(6):933–8. doi: 10.1200/jco.2008.18.0679. [DOI] [PubMed] [Google Scholar]
- 19.Mayer EL, Gropper AB, Neville BA, Partridge AH, Cameron DB, Winer EP et al. Breast cancer survivors’ perceptions of survivorship care options. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2012;30(2):158–63. doi: 10.1200/jco.2011.36.9264. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Nyarko E, Metz JM, Nguyen GT, Hampshire MK, Jacobs LA, Mao JJ. Cancer survivors’ perspectives on delivery of survivorship care by primary care physicians: an internet-based survey. BMC family practice. 2015;16:143. doi: 10.1186/s12875-015-0367-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Sisler JJ, Taylor-Brown J, Nugent Z, Bell D, Khawaja M, Czaykowski P et al. Continuity of care of colorectal cancer survivors at the end of treatment: the oncology-primary care interface. Journal of cancer survivorship : research and practice. 2012;6(4):468–75. doi: 10.1007/s11764-012-0235-9. [DOI] [PubMed] [Google Scholar]
- 22.Hudson SV, Ohman-Strickland PA, Bator A, O’Malley D, Gundersen D, Lee HS et al. Breast and prostate cancer survivors’ experiences of patient-centered cancer follow-up care from primary care physicians and oncologists. Journal of cancer survivorship : research and practice. 2016;10(5):906–14. doi: 10.1007/s11764-016-0537-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Oeffinger KC, McCabe MS. Models for delivering survivorship care. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2006;24(32):5117–24. doi: 10.1200/jco.2006.07.0474. [DOI] [PubMed] [Google Scholar]
- 24.Loonen JJ, Blijlevens NM, Prins J, Dona DJ, Den Hartogh J, Senden T et al. Cancer Survivorship Care: Person Centered Care in a Multidisciplinary Shared Care Model. International journal of integrated care. 2018;18(1):4. doi: 10.5334/ijic.3046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Nekhlyudov L, O’Malley DM, Hudson SV. Integrating Primary Care Providers in the Care of Cancer Survivors: Gaps in Evidence and Future Opportunities. The Lancet Oncology. 2017;18(1):e30–e8. doi: 10.1016/S1470-2045(16)30570-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Kinahan KE, Kircher S, Altman J, Rademaker A, Salsman JM, Didwania A et al. Promoting the Shared-Care Model for Adolescent and Young Adults With Cancer: Optimizing Referrals and Care Coordination With Primary Care Providers. Journal of the National Comprehensive Cancer Network : JNCCN. 2017;15(1):38–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Cavallo J Building a Multidisciplinary Model of Care for Cancer Survivors. The ASCO Post. [Google Scholar]
- 28.Oeffinger KC. Shared care of adult survivors of childhood cancers. The Lancet Oncology. 2008;9(3):191–3. doi: 10.1016/s1470-2045(08)70044-5. [DOI] [PubMed] [Google Scholar]
- 29.Koczwara BW, Kate; Jon Emery; Geoff Mitchell; Danielle Mazza; Patsy Yates; Ferguson Julia F;. Principles Statement: Shared Care. In: Cancer Research in Primary Care. 2016. http://pc4tg.com.au/wp-content/uploads/2016/07/PC4-Principles-Statement-shared-care-2016-1.pdf. Accessed January 8, 2020. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.