1 Personalizing Cancer Screening for the Older AdultThe positive potential is not from the aging society, rather from the ability to make aging a little bit easier through using the potential of new technologies and advanced understanding of the science of screening delivery,

2 Disclosure of Financial RelationshipsErica S. Breslau, PhD, MPH Has no relationships with any proprietary entity producing health care goods or services consumed by or used on patients.

3 Background Cancer is the second leading cause of death among those 65 and older Population screening programs are being debated as they pertain to age Aging society has the potential for a number of significant outcomes, including personalization of cancer screening The positive potential is not from the aging society, rather from the ability to make aging a little bit easier through using the potential of new technologies and advanced understanding of the science of screening delivery,

4 Broadening the definition of Personalized medicine andclinical practice “Personalized cancer medicine is defined as medical care based on the particular biological characteristics of the disease process in individual patients Traditional definition was applied to genomics and proteomics and was based on susceptibility to particular disease or response to specific treatment Expansion of definition incorporates how patient factors influence screening decisions

5 Factors that influence personalized screening decisionsDecisions to screen should personalized: Evidence for positive outcomes Individual health status Genetic profile Benefits and harms of the test Patient preference

6 Screening and Surveillance Processes Across the Cancer Care ContinuumDiagnosis Cancer or Precursor Treatment Post Treatment Survivorship Risk Assessment Primary Prevention Detection End of Life Types of Care Transitions in Care Impact Outcomes Patient Risk status Biologic outcomes Health related quality of life & well- being Quality of death Financial burden Patient experience Quality measures The important distinction is that the guidance for screening should be based on the patient’s personal health information rather than just based on age and organ site as our current guidelines encourage. By this I mean that decisions to screen for cancer should be patient-specific and I would suggest based on five elements to support the right care decision - Beginning with the evidence for positive health outcome which is based on increasingly precise and predictive health care that the person is healthy enough to undergo treatment should a positive test result be revealed. We lack randomized trial of women 75 and older in general , especially those in good health, and particularly for those with ethnic differences . For example, Asian women have a lower incidence of breast cancer and a greater breast density, which might attenuate the benefits of screening. Based on an individual’s health status. The frequency of screening rounds should be adapted to the individual level of risk. Their genetic profile. We know that the incidence of interval cancers is increased in familial risk women as the absolute number of breast cancer is higher in this subgroup, and because of the more aggressive characteristics of the tumor. Their capacity to comprehend the benefits and harms of the test. Population based screening induces significant harms to many healthy individuals: false positives, over-diagnosis and although tumors are detected at an earlier stage, does it save lives. Further false positives alter unnecessary quality of life in many healthy persons. I have not even mentioned the psychological distress and physical discomfort And finally to complete the cycle, the individual’s preference Even though a patient-centered approach necessitates a time-consuming cross-talk between the patient and the clinician, results from a new study by Elston Lafata suggest that it increases adherence to a screening strategy since it is perceived as adapted to a personal case. Population Morbidity Mortality Cost-effectiveness

7 Agenda Framework for Screening Decision MakingLouise C. Walter, M.D., University of California, San Francisco, CA Decision Aids. The answer we’ve been looking for? Mara Schonberg, M.D., M.P.H., Beth Israel Deaconess Medical Center, Boston, MA Implementing Personalized Medicine in Primary Care Sherri Sheinfeld Gorin, Ph.D., Columbia University, New York, NY Future Thinking Erica S. Breslau, Ph.D., M.P.H., National Cancer Institute, Bethesda, MD

8 Next Speaker: Dr. Louise WaltersShe will discuss how to communicate with patients around these core principles of personalized screening decisions, including the use of decision aids.

9 Framework for Screening Decision MakingLouise C. Walter, MD Professor of Medicine Division of Geriatrics University of California, San Francisco San Francisco VA Medical Center

10 Disclosure of Financial RelationshipsLouise C. Walter, MD Has no relationships with any proprietary entity producing health care goods or services consumed by or used on patients.

