National trends in prostate cancer screening among older American men with limited 9-year life expectancies: Evidence of an increased need for shared decision making
Published in part as an abstract in conjunction with the 2013 American Society of Clinical Oncology (ASCO) Annual Meeting; May 31 to June 4, 2013; Chicago, Illinois.
Abstract
BACKGROUND
Prostate-specific antigen (PSA) screening for prostate cancer remains controversial. Most groups recommend informed decision making for men with 10 years of remaining life expectancy. The primary objective of this observational cohort study was to investigate the association between predicted 9-year mortality and prostate cancer screening among American men aged ≥65 years in 2005 and 2010. The second objective was to analyze the proportions of men who discussed screening with their physicians.
METHODS
Data were extracted from the 2005 and 2010 National Health Interview Surveys. Men aged ≥65 years without prostate cancer were divided into predicted 9-year mortality quartiles. The proportions of men confirming a screening PSA within the prior year were determined. Logistic regression was used to compare screening rates.
RESULTS
Screening rates for men aged ≥65 years were 48% in 2005 and 48% in 2010 (P = .9). Men ages 65 to 74 years who had <27% predicted 9-year mortality were most commonly screened, with 56% screened in 2010, compared with 34% of men aged ≥75 years with >75% predicted 9-year mortality. Approximately 55% of screened men aged ≥75 years who had ≥53% predicted 9-year mortality recalled discussing the advantages of screening, whereas 25% recalled discussing the disadvantages.
CONCLUSIONS
Prostate cancer screening with PSA did not differ significantly between 2005 and 2010 for men aged ≥65 years based on predicted 9-year mortality. Approximately 33% of older men with a high likelihood of 9-year mortality were screened despite minimal clinical benefit. Twice as many men recalled discussing the potential advantages of screening compared with the disadvantages. Cancer 2014;120:1491–1498. © 2014 American Cancer Society.
INTRODUCTION
In 2012, the United States Preventive Services Task Force (USPSTF) released a grade D recommendation discouraging prostate-specific antigen (PSA) screening for prostate cancer.1 The USPSTF noted that the rate of prostate cancer over-diagnosis, defined as the detection of disease that otherwise would not have shortened a man's lifespan or resulted in adverse health effects, was directly related to age at diagnosis. Consequently, older men with shorter remaining life expectancies (RLEs) are at a higher risk of over-diagnosis. In 2008, the USPSTF issued a grade D recommendation against PSA-based prostate cancer screening for men ages ≥75 years, noting a low likelihood of benefit from screening a population with an average RLE of approximately 10 years.2 Many other organizations have released guidelines emphasizing the role of life expectancy in screening decisions. For example, recommendations from the American Cancer Society (ACS) emphasized discussing PSA-based screening starting at age 50 years with men who have at least 10 years of RLE and who are at average risk of prostate cancer.3 The American Urological Association (AUA) recommended physicians engage in shared decision making with men ages 55 to 69 year but discouraged routine PSA-based screening for men aged ≥70 years. The AUA also discouraged the use of PSA screening for prostate cancer among men of any age with less than a 10-year to 15-year life expectancy.4 The National Comprehensive Cancer Network (NCCN) recommended discussing risks and benefits regarding digital rectal examinations and PSA testing starting at age 40 years. The NCCN guidelines also emphasized that life expectancy, on average, decreases to less than 10 years in American men starting at age 75 years, suggesting that there are minimal benefits of routine screening for men aged >75 years.5
The importance of age and RLE in guiding prostate cancer screening decisions was further heightened by the absence of data pertaining to older men in 2 large randomized controlled trials of PSA screening, both of which excluded men aged >74 years.6, 7 The European Randomized Study of Screening for Prostate Cancer (ERSPC) demonstrated a 20% relative reduction in prostate cancer-specific mortality benefit among men ages 55 to 69 years.7 The incidence of prostate cancer in older Swedish men at a single ERSPC site did not differ significantly between screened men and nonscreened men 9 years after screening efforts were completed.8 The United States Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial reported no mortality benefit from PSA screening among men ages 55 to 74 years, but that trial was limited because approximately 50% of men were screened before enrollment, and >50% of men in the control arm obtained a PSA level.6 Although the mortality benefits from general PSA screening are uncertain and modest at best, there are important harms associated with PSA screening, including false-positive results.6, 7 Over-treatment of prostate cancer is associated with a risk of urinary incontinence, erectile dysfunction, and bowel toxicity.9, 10
