• long-term cancer survivor;
  • multidimensional survivorship care;
  • survivorship care plans;
  • primary care physicians;
  • physician survey


  1. Top of page
  2. Abstract
  6. Conflict of Interest Disclosures
  7. References


The number of long-term US cancer survivors is expected to double by the year 2050. Although primary care physicians (PCPs) provide the majority of care for long-term cancer survivors, to the authors' knowledge, few data to date have detailed PCP practice patterns, attitudes, and challenges in caring for long-term cancer survivors.


Self-administered surveys were mailed to 406 community- and academic-based general internal medicine physicians in Denver, Colorado. Survey development included in-depth physician interviews and pretesting. Of the 299 responses, 72 were ineligible; an analysis of the data from 227 surveys is presented.


The response rate was 76%. Community-based PCPs comprised 70% of completed surveys. Reported care patterns were assessed to create a multidimensional care score reflecting levels of attention to 4 areas of survivorship care: monitoring for cancer recurrence, management of late effects, sexual functioning, and mental health. Only 24% of PCPs met criteria for routinely providing more multidimensional survivorship care. More recent medical school graduates reported providing less multidimensional survivorship care when compared with their more experienced colleagues. Approximately 82% of PCPs believed that primary care guidelines for adult cancer survivors are not well defined, and 47% of PCPs cited inadequate preparation and lack of formal training in cancer survivorship as a problem when delivering care to long-term survivors.


Although PCPs provide the bulk of care for long-term survivors within the survivorship phase of the cancer trajectory, only a small subset have reported providing multidimensional survivorship care. Results underscore a need for substantially increased training in survivorship care to support the delivery of multidimensional primary care for long-term survivors. Cancer 2009;115(18 suppl):4409–18. © 2009 American Cancer Society.

Greater than 67% of individuals diagnosed in 2009 with cancer can expect to survive at least 5 years, thus joining the current ranks of the 11 million long-term cancer survivors in the United States.1 With the number of cancer survivors expected to grow dramatically in the next decades,2 attention is increasingly turning toward the unique and complex challenges that characterize the survivorship phase of the cancer trajectory.3 The challenges for long-term survivors include a broad range of medical and psychosocial late effects that call for comprehensive screening, evaluation, and management.4 During the post-treatment phase of care, as survivors transition away from intensive oncology, care primary care physicians (PCPs) play a critical role in caring for long-term survivors.

Several recent reports have acknowledged the importance of the role of PCPs in survivorship care5, 6 and advocated the value of sharing care of survivors between oncologists and PCPs.7-9 The Institute of Medicine's (IOM) 2006 report, From Cancer Patient to Cancer Survivor: Lost in Transition, described the coordination of care between cancer specialists and PCPs as one of the essential components of survivorship care. In studies of follow-up care for long-term survivors, recent investigations examined the role that PCPs play in specific groups of cancer survivors such as breast cancer and colorectal cancer survivors in the context of both monitoring for cancer recurrence as well as preventive care.10, 11 In light of growing acknowledgment that PCPs and oncologists need to act as team players,6 several understandable questions arise regarding how such a team might function. For example, who is responsible for coordinating the team and who takes the lead on delegating tasks? Should PCPs along with oncologists and patients all function as co-captains, and if so, who else would be on the clinical team? Although the notion of coordinated care is underscored in the IOM report, the report also notes that there are various ways in which survivorship care may be delivered and little evidence as to which strategies are best. Not surprisingly, there is now tremendous focus on the need to develop new paradigms for survivorship care planning, such as the need to establish survivorship care plans that can serve to document and connect the phase of successful cancer treatment with a phase of multidisciplinary follow-up care.12 Survivorship care plans, including a comprehensive care summary and follow-up plan, have been recommended by both the IOM and the President's Cancer Panel as a key tool for informing PCPs in the care needs of survivors after their cancer treatment.