11 Objectives Describe a framework for personalizing cancer screening decisions in older adults Consider life expectancy/health when making screening recommendations Understand the importance of factoring patient preferences into screening decisions The objectives of this talk are to….

12 Cases Mrs. A 70 y/o woman with Alzheimer’s dementia (MMSE=10/30) and functionally dependent in many ADLs. She lives with her daughter who brings her in for a routine check-up. She has no history of any cancer screening tests. Mrs. B 80 y/o woman with a history of osteoarthritis. She walks 2 miles a day and cares for her older sister. She has not seen a doctor in several years but decides to come in for a routine check-up. She has no history of any cancer screening tests. So I thought I would start out with two cases to get you thinking about how you currently make screening recommendations to older patients. Mrs. A….. Mrs. B…. What screening tests would you recommend for Mrs. A and Mrs B? Let’s start with Mrs A…How many of you would recommend a screening mammogram for Mrs A? A screening Pap smear? FOBT? How about Mrs B…mammogram? Pap smear? FOBT? So there is lots of variation in what people here would recommend. So think about how you made your decision about what screening test you would recommend to Mrs A and Mrs B and we will come back to these cases at the end to illustrate some of the main points of the talk.

13 Uncertainty Uncertainty about when to screen asymptomaticelderly patients for cancer Most trials of cancer screening tests have excluded patients over age 75 Extrapolate data to older patients Data from randomized trials not always applicable to an individual patient Trials do not address individual characteristics that may change the likelihood of benefit vs. harm As you were thinking about whether to screen Mrs A or Mrs B for cancer you probably realized that there is a great deal of uncertainty about screening asymptomatic older persons for cancer. This uncertainty stems partly from the fact that most trials of cancer screening tests have … Requires clinicians to extrapolate data about the efficacy of a screening test performed in younger people to older people. In addition, even if you think a screening test is likely to perform just as well in older people as younger people there is still uncertainty because data from RCT are not always applicable to an individual patient. Trials reveal the average effectiveness of an intervention and in general do not address individual characteristics (like health status) which may change the likelihood receiving benefit vs harm from a screening test. For example, none of the RCTs address the question, What is the likelihood of benefit of mammography for a woman with moderate dementia or CHF? Yet these are the questions clinicians are having to address all the time in our clinics because many older patients that we see have significant comorbid illnesses.

14 What to Do? Guidelines used to be based on age cutoffs and were conflicting Mammography Guidelines (until 2002) USPSTF: Stop mammography at age 70 American College of Physicians: Stop at age 75 American Geriatrics Society: Stop at age 85 American Cancer Society: No upper age limit Now guidelines agree that screening: Should continue if an older person is healthy Should stop if an older person is in poor health—has a limited life expectancy So what to do? Well, often we can look to guidelines for help but when I first got interested in cancer screening in the elderly in the late 1990’s I found that the guidelines were usually based on age cutoffs and were often conflicting. For example, mammography guidelines up until 2002 ranged from the USPSTF recommending up until 2002, to stop mammography at age 70, ACP.., AGS… while the American Cancer Society did not mention stopping, everyone should be screened. Luckily, now guidelines are more in agreement that screening should continue if an older woman is healthy and should stop if an older woman is unhealthy, has a limited life expectancy.

15 Screening Not Optimally TargetedScreening is not being optimally targeted Healthy older patients often under-screened Older patients in poor health often over-screened Opportunities for screening usually rely on patients visiting clinicians for medical problems Targets screening to sick patients; misses healthy Screening often viewed as a simple check-box rather than a personalized decision Quality indicators encourage high screening rates regardless of health So the conclusions from these studies is that screening is not being optimally targeted. Healthy older patients are often underscreened while older patients in poor health are often overscreened. Reasons for this may be that opportunities for screening usually rely on patients visiting clinicians for medical problems which of course targets screening to sick patients and misses healthy older adults who rarely see clinicians. And another reason may be that most quality indicators encourage cancer screening regardless of health or life expectancy.