It is important to understand, given the potential benefits and harms associated with prostate cancer screening and subsequent treatments, how effectively PSA-based screening programs are targeting men who are most likely to benefit—primarily those men with longer RLE and older men who may have more aggressive cancers.11 Our primary objective was to investigate the association of estimated RLE, determined using a validated 9-year mortality index, with rates of PSA-based prostate cancer screening in men aged ≥65 years in 2005 and 2010. We used an established index for men aged ≥65 years using data regarding functional measures, illnesses, behaviors, and demographics to predict 9-year mortality.12 Because the index has been validated in men aged ≥65 years, our study was restricted to men in this age group. Our second objective was to investigate how the rates of prostate cancer screening differed among men with better or worse 9-year predicted mortality between 2005 and 2010. These years marked a time interval which witnessed the release of widely publicized guidelines, such as the USPSTF guidelines from 2008, emphasizing the importance of age and RLE in screening decisions.2 Our third objective was to examine the proportions of men aged ≥75 years who were screened for prostate cancer and recalled their physicians recommending PSA testing for prostate cancer detection. In addition, we investigated the proportion of screened men aged ≥75 years who recalled their physicians discussing the advantages and/or disadvantages of PSA-based screening for prostate cancer. We focused our investigation of shared decision making on men aged >75 years, because the balance between advantages and disadvantages of screening for prostate cancer becomes more nuanced and important for patients to understand as their RLE decreases.
MATERIALS AND METHODS
Study Population
Data were extracted from the 2005 and 2010 National Health Interview Survey (NHIS). The NHIS is a public-access database containing deidentified patient data collected from a cross-sectional, in-person household survey administered continuously throughout the year by employees of the US Census Bureau. For the current study, we analyzed data collected during 2005 and 2010. The survey provided a representative sample of the US population and was performed on a continuous basis throughout each year on approximately 87,500 individuals, with final response rates of 69% and 61% in 2005 and 2010, respectively.13 The Cancer Control Supplement contained specific questions on cancer screening and was most recently released in 2005, 2008, and 2010. Interview participants were randomly selected from the population, and it was exceedingly unlikely that an individual was included in both the 2005 and 2010 survey responses. This study was exempt from review by an institutional review board.
Data Analysis and Statistical Methods
The NHIS included variables asking men whether they had visited a medical specialist or general physician in the year before the survey. Men aged ≥65 years who confirmed they had visited either a general physician or a medical specialist in the year before the survey were included. The men were separated into 2 age groups: men ages 65 to 74 years and men aged ≥75 years. Men who had a personal history of prostate cancer were excluded. The NHIS asked men why their most recent PSA test was recorded, with possible responses including: “as part of a routine examination,” “because of a problem,” “other,” or “don't know.” We excluded men who reported a PSA in the previous year for reasons other than “as part of a routine examination” or who were missing PSA testing data in the NHIS database. The men were separated into subgroups using an established index that uses data regarding functional measures, illnesses, behaviors, and demographics to predict 9-year mortality in the NHIS database (the index is listed in Table 1).12 This index has been validated in men aged ≥65 years; therefore, we limited our analysis to men in this age group.12 Individuals were grouped into quartiles based on the predicted likelihood of 9-year mortality: <27%, from 27% to 52%, from 53% to 75%, and >75%. Adults who had a >50% risk of mortality over a given time frame generally expected are to have an RLE less than the given time frame.14 Therefore, the first 2 quartiles (<27% and 27%-52% of 9-year mortality) were categorized as >9-year estimated RLE, and the third and fourth quartiles (53%-75% and >75% 9-year mortality) were categorized as <9-year estimated RLE.