To the best of our knowledge, to date, most of the discussion concerning survivorship care has been driven almost exclusively from the perspective of oncologists. Given the role that PCPs play in survivorship care, it is striking that perceptions of PCPs regarding caring for long-term cancer survivors, their attitudes about providing this care, and the barriers or challenges in caring for survivors have not been well-described. Although the need for survivorship care planning has been emphasized,13 there is in addition a lack of information that directly addresses PCP preferences concerning what information would be most useful in the construction of comprehensive care summaries and follow-up patient care plans. One of the few studies conducted with PCPs regarding cancer survivorship focused on their views regarding the transfer of care for breast and colorectal cancer patients from oncologist to primary care follow-up care.14 Greater than half of the PCPs in the study rated the transfer of care as fair or poor, and a majority of PCPs were uncertain about the appropriate frequency and duration of surveillance testing for cancer recurrence.

Despite the clearly growing consensus regarding the importance of PCPs in survivorship care, there remains a distinct lack of clarity concerning the role that the PCP should play, and how PCPs view their role and their preparedness. The current PCP survey was designed to further our understanding of the challenges in survivorship care from the PCP's point of view and to investigate how factors such as practice setting and PCP experience may be associated with the PCP's attitudes and approaches to survivorship care. Specifically, we hypothesized that PCPs practicing in an academic setting and PCPs who completed their medical training more recently, would believe more prepared to manage long-term effects of treatment and be more likely to provide the multidimensional care (MDC) necessary to address the specific needs of cancer survivors. We also hypothesized that PCPs who spend more hours in direct patient care would be more likely to provide multidimensional survivorship care.


  1. Top of page
  2. Abstract
  6. Conflict of Interest Disclosures
  7. References

Survey Variables

A survey instrument was developed using interviews with PCPs and was revised after pretesting with both academic- and community-based PCPs. The survey covered the following domains: personal and professional characteristics, including age, sex, and years in clinical practice; practice characteristics; attitudes regarding survivorship care; perceived barriers to survivorship care; perceptions of preparedness to evaluate and manage long-term effects; routine survivor care practices; and preferences for receiving survivorship information. The survey consisted of 45 multiple choice items presented in a 5-page printed color booklet entitled, “Caring for Cancer Survivors: Challenge and Opportunity.” The front cover of the survey included brief instructions, as well as a single eligibility item asking respondents if they provided ‘direct outpatient primary care.’ Respondents who answered “no” were instructed to stop and return the uncompleted survey.

PCP attitudes regarding survivorship care were evaluated by means of 3 questions, each rated on a 5-point Likert scale (Strongly Disagree to Strongly Agree). Perceived barriers to care were evaluated using 10 items rated on a 4-point scale (No Problem to Big Problem). PCPs were asked if they were aware of the 2006 IOM report. Preparedness to evaluate and manage long-term effects of cancer treatment were evaluated with 2 statements, rated on a 4-point scale (Very Unprepared to Very Prepared), and routine survivorship care practices were evaluated by means of 5 items, rated on a 4-point scale (Never to Always). PCP preferences for receiving information concerning survivorship care were evaluated with 8 items, each rated on a 4-point scale (Very Helpful to Not at All Helpful).

Practice characteristics, PCP attitudes, perceived barriers to care, and perceived preparedness were analyzed as categorical outcomes and frequencies and 95% confidence intervals (95%percnt; CIs) were reported. The years in practice variable was dichotomized (≥10 years vs <10 years). Time spent in direct patient care was dichotomized using the observed median (≥32 hours/week vs <32 hours/week). For the preparedness variables, response categories were collapsed to create a dichotomous variable Prepared/Unprepared. The barriers to care variables were also dichotomized into Moderate/Big Problem versus No/Small Problem.

Survey Administration

Details of the survey methodology have been reported elsewhere,15 and are briefly summarized here. The survey frame was the Division of General Internal Medicine, Department of Internal Medicine at the University of Colorado, Denver School of Medicine. Contact information for potentially eligible participants was taken from 2 departmental mailing lists: 1) all regular General Internal Medicine faculty practicing in university clinics (excluding hospitalists) (n = 126) and 2) all community-based affiliates who maintain unpaid clinical appointments with the internal medicine department as part of their role of precepting or teaching medical students or residents (n = 280). All PCPs were sent a survey packet including the survey, a cover letter, postage-paid return envelope, and an incentive by means of Federal Express second-day air delivery. The cover letter informed participants that the survey was for PCPs and requested they return the completed paper survey or complete the survey on-line at a URL provided in the letter. As part of a separate substudy examining monetary versus nonmonetary incentives,15 potential participants were randomized to 1 of 3 incentive conditions: 1) a $20 bill; 2) a USB flash drive; and 3) both a $20 bill and a flash drive. Potential participants with a known E-mail address were sent an E‒mail announcement containing an abbreviated version of the cover letter as well as an invitation to complete the survey online by selecting a Web-link provided in the E-mail. Approximately 2 weeks after the initial mailing, a reminder letter and a second copy of the survey were sent to all nonresponders by first class US mail. A reminder E-mail was also sent to all nonresponders with a known E-mail address. If surveys were undeliverable due to a bad address, attempts were made to identify a current address by using the Colorado State Board of Registration in Medicine online database and a new mailing was initiated. No other tracing measures were conducted.