16 (N = 4,792 women in California)Among Women of Similar Ages, Mammography Screening Rates DO NOT Decline with Worsening Health (N = 4,792 women in California) Health Quartile (SF-12) For example, among women of similar ages, mammography screening rates did not decline with worsening health. This was a population-based study of 4,792 women in California with % reporting a mammogram within 2 years on the Y axis and age on the X-axis. You can see that overall screening rates decreased with age. However, within each age group mammography rates are the same regardless of health quartile. In fact, often, the mammography rate is the same or higher for women in worst health (the red bar) compared to those in best health (the green bar). In addition, women aged in the worst health were more likely to report screening mammography than women aged in best health even though healthy year-old women likely have greater life expectancies than unhealthy women in their late 70’s. (12-items: self-rated health and limitations in physical function) Walter LC. Ann Intern Med. 2004;140:

17 Among Persons of Similar Ages, Colorectal Cancer Screening DOES NOT Decline with Worsening HealthCharlson Comorbidity Score (N = 27,068 veterans) And lastly my most recent study found that among persons of similar ages, colorectal cancer screening did not decline with worsening health. This was a study of approximately 27,000 veterans across 4 VAs with the % of persons who had colorectal cancer screening within the past 2 years on the y-axis. You can see that colorectal cancer screening rates are relatively low (much lower than the PSA screening rates on the last slide) and decrease with advancing age. However, within each age group colorectal cancer screening rates are fairly similar regardless of the Charlson score. In fact, even among those aged in best health (green bar) screening rates are only 51% and if you look at those 80 years and older in worst health still a third of them are being screened so not that big of difference in the % of people being screened despite dramatically different life expectancies between these groups. Walter LC. Ann Intern Med. 2009;150:

18 Screening Decisions Age should not be the most important factor in screening decisions “One-size-fits-all” approach to medical care based on age does not work in diverse elderly population Variation in comorbidity/life expectancy/preferences Present a framework to guide how to think through cancer screening decisions in elderly Incorporates individual characteristics (health/life expectancy) and preferences into decisions So this is great because age should not be the most important factor in screening decisions because the one-size-fits all approach to medical care does not work in the diverse elderly population which has great variation in health status/life expectancy/ and preferences. Therefore, I will present a framework I developed to help guide how to think through cancer screening decisions in the elderly that incorporates individual characteristics, such as health and life expectancy, as well as patient preferences, into screening decisions. This framework really emphasizes patient-centered care.

19 Framework for Individualized DecisionsEstimate life expectancy Determine potential benefits of screening Determine potential harms of screening Weigh potential benefits and harms according to an individual’s values and preferences So here is the framework. One of the first things to do is to estimate a person’s life expectancy and we’ll talk about how to do this. Then you need to determine potential benefits of screening. Has this test ever been proven to do anything good? And then you need to determine the potential harms of screening. All tests have downsides. And lastly it is important to weigh the potential benefits and harms of a screening test according to an individual’s values and preferences because different people are going to place different degrees of importance on the potential benefits and harms of a test. So this talk will go through each of these 4 components of the framework.

20 Life Expectancy for WomenYears Well, the first step in the framework is to think about life expectancy. I present here a graph of the life expectancies for women in the US based on US Life Tables. The green line represents the healthiest women in the US, those in the top 25th percentile at each age (these are active women with no significant comorbid illnesses), the blue is the 50th percentile (average life expectancy at each age) and the red represents the lowest 25th percentile of survival at each age (these are women who are bedbound or those with severe comorbid illnesses like dementia or CHF). So you can see that there is great variation in how long people of similar ages live. For example, if you take all 80 y/o women in the US, 25% will live more than 13yrs. Half will live close to 10 years and 25% of the sickest 80 y/o women will live less than 5 years. It also shows that average LE does not fall below 5 yrs until a woman is well over 85. The graph for men shows the same trends except at each age the values are 2-3 years less because men don’t live as long as women. But there is still a lot of variation in how long older people of similar ages live which is really the take home point of this graph. Age (years) Walter LC. JAMA 2001;285:

21 Life Expectancy Number/severity of comorbid conditions and functional impairments stronger predictors than age Life expectancy substantially below average CHF (Class III, IV), ESRD, Severe COPD (home O2), Severe dementia (MMSE < 10), Dependency in many ADL Life expectancy substantially above average No comorbid conditions or functional impairments; active Prognostic indices that combine age, comorbidity, functional status, laboratory results, etc. can supplement clinical judgment And while I agree that you can’t look at someone and say you are going to live exactly 9.5 years, I do think that it is possible to estimate whether a person is likely to live substantially longer or shorter than an average person of their age group. And we do this all the time when making medical recommendations for surgery or other interventions. We think about the number and severity of comorbid conditions and functional impairments because these are much stronger predictors of life expectancy than age alone. So when I see an older pt in my office I ask myself does this person have a life expectancy that is substantially below average. Do they have……All these are associated with very high 5yr mortality rates. On the flip side I ask myself if this person has a life expectancy substantially above average…do they lack comorbid conditions or functional impairments are they very active. And while I agree that estimating LE is not perfect, I think that considering a person’s health status gives a more accurate estimation of their likelihood of receiving benefit or harm from a screening test than focusing on age alone.

22 Lag-Time to Benefit Benefit of screening does NOT occur immediatelyScreening results in benefit by finding cancers at an early stage, which would have caused symptoms or killed a person years later A life expectancy of > 5-10 yrs is required to have some chance of survival benefit from screening RCTs of mammography and FOBT show survival curves of screened vs. unscreened do not separate significantly until > 5-10 years after start of screening And the reason why life expectancy is so important to think about in screening decisions is that the benefit of screening does not occur immediately. For cancer tests, screening results in benefit by finding cancers at an early stage which would have caused symptoms or killed a person years later. If a person is not going to live those years later they are not going to benefit from a cancer screening test. And in general a LE of greater than 5 years is required to have some chance of survival benefit from cancer screening. This is based on RCTs…. Lee S. et al. BMJ Jan (on-line)

23 Harms of Screening ImmediateOlder adults with life expectancy < 5 years Unlikely to benefit from cancer screening Increased risk for experiencing harm from screening Complications due to inaccurate test results Identification and treatment of clinically unimportant disease that would never have progressed to symptoms in patient’s lifetime Psychological distress Yet the harms of screening are immediate. So while older adults with a life expectancy less than 5 years are unlikely to benefit from cancer screening they are often at increased risk for experiencing harm from screening such as complications from additional diagnostic procedures due to inaccurate test results...so essentially a false positive result and all the tests emanating from that result. Identification and treatment of clinically unimportant disease that would never have progressed to symptoms in a patient’s lifetime but by finding it and treating it you cause significant harm to the person. And psychological distress from the above 2 items or undergoing the screening test itself.

24 Harms of Screening MammographyInaccurate “False-positive” results: ~50% of women yrs will have false-positive after 10 mammograms biopsies, ultrasounds, etc. Clinically unimportant cancers (overdiagnosis): ~10-25% of breast cancers detected by screening would remain dormant and never progress to symptoms More likely in women in poor health Detection by screening often leads to unnecessary treatment and complications So I conducted a sutdy at On Lok, which is part of PACE, the program of all-inclusive care for the elderly, which is a capitated health program for the frail elderly. In the mid 1990’s CA state auditors insisted that mammograms be performed in all women who enrolled in On Lok despite poor health and advanced age as per the American Cancer Society’s guidelines. It didn’t matter that 50% of patients in this program have dementia or that the median life expectancy is 4 years. So On Lok complied with their request but we decided to assess the harms of this universal screening mammography policy in frail older women. Fletcher SW. Epidemiol. Rev. 2011