| Variable | Score: No. of Points |
|---|---|
| Age, y | |
| 65-69 | 0 |
| 70-74 | 1 |
| 75-79 | 3 |
| 80-84 | 5 |
| ≥85 | 7 |
| Sex | |
| Women | 0 |
| Men | 3 |
| Body mass index <25 kg/m2 | 2 |
| Self-reported health status | |
| Excellent/very good | 0 |
| Good | 1 |
| Fair/poor | 2 |
| Comorbidities | |
| Emphysema/chronic bronchitis | 2 |
| Cancer, excluding nonmelanoma skin cancers | 2 |
| Diabetes | 2 |
| Requires help with everyday household chores, doing necessary business, shopping, or getting around for other purposes | |
| Requires help | 2 |
| Difficulty walking a quarter mile/3 city blocks without special equipment or assistance | |
| Not difficult | 0 |
| A little difficult | 3 |
| Cannot do at all/does not walk | 3 |
| Smoking status | |
| Never smoker | 0 |
| Former smoker | 1 |
| Current smoker | 3 |
| No. of hospitalizations during previous y | |
| No hospitalizations | 0 |
| 1 Hospitalization | 1 |
| ≥2 Hospitalizations | 3 |
- a See Alemozaffar 2011.10
The primary outcome of interest was a screening PSA, defined as a PSA ordered as “part of a routine examination” in the 12 months before the survey administration. The secondary outcome of interest was the proportion of men aged ≥75 years who were screened for prostate cancer in the year before the 2010 survey who recalled discussing with their physician either the advantages, disadvantages, or controversy of using PSA testing for prostate cancer screening. The men also were asked whether their physicians ever recommended screening for prostate cancer using PSA testing. Sampling weights were used in all analyses to determine the proportion of men aged ≥65 years that received a PSA screen. Logistic regression was then used to compare the proportions of men screened in 2005 and 2010. Statistical significance was considered to be a 2-sided P value < .05. All analyses were performed using Stata version 11.0 (StataCorp, College Station, Tex).
RESULTS
The final cohort included n = 1613 (2005) and n = 1475 (2010) men. After the application of survey weights, these cohorts represented 10.3 million and 11.7 million men in 2005 and 2010, respectively (Table 2). There were no significant differences in estimated 9-year mortality between the years 2005 and 2010 for either age group (P = .6 for men ages 65-74 years; P = .7 for men aged ≥75 years).
| Probability of 9-Year Mortality | 2005 | 2010 |
|---|---|---|
| Men ages 65-74 ya | ||
| <27% | ||
| Proportion of men (95% CI), % | 54 (50.4-57.6) | 52.4 (48.7-56.1) |
| No. of men in sample | 482 | 451 |
| Survey-weighted no. of men in US population | 3,300,000 | 3,900,000 |
| 27%-52% | ||
| Proportion of men (95% CI), % | 30.7 (27.6-33.7) | 32.9 (29.1-36.7) |
| No. of men in sample | 305 | 313 |
| Survey-weighted no. of men in US population | 1,800,000 | 2,400,000 |
| 53%-75% | ||
| Proportion of men (95% CI), % | 12.3 (9.8-14.8) | 11.9 (9.6-14.2) |
| No. of men in sample | 113 | 111 |
| Survey-weighted no. of men in US population | 740,000 | 880,000 |
| >76% | ||
| Proportion of men (95% CI), % | 3 (1.8-4.2) | 2.8 (1.5-4) |
| No. of men in sample | 28 | 27 |
| Survey-weighted no. of men in US population | 180,000 | 210,000 |
| Men aged ≥75 yb | ||
| <27% | ||
| Proportion of men (95% CI), % | 6.1 (4-8.2) | 8.4 (5.9-11) |
| No. of men in sample | 37 | 43 |
| Survey-weighted no. of men in US population | 260,000 | 370,000 |
| 27%-52% | ||
| Proportion of men (95% CI), % | 38 (34.1-41.9) | 30.3 (25.8-34.7) |
| No. of men in sample | 248 | 168 |
| Survey-weighted no. of men in US population | 1,600,000 | 1,300,000 |
| 53%-75% | ||
| Proportion of men (95% CI), % | 28.6 (25-32.1) | 36.3 (32-40.6) |
| No. of men in sample | 205 | 112 |
| Survey-weighted no. of men in US population | 1,200,000 | 1,600,000 |
| >76% | ||
| Proportion of men (95% CI), % | 27.3 (24-30.7) | 25 (21.4-28.7) |
| No. of men in sample | 195 | 150 |
| Survey-weighted no. of men in US population | 1,200,000 | 1,100,000 |
- Abbreviations: CI, confidence interval.
- a P = .9 for men ages 65 to 74 years between 2005 and 2010
- b P = .9 for men aged ≥75 years between 2005 and 2010.
Overall, there were no significant differences in PSA screening between 2005 and 2010 when screening patterns were examined by age and predicted 9-year mortality for men ages 65 to 74 years (<27%, 27%-52%, 53%-75%, and >76% predicted 9-year mortality; P = .5, P = .9, P = .6, P = .8, respectively) (Fig. 1) and for men aged ≥75 years (P = .4, P = .7, P = .4, and P = .7, respectively) (Fig. 2).
The most frequently screened men were the healthiest men ages 65 to 74 years who had a <27% chance of dying in the next 9 years. In that group, 58.3% (95% confidence interval [CI], 53.6%-63.1%) were screened in 2005, and 56.1% (95% CI, 50.6%-61.5%) were screened in 2010. Less healthy men aged >65 years were screened less frequently; and, among these screened men, approximately one-quarter (26.9%; 95% CI, 10.7%-43.0%) had a ≥76% likelihood of 9-year mortality, and approximately half (58.3%; 95% CI, 53.6%-63.1%) had a ≥27% likelihood of 9-year mortality.
Of the men aged ≥75 years who had a ≥76% likelihood of 9-year mortality, 35.7% (95% CI, 28.4%-43.1%) were screened for prostate cancer in 2005, representing an estimated 420,000 screening PSA tests. In 2010, 33.7% (95% CI, 24%-43.3%) of the men in this subgroup were screened for prostate cancer, representing an estimated 367,000 screening PSA tests.
The vast majority of men aged ≥75 years who were screened for prostate cancer in 2010 remembered their physician recommending screening for prostate cancer (Table 3). Among screened men aged ≥75 years, 96.3% (95% CI, 91.1%-101.6%) of individuals with less than a 53% predicted probability of 9-year mortality recalled their physician recommending PSA screening. This proportion was 94.2% (95% CI, 90.6%-97.9%) among sicker men aged ≥75 years who had a predicted 9-year mortality rate ≥53%. However, significantly fewer men in both groups recalled being told about the specific advantages of PSA screening: 54.1% (95% CI, 38.9%-69.3%) and 55% (95% CI, 45.3%-64.8%) of men aged ≥75 years with low (<53%) or high (≥53%) predicted 9-year mortality, respectively, recalled discussing the advantages of screening with their physicians. The proportions of screened men aged ≥75 years who recalled discussing the disadvantages of PSA testing with their physicians decreased further to 21.8% (95% CI, 9.4%-34.2%) and 24.7% (95% CI, 16.5%-33.0%) for those men with low (<53%) or high (≥53%) predicted 9-year mortality, respectively. A minority of screened men aged ≥75 years with low (<53%) or high (≥53%) predicted 9-year mortality discussed the controversy of prostate cancer screening with their physicians, and an estimated 18.3% of men aged ≥75 years who had <53% predicted 9-year mortality recalled being told about the controversy of screening. This number increased to 23.1% among men aged ≥75 years who had ≥53% predicted 9-year mortality.
| Probability of 9-Year Mortality Before 2010 Survey | Physician Recommended PSA Screening | Physician Discussed Advantages | Physician Discussed Disadvantages | Physician Discussed Controversy |
|---|---|---|---|---|
| Screened men aged ≥75 y | ||||
| <53% | ||||
| Proportion of men (95% CI), % | 96.3 (91.1-101.6) | 54.1 (38.9-69.3) | 21.8 (9.4-34.2) | 18.3 (5.9-30.7) |
| Total no. of men | 52 | 29 | 11 | 9 |
| ≥53% | ||||
| Proportion of men (95% CI), % | 94.2 (90.6-97.9) | 55 (45.3-64.8) | 24.7 (16.5-33) | 23.1 (15.2-31.1) |
| Total no. of men | 159 | 85 | 41 | 33 |
| Unscreened men aged ≥75 y | ||||
| <53% | ||||
| Proportion of men (95% CI), % | 61.6 (49.3-73.9) | 25.7 (13.4-38.1) | 16.6 (7-26.2) | 13.7 (4.2-23.2) |
| Total no. of men | 33 | 16 | 10 | 8 |
| ≥53% | ||||
| Proportion of men (95% CI), % | 53.8 (47.4-60.1) | 33.9 (27.2-40.6) | 18.4 (13.2-23.6) | 12.1 (7.5-16.7) |
| Total no. of men | 145 | 87 | 48 | 31 |
- Abbreviation: CI, confidence interval.
In contrast, a narrow majority of men aged ≥75 years who were not screened for prostate cancer in 2010 recalled their physicians recommending PSA testing, with 61.6% (95% CI, 49.3-73.9) and 53.8% (95% CI, 47.4-60.1) of men aged ≥75 years who had <53% and ge;53% predicted 9-year mortality, respectively, recalling that their physician recommended screening. A minority of these same unscreened groups recalled their physicians discussing the advantages, disadvantages, or controversy of prostate cancer screening using PSA testing (Table 3).
DISCUSSION
Despite the increased emphasis placed on limiting prostate cancer screening in older men and those with limited 10-year life expectancies according to the 2008 USPSTF recommendation, we observed that PSA screening patterns among men aged ≥65 years, when analyzed on the basis of predicted 9-year mortality, did not change significantly between 2005 and 2010. We observed that 41% of men aged ≥75 years who had >52% predicted 9-year mortality were screened in 2005, similar to the 43% screened in 2010 (P = .6), despite the low likelihood that these men would survive another 10 years. Furthermore, we observed that a large number of younger, healthier men who were more likely to benefit from the early detection of prostate cancer were not screened for prostate cancer, and approximately half of men (52.4%) ages 65 to 74 years who had a <27% probability of 9-year mortality were screened for prostate cancer in the year before the 2010 survey. Multiple professional organizations have recommended informed decision making between patient and physician, with PSA screening appropriately offered to men who have an estimated RLE of at least 10 years.3-5 Nevertheless, we observed that a significant minority of men who had a high likelihood of 9-year mortality continued to undergo prostate cancer screening in 2010, whereas large numbers of healthier men who had a lower likelihood of 9-year mortality were not screened. Furthermore, our results suggest that, although the vast majority of screened men ages ≥75 years recalled their physicians recommending screening for prostate cancer, only half of those men recalled discussing the potential advantages of PSA screening, and only about a quarter of screened men aged ≥75 years recalled their physicians discussing the potential disadvantages of PSA screening. These results suggest that many men aged ≥75 years who were screened for prostate cancer in 2010, most of whom were unlikely to benefit from prostate cancer screening or a new diagnosis of early stage prostate cancer, did not participate in a shared decision-making process.
Our findings are consistent with other studies demonstrating persistently elevated screening rates among men with lower RLEs in the civilian population and among elderly men in the Veterans Affairs medical system.15-17 These recent findings, when interpreted alongside other recent studies demonstrating unchanged PSA screening rates based on age,18 suggest that physician screening decisions were not altered significantly by either the age-specific or life expectancy-specific portions of recent professional society recommendations released by various medical organizations.
Persistently elevated rates of prostate cancer screening in patient populations with lower RLE and higher mortality risk likely stemmed from numerous factors. Recent surveys of primary care physicians have cited concerns regarding malpractice, patient expectations, time constraints, patient interest, and competing interests as barriers to discontinuing cancer screening.19, 20 Concerns regarding malpractice were especially prevalent among community physicians.19 The majority of survey respondents in one study also expressed difficulty assessing patient RLE,20 and difficulties discussing life expectancy with older patients have been described.21 A recent study demonstrated that, whereas nearly 90% of older Americans value discussions regarding the benefits and drawbacks of cancer screening, the majority of this same population (62%) believed that their own life expectancy should not influence decision making with regard to personal cancer screening decisions. Nearly half (48%) of respondents preferred that physicians not discuss life expectancy at all, and 43% expressed a willingness to pursue screening even against a physician's recommendations.21
Not surprisingly, the decision-making process after a “positive” PSA screening test is difficult and fraught with many of the same complexities as the process of deciding whether to initially offer a patient a PSA test. Various clinical tools for predicting RLE in patients with newly diagnosed prostate cancer have been developed with various degrees of success and implementation.22 These tools may be beneficial when determining the proper course of treatment for a newly diagnosed cancer. Ideally, however, clinicians will have already incorporated a patient's age and comorbidities into the initial screening decision, well before a diagnosis is given to the patient and the subsequent decision of how to best treat a newly diagnosed cancer must be made.
This study has several strengths. First, we investigated prostate cancer screening patterns in elderly men both by age, in accordance with age-based screening cutoffs, and by predicted 9-year mortality, in accordance with guidelines emphasizing the use of health status to guide screening decisions. Second, to our knowledge, this is the first study to investigate the association between prostate cancer screening in elderly men and predicted 9-year mortality, an interval that approaches the 10-year life expectancy suggested by multiple society guidelines as necessary to consider prostate cancer screening.3-5 Incorporating life expectancy into screening decisions is especially important among elderly men who are older than age-based cutoffs and who are unlikely to experience benefits of early prostate cancer detection. Older men, especially those aged >75 years, were well represented in our analysis. Third, these data are presented in conjunction with respondents' self-reported conversations with their physicians regarding the advantages and disadvantages of prostate cancer screening, key components of the informed decision-making process. This provides insight into which factors were discussed before making a prostate cancer screening decision. Fourth, the study was performed using the population-based NHIS data set, a civilian database, and the results may be generalized to the population of the United States. Finally, by excluding men with a history of prostate cancer, we were able to focus solely on PSA tests ordered as part of a routine examination, presumably for the early detection of new cancers, as opposed to tests ordered in response to a problem.
The primary weaknesses of this study were the deidentification of NHIS data and the inability to link responses of survey respondents to a clinical or claims database. This could lead respondents to either overestimate (for example, if a physician stopped screening for prostate cancer without informing the patient) or underestimate (for example, if the patient was not told they were screened) screening behaviors. Nevertheless, previous work has demonstrated good agreement between self-reported PSA testing and independent medical record audit; indeed, a slight, nonsignificant tendency for patients to under-report PSA testing compared with medical records has been demonstrated.23 Furthermore, the rates of prostate cancer screening in this study were similar to rates previously published in the Veterans Affairs medical system, a study performed with clinically linked data, indicating that estimates from the NHIS were accurate despite the lack of a linked claims database.17 A drawback of this study was the lack of direct observations of conversations between providers and patients. As such, recall bias may exist with regard to patients' recollections of discussions, with patients who were screened fixating on the perceived advantages mentioned during discussions and unscreened patients recalling discussing the disadvantages of screening. This recall bias suggests that unscreened men would recall conversations about the disadvantages of screening at a higher rate than those men who were screened. Instead, we observed that nearly twice as many screened men recalled having a discussion about the disadvantages of PSA testing compared with unscreened men, suggesting that the effects of recall bias in the sample may have been minor and that unscreened men aged >75 years generally engaged in less shared decision making.
An additional limitation of this study was that the NHIS database does not provide information regarding which physician ordered a PSA test. For example, it is impossible to know whether a primary care physician or a specialist, such as an oncologist or urologist, ordered a PSA test. As such, we could not comment regarding whether primary care physicians or specialists were relatively more likely or less likely to respond to society-based prostate cancer screening guidelines between 2005 and 2010, an interesting question that remains unanswered in the literature. Similarly, the NHIS database did not include information about practice environment. This limited our ability to analyze prostate cancer screening trends in different practice settings (academic centers, multispecialty group practices, private practices, etc). A final limitation of this study was the potential for self-selection bias. Respondents to the face-to-face survey were allowed to not participate, as evidenced by the final response rates of 69% and 61% in 2005 and 2010, respectively.13 No data were collected from nonrespondents, and, as such, we were not able to tell whether they differed from the survey respondents in a significant manner.
Conclusion
In conclusion, prostate cancer screening patterns in the United States in both 2005 and 2010 demonstrated that the estimated risk of 9-year mortality was largely under-used for making prostate cancer screening decisions among older men. Many older, less healthy American men with limited life expectancies were inappropriately screened for prostate cancer despite a very low likelihood of meaningful clinical benefit; whereas many younger, healthier men with longer life expectancies were not screened. This study also demonstrated that, based on patients' self-reported conversations with physicians, many older men aged >75 years with a high risk of 9-year mortality were screened for prostate cancer without participating in a truly informed decision-making process. This lack of improvement in appropriately targeting prostate cancer screening efforts in the years from 2005 to 2010 may have stemmed from discomfort with calculating and discussing life expectancies, time constraints in the clinic, limited awareness of national guidelines regarding prostate cancer screening, patient preferences, and malpractice concerns. Additional steps in patient and provider education and the economics of screening provision are needed to appropriately alter provider practice patterns.
FUNDING SUPPORT
No specific funding was disclosed.
CONFLICT OF INTEREST DISCLOSURES
Dr. Eggener reports the receipt of consulting fees and honoraria from Janssen Pharmaceuticals, consulting fees from Genomic Health, and consulting fees and research funding from Myriad Genetics.