Surveys were mailed to 406 potential subjects. Of these, 15 surveys were undeliverable because of bad addresses. Of the remaining 319 surveys, 299 were returned, yielding a 76% response rate (Table 1). Of the 299 returned, 72 were returned by PCPs who were ineligible because they did not provide direct outpatient primary care. The results of 227 surveys are presented. Approximately 59% of the respondents were men, and 85% of PCPs were Caucasian. The great majority of the respondents were board certified in Internal Medicine, and 14% reported board certification in at least 1 other specialty, including Pediatrics and Obstetrics/Gynecology. The average number of years in practice was 21.2, and the average time spent in direct patient care was 32.8 hours/week.

Table 1. Characteristics of Primary Care Provider Participants
  1. SD indicates standard deviation.

Gender (N = 224)  
Race (N = 222)  
 Black/African American31.4
 Hispanic or Latino52.3
Years in practice (N = 227)  
 Mean (SD)21.210.4
 <10 y3816.7
 ≥10 y18983.3
Time spent in direct patient care (h/wk) (N = 226)  
 Mean (SD)32.816.2
Board certification (N = 223)  
 Internal medicine only19185.7
 ≥2 specialties3214.3

PCPs were also asked about addressing important individual components of survivorship care: monitoring for cancer recurrence, screening for medical late effects, sexual dysfunction, and mental health issues. For each of these individual components of survivorship care, frequencies for delivery of care were determined. Furthermore, a dichotomous MDC variable was created to reflect levels of attention to these 4 components of survivorship care. For each component, categories were collapsed to create a dichotomous variable to reflect delivery of care for that component (Never or Rarely vs Usually or Always). Prior to collapsing categories, distributions were examined to confirm appropriateness of collapsing categories. Items were collapsed because there were a small number of responses in the intermediate categories and we were able to retain the majority of the information by collapsing them. Delivering MDC was defined as consistent delivery of all 4 components.

Statistical Analysis

To evaluate the hypothesis that PCPs practicing in an academic setting and PCPs who completed their medical training more recently were more likely to deliver MDC, associations with practice setting (academic-based vs community-based) and years in practice were examined. Time spent in direct patient care was also examined. Univariate associations were evaluated using the Fisher exact test. In addition, these variables were included in a multivariate logistic regression model of factors associated with providing MDC. Odds ratios (OR) and 95% CIs were determined, and P values in the regression model refer to the Wald chi-square test. All tests were conducted at a .05 level of significance without adjustment for multiple testing.


  1. Top of page
  2. Abstract
  6. Conflict of Interest Disclosures
  7. References

Attitudes Toward and Preparedness for Providing Survivorship Care

Table 2 presents the attitudes of PCPs toward certain aspects of cancer survivorship care. Nearly one-quarter (24%) of PCPs were unsure of whether late effects of cancer treatment are usually minimal. In response to the question regarding whether late effects of treatment should be managed primarily by oncologists, 58% of PCPs disagreed. When asked whether they agree that primary care guidelines are not well-defined for cancer survivors, the majority (82%) of PCPs agreed, with 21% of PCPs strongly agreeing. Furthermore, when asked about the frequently cited IOM report describing the healthcare needs of cancer survivors,2 >90% of the PCPs answered that they were not aware of this report.

Table 2. Attitudes of Participants Regarding Survivorship Care and Preparedness to Evaluate and Manage Late Effects
  1. IOM indicates the Institute of Medicine.

Late effects of treatment are usually minimal and typically don't require medical intervention (N = 218) 
 Strongly disagree7.3
 Strongly agree0.9
Late effects of adult cancer should primarily be managed by oncologists (N = 218) 
 Strongly disagree6.0
 Strongly agree1.8
Primary care guidelines for adult cancer survivors are not well defined (N = 218) 
 Strongly disagree0.9
 Strongly agree21.1
Knowledge of the IOM report (N = 225) 
Preparedness to evaluate long-term effects (N = 215) 
Preparedness to manage long-term effects (N = 215) 

Preparedness and Barriers to Care

When asked about their preparedness to evaluate and manage the long-term effects of cancer treatment, nearly half of PCPs (47.9%) reported feeling unprepared to evaluate the long-term effects of adult cancer survivors and 49.8% reported feeling unprepared to manage these long-term effects (Table 2). The commonly endorsed barriers to delivering survivorship care were: lack of standards for survivorship care (52.5%); inadequate preparation/formal training around survivorship (47.2%); and limited access to mental health referrals for cancer survivors (45.7%). In addition, approximately 42% of PCPs reported a lack of time to adequately address survivorship issues and 36% of PCPs reported that inadequate access to patients' cancer treatment history is a problem (Table 3).

Table 3. Perceived Barriers to Care
Barriers to Care*Percentage95% CI
  • 95% CI indicates 95% confidence interval.

  • *

    Percentage of participants who indicated an issue was a moderate problem or big problem. Percentages are of participants for whom data were available.

Lack of standards of care for long-term adult cancer survivors52.545.7-59.3
Inadequate preparation/formal training around survivorship issues47.240.4-54.1
Limited access to mental health referrals for cancer survivors45.738.9-52.5
Lack of time to adequately address cancer survivorship issues42.435.7-49.3
Inadequate access to patients' cancer treatment history36.129.7-42.8
Patient anxiety or fears about health28.822.9-35.3
Lack of practical experience in caring for cancer survivors22.416.9-27.9
Limited access to cancer specialists when needed10.56.8-15.3
Limited access to noncancer specialists such as cardiac or endocrine specialists7.84.6-12.1
Patient reluctance to discuss previous cancer history2.30.8-5.3

Delivering Survivorship Care

Responding to questions regarding how regularly they address various aspects of multidimensional survivorship care (Table 4), 84.4% of PCPs reported that they usually or always monitor patients for cancer recurrence. Greater than one-third of PCPs (37.3%) reported never or only rarely screening for medical late effects, 62% of PCPs reported never or only rarely addressing issues of sexual dysfunction, and 33% reported never or rarely addressing mental health issues. Only 24.4% of PCPs endorsed regularly addressing all 4 of these areas, thus meeting study criteria for delivering multidimensional survivorship care.

Table 4. Multidimensional Care Components
VariableCommunity Physician, %Academic Physician, %Overall, %
N = 149N = 69*N = 218*
  • *

    For questions regarding screening for medical late effects and sexual dysfunction, there were data available for 68 academic-based physicians and a total of 217 physicians overall.

Screening for medical late effects   
 Never or rarely32.947.137.3
 Usually or always67.152.962.7
Monitoring for cancer recurrence   
 Never or rarely16.114.515.6
 Usually or always83.985.584.4
Sexual dysfunction   
 Never or rarely55.777.962.4
 Usually or always44.323.537.6
Mental health issues   
 Never or rarely34.230.433.0
 Usually or always65.869.667.0
Multidimensional care   
 Yes (address all 4 issues)30.211.824.4
 No (<4 issues addressed)69.888.275.6

In multivariate analyses, PCP experience was found to be significantly associated with consistency of delivering MDC (Table 5). In contrast to expectations, PCPs in practice ≥10 years were 7 times more likely to deliver MDC (OR, 7.17; 95% CI, 1.7-30.9 [P = .008]) than PCPs with <10 years in practice. As expected, those reporting at least 32 hours of direct patient care per week were nearly twice as likely to deliver MDC (OR, 1.98; 95% CI, 1.0-3.8 [P = .04]) as those who provided <32 hours of direct patient care. Contrary to the study hypothesis, community-based PCPs and not those in an academic-based practice reported delivering MDC more consistently—in fact, they were more than 3 times as likely to deliver MDC (OR, 3.25; 95% CI, 1.4-7.3 [P = 005]).

Table 5. Correlates and Logistic Regression Analysis of Multidimensional Care
VariableOR95% CIP
  1. OR indicates odds ratio; 95% CI, 95% confidence interval.

Univariate associations   
 Physician experience, y, ≥10 vs <107.171.7-30.9.008
 Time in direct patient care, h/wk, ≥32 vs <321.981.0-3.8.04
 Practice setting, community vs academic3.251.4-7.3.005
Multivariate model   
 Physician experience5.501.3-24.2.02
 Time in direct patient care1.470.7-3.0.30
 Practice setting2.160.9-5.3.09

Further analysis revealed that the correlation between practice setting and years in practice and the delivery of MDC were potentially confounded, becauses community practice respondents were found to be a higher average number of years in practice (mean years, 23.3) compared with academic practice respondents (mean years, 16.8; t = 4.63 [P < .0001]). Similarly, community practice respondents also reported a higher number of hours per week in direct patient care (mean number of hours, 37.5) compared with their academic peers (mean number of hours, 22.6; t = 7.82 [P < .0001]). To clarify the correlation between these PCP factors and delivery of MDC, a logistic regression was performed using years in practice, hours of direct care, and practice setting as covariates. In this model, only years in practice was found to be independently associated with delivery of MDC (OR, 5.5; 95% CI, 1.3-24.2 [P = .02]). Of note, the model indicated that community-based PCPs were more likely to deliver MDC after adjusting for years in practice, but this association was not found to be statistically significant (OR, 2.16; 95% CI, 0.9-5.3 [P = .09]).

PCP Information Preferences

When asked about their preferences, PCPs overwhelmingly endorsed several practical options for obtaining information regarding cancer survivorship care (Table 6). Approximately 98% of PCPs would like to receive additional descriptions of patients' previous cancer treatments, including therapeutic exposures, information regarding short-term complications, and individualized guidelines for care. Greater than 85% of PCPs would like to have clinical practice guidelines for adult survivorship care, both in print and online, and 81% of PCPs would like greater opportunity to consult with adult cancer survivor specialists. Approximately three-quarters of PCPs (74%) also wanted to have continuing medical education (CME) opportunities in the area of cancer survivorship.

Table 6. Information Preferences
  1. CME indicates continuing medical education.

Clinical practice guidelines for survivorship care (print) (N = 217) 
 Very/somewhat helpful92.2
 Not very/not at all helpful7.8
Clinical practice guidelines for survivorship care (online) (N = 217) 
 Very/somewhat helpful91.2
 Not very/not at all helpful8.8
Opportunities for consultations with adult cancer survivor specialists (N = 215) 
 Very/somewhat helpful85.1
 Not very/not at all helpful14.9
Description of patients' previous cancer diagnosis and treatment summary (N = 217) 
 Very/somewhat helpful96.3
 Not very/not at all helpful3.7
Description of patients' previous cancer treatment including therapeutic exposures and short-term complications (N = 216) 
 Very/somewhat helpful97.2
 Not very/not at all helpful2.8
Description of patient's previous cancer treatment including therapeutic exposures, short-term complications, and individualized guidelines for care (N = 216) 
 Very/somewhat helpful97.7
 Not very/not at all helpful2.3
Individualized patient recommendations for managing late effects of adult cancer treatment (N = 216) 
 Very/somewhat helpful96.8
 Not very/not at all helpful3.2
CME opportunities in cancer survivorship (N = 217) 
 Very/somewhat helpful77.9
 Not very/not at all helpful22.1


  1. Top of page
  2. Abstract
  6. Conflict of Interest Disclosures
  7. References

To our knowledge, the current study represents the first survey to assess the attitudes, barriers, practices, and information preferences of PCPs regarding cancer survivorship care, particularly from the perspectives of both community-based and academically based clinicians. The results of our survey indicate that PCPs need and want more preparation, training, and practical guidelines to care for their long-term cancer survivors. Although PCPs desire to be closely involved in delivering survivorship care, it was striking that nearly half of the PCPs surveyed felt unprepared to both evaluate and manage the long-term effects of their cancer survivors. This study elucidates the primary barriers that PCPs perceive in delivering care to long-term cancer survivors, including lack of preparation, time, and clinical guidelines. Results also suggest that there are distinct components of multidimensional survivorship care that are not routinely addressed for cancer survivors and that, in contrast to the hypotheses, PCPs who have finished medical training more recently may be less likely to provide multidimensional survivorship care in comparison with their more experienced colleagues.

These results suggest that there may be a fair amount of uncertainty and ambiguity for PCPs surrounding the consequences of cancer treatment and who is primarily responsible for that treatment. For example, although the majority of PCPs disagreed that late effects of cancer treatment are usually minimal, approximately 25% of PCPs neither agreed nor disagreed with this opinion. Similarly, although the majority also disagreed with the statement that the management of long-term effects of treatment should primarily be managed by oncologists, once again approximately 25% of the PCPs neither agreed nor disagreed. As has been pointed out in previous studies focusing on the long-term care of cancer survivors, there is frequently a lack of clarity regarding who provides what kinds of ongoing follow-up care to cancer survivors.7 For example, it has been shown that, although preventive care such as cholesterol screening and bone densitometry increases when survivors are followed primarily by PCPs, cancer-related screenings decrease as oncology visits decrease.16 As oncologists are becoming progressively less involved in long-term survivorship care, it is critical to establish not only what the expectations are regarding potential late effects of treatment but clear guidelines for who is responsible for evaluating and managing them. These finding certainly also support the increasingly widespread call for survivorship care plans10 that help clarify and define the roles and responsibilities of PCPs in the context of delivering survivorship care.

The results of the current study suggest that the information and guidelines currently available are not well disseminated to PCPs. The number 1 barrier to care that greater than half of the responding PCPs cited was a lack of standards of care for survivors. It is remarkable that, despite the somewhat recent development of more than 1 set of survivorship guidelines, such as those put forth by the American Society of Clinical Oncology17 as well as the National Comprehensive Cancer Network,18 most PCPs do not appear to be aware of them. Amplifying this issue, nearly all of the PCPs surveyed were not aware of the IOM report.2 Although this report is frequently referenced as a foundational document that lays out the groundwork for establishing quality follow-up for cancer survivors, the findings of the current study underscore the concern that current conversations regarding survivorship are typically driven by oncologists and do not adequately represent the perspective of PCPs. Such findings underscore the need for significantly more effective and efficient dissemination of these resources. We acknowledge that the dissemination of knowledge is a challenge, although it is possible that factors previously used to successfully spread clinical innovation across large healthcare systems could be used. For example, leadership support, staff training, and optimizing the use of organizational infrastructure to support program implementation19 are all tools that have been used previously and may be helpful in spreading knowledge of resources such as the IOM report.

In addition to better dissemination of currently available survivorship care resources, there has also been a call for the development of new resources and clinical tools such as survivorship care plans10 and interoperable health records. As previously outlined, survivorship care plans can summarize critical pieces of information needed by PCPs to collaborate in long-term survivorship care. This information includes description of cancer type, treatment history, long-term effects of cancer treatment, specific information and recommendations concerning follow-up screening, and other preventive and necessary health maintenance practices.10 Another pressing need is for the availability to exchange records between different medical systems (interoperability) that would facilitate the co-ordination of survivorship care delivery between oncologists and PCPs, especially when these clinicians work in nonaffiliated healthcare facilities. Often a cancer patient receives treatment at a cancer center that does not have any kind of shared medical record, either electronic or paper, with the patient's PCP practice. This kind of fragmented system of care is particularly problematic when there is a need not only for seamless transition but potential ongoing collaboration between PCPs and oncologists when it comes to cancer survivorship.2 In addition, it is imperative that such resources are integrated so that, for example, survivorship care plans are part of a patient's interoperable health record that can be easily accessed by various clinicians involved in survivorship care.

Respondents suggested several barriers that are problematic when delivering survivorship care. Several of these barriers not only reflect issues in delivering survivorship care (eg, inadequate access to treatment histories), but also echo many of the broader systems' issues that are well-known to current primary care medicine practices in general, such as the lack of adequate time and access to mental health referrals. Given that the typical office visit in the United States lasts approximately 16 minutes,20 it has been acknowledged that this is not nearly enough time to adequately address complicated medical problems.21 In light of the limits of the current US healthcare system, it is unlikely that PCPs will be able to routinely make changes such as seeing cancer survivors for routinely longer or more frequent visits. Thus, these findings underscore the imperative for effectively disseminating patient information such as treatment histories, guidelines, and decision aids for PCPs. These findings also echo the concern that it may be difficult for clinicians to access appropriate mental health resources for cancer survivors. If PCPs lack the resources to refer cancer survivors for mental health follow-up, it raises the likelihood that mental health may not always get adequately assessed, in part because PCPs may be less likely to dwell on a type of problem in which there are not always obvious next steps to be taken. Certainly the integration of mental health services with primary care is a serious challenge that has been widely acknowledged22 and defies easy answers. Fortunately, with regard to the needs of cancer survivors, there are several organizations that have developed highly innovative resources and easy to use educational materials for cancer survivors about mental health as well as other survivorship needs such as sexual health, employment, and health insurance. Three such organizations that provide extensive materials and resources, including mental health referrals, that can be accessed by both PCPs and patients are the Lance Armstrong Foundation,23 the American Psychosocial Oncology Society,24 and CancerCare.25

Turning to the delivery of survivorship care, we assessed the delivery of MDC based on PCP reports of whether they routinely address 4 fundamental aspects of survivorship care: screening for medical late effects, monitoring for cancer recurrence, addressing sexual dysfunction, and addressing mental health issues. Although the majority of PCPs, both community-based as well as academically based, reported that they usually or always monitor for cancer recurrence, the rates of compliance with the other 3 aspects of MDC were significantly lower. Multivariable analysis suggests that PCPs in practice >10 years were more likely to deliver MDC. Although there is clearly growing attention being paid to cancer survivorship as a distinct aspect of care, it is interesting to contemplate why younger PCPs, that is, those PCPs who have been in practice for <10 years, were significantly less likely to deliver MDC than their more experienced peers. Certainly, it is arguable that some of the aspects that we identified as key components of care, such as sexual dysfunction and mental health, can be anxiety‒provoking for clinicians to address. However, sexual dysfunction is a major source of distress to a significant number of cancer survivors.26 Moreover, it has been shown that, although patients often do not want to bring up the topic for fear of embarrassing their PCPs, they very much want to talk about this issue with their PCP.27 If PCPs do not have much experience in asking about a potentially sensitive topic such as sexual dysfunction or if they have concern that they would not know how to follow-up with a patient once they “open the door” to such a topic, it is understandable that the topic often goes unaddressed. Similarly, if PCPs do not believe that they have a referral for patients if mental health issues are revealed, it is again understandable that PCPs may be reluctant to address such an issue. Certainly these findings raise relevant questions regarding the availability of ongoing teaching and training opportunities for PCPs, especially for those who have either finished training more recently or for those who simply have less practical experience with cancer survivors.

The current study has several limitations. The PCPs surveyed were from 1 particular region of the country (metropolitan Denver) and 1 health system, which limits the generalizability of the findings. Although the response rate was significantly higher than expected compared with the typical response rate of most PCP surveys,28 it remains uncertain how representative these respondents are relative to a broader sample of PCPs from across the United States. Furthermore, this survey did not attempt to assess actual knowledge about survivorship care and it was based on self-report of behavior. However, self-reporting behavior would have more likely resulted in an under-reporting of perceived preparedness to evaluate and manage survivorship care as well as an under-reporting of addressing specific components of care.

In summary, the findings of the current study underscore both the desire for PCPs to be more involved in and knowledgeable about delivering cancer survivorship care as well as the significant amount of work to be done in helping patients make the successful transition from oncology treatment to long-term survivorship care. Because one of the major goals within the emerging field of cancer survivorship is to improve the ability of PCPs to care for long-term cancer survivors, it is particularly important to identify their challenges, preferences, and attitudes to begin developing efficient tools for PCPs that will be both useful and easily disseminated to a wide range of providers. Such intervention may focus on not only enhancing awareness of late effects and guidelines for care, but also providing skills to communicate about and manage these late effects in the medical encounter.

Conflict of Interest Disclosures

  1. Top of page
  2. Abstract
  6. Conflict of Interest Disclosures
  7. References

Cosponsored by the National Cancer Institute's Office of Cancer Survivorship, the Office of Cancer Survivorship of the Centers for Disease Control and Prevention, and the American Cancer Society's Behavioral Research Center.


  1. Top of page
  2. Abstract
  6. Conflict of Interest Disclosures
  7. References