25 Harms of Colorectal and Cervical Cancer ScreeningColorectal Cancer Screening False-positives: ~1 in 10 for FOBT Complications: 3 colonoscopies per 1,000 screened (perforation, bleeding, cardio/pulmonary events) Cervical Cancer Screening Of ~2,500 women yrs screened  110 had abnormal Pap  33 colposcopies, 35 endocervical curettages, 39 biopsies, 4 dilation & curettages  1 woman with mild-moderate cervical dysplasia Whitlock EP et al. AHRQ pub EF-1; 2008 For example, colorectal and cervical cancer screening have harms. For CRC screening 1 in 50 people who undergo FOBT will have a false-positive result and the work-up of a positive result is not benign. Serious complications are estimated to occur in 1 per 1000 screening colonoscopies..including perforation, serious bleeding and stroke…and this is in healthy people…the complication rate is likely much greater in older persons with severe comorbid diseases. And even cervical cancer screening has its downsides….One study of around 2500 women aged who had a screening pap smear within 2 yrs of a normal pap…. So 109 women were harmed in this study. Sawaya G. Ann Intern Med. 2000:

26 Psychological Harm Many important burdens not measured in RCTsEmotional pain of cancer diagnosis in people whose lives were not extended by screening Alarm of false-positive results Stress of undergoing the screening tests Magnitude of psychological harm is individual Cognitive or sensory problems may make tests and follow-up procedures more difficult, painful or frightening And then there is psychological harm that screening tests may cause pts. Many burdens that are important to pts are not measured in clinical trials such as the emotional pain of a diagnosis of cancer in people whose lives were not extended by screening, or the alarm of a false-positive result. For mammography studies have shown that this alarm can persist for over 6 months and affect mood and activities. And then there is the stress of undergoing the screening tests. Some people have poo-pooed this saying how bad can this stress be compared to preventing death from cancer but I think the magnitude of psych harm is individual……more frightening. This should be considered because it raises the bar on the amount of benefit this test better give a patient. Obviously, the magnitude of stress or psychological harm causes by screening is somewhat dependent of the individual. We all know there are some pts at higher risk for psychological harm from screening and we should think about whether the patient we’re thinking about screening has cognitive or sensory problems that would make screening and follow-up testing particularly painful or frightening. Would the pt understand why her breasts needed to be squeezed during a mammogram or would she more likely get agitated or afraid? These are important things to think about because they raise the bar for the amount of potential screening benefit that is needed to offset the potential harms. Sox. JGIM 1998;13:424-5

27 Preferences Assess how patients view potential harms/benefits and integrate values/preferences into decisions Different from public health strategy in which experts weigh benefits/risks and decide what is best for a population Since many decisions in older patients are “close calls,” need to consider values/preferences Harms look larger to some people Non-mortality benefits considered more substantial to some people (e.g., “peace of mind”) So that brings us to preferences….

28 Cases Mrs. A – 70 y/o woman with severe dementiaMrs. B – 80 y/o woman with osteoarthritis Estimate life expectancy Mrs. A is younger but has severe dementia and functional dependency, so life expectancy < 5 years Mrs. B is probably in upper quartile of life expectancy for her age, so likely to live > 13 years Probability of benefit Mrs. A unlikely to benefit since life expectancy < 5 yrs Mrs. B has reasonable likelihood of benefit given her substantial life expectancy So let’s return to our two cases to finish up with cancer screening decisions.

29 Cases Probability of harm Values and preferencesMrs. A has severe dementia so tests may cause distress and if cancer identified likely is unimportant Mrs. B understands and accepts risks of tests Values and preferences Mrs. A has avoided doctors and becomes agitated if anything interrupts her daily routine Mrs. B worries about her health and wants a mammogram, FOBT, and Pap smear Screening recommendations Recommend AGAINST screening Mrs. A Recommend screening Mrs. B Realize cases a bit black and white but point is to emphasize that age was not the main driver of cancer screening decisions in these patients.

30 Personalized DecisionsDo Screen Don’t Screen Likelihood of Benefit Likelihood of Harm So I summarize this idea of individualized decisions with a scale which shows the need to weigh the likelihood of benefit vs the likelihood of harm when making recommendations for or against an intervention. And that patient preferences should act like a moveable fulcrum of a scale that shifts the magnitude of the harms and benefits that are needed to tip the decision towards recommending a preventive intervention or not recommending an intervention. For some patients any chance that they would reduce their risk of dying of cancer will tip the scale towards Screening because the benefits are much more important to them than the harms. Then there are others who don’t want to take any chance that the test or intervention might harm them so they’d rather take their own chances tipping the scale towards don’t screen. Patient Preferences (moveable fulcrum)

31 Conclusion Be thoughtful about cancer screening (rather than a checkbox approach based on age) Personalize decisions by thinking about Life expectancy and Health Status Benefits and Harms Patient Preferences These are core principles of high quality medical decision making and should be applied to screening decisions So in conclusion, be thoughtful about cancer screening in older patients rather than following a checkbox approach. Think about life expectancy and comorbidity and how these impact benefits and harms and be aware of patient preferences. These are core principles of good medicine and are necessary for maximizing informed cancer screening decisions. Screening should be an individualized and informed decision.

32 Next Speaker: Dr. Mara SchonbergShe will discuss how to communicate with patients around these core principles of personalized screening decisions, including the use of decision aids.

33 Decision Aids—the answer we’ve been looking for?Mara A. Schonberg MD, MPH, Division of General Medicine and Primary Care Beth Israel Deaconess Medical Center Harvard Medical School May 3, 2013

34 Disclosure of Financial RelationshipsMara Schonberg, MD, MPH Has no relationships with any proprietary entity producing health care goods or services consumed by or used on patients.

35 Funding K23/Paul B. Beeson Career Development Award in Aging K23 Benefits and Burdens of Screening Oldest- old Women: The Case of Mammography No conflicts of interest to report

36 Overview None of the RCTs evaluating mammography screening included women 75+ Harms: pain; anxiety; complications from follow-up tests (e.g., breast biopsy), and overdiagnosis (finding tumors that would never cause symptoms in an older woman’s lifetime) Risks of treatment increase with age

37 Policy USPSTF state that there is insufficient evidence to recommend mammography screening for women 75+ years Others organizations (AGS/ACS) recommend clinicians consider patient life expectancy and/or health All recommend that older women be informed of the pros and cons of mammography screening However, few older women are adequately informed Most overestimate the benefits and underestimate the risks

38 Aim To develop decision aids (DA) for adults 75+ years and their clinicians to improve decision-making around mammography screening

39 Decision Aids Leaflets, video, interactive media, etc.Designed to help patients understand risks and benefits of a test, clarify preferences and values, and be more involved in decision-making Differ from usual health education materials because of their detailed, specific, & personalized focus on options and outcomes of a decision Helpful when ratio of benefits to risks are uncertain May be used to prepare for a visit or during a visit to facilitate decision-making Attempt to present the information necessary for a decision without bias Helpful when ratio of benefits to risks are uncertain as is the case for many cancer screening tests for older adults DAs may be designed as leaflets, interactive media, or video, etc. DAs are intended to supplement rather than replace patient-physician interaction. Attempt to present the information necessary for decision-making without bias specific and personalized focus on options and outcomes.

40 Decision Aids to improve cancer screening decisions among older adultsA Decision Aid on mammography screening for women 75+ years Eprognosis.org Mobile application on cancer screening for older adults

41 Development of the mammography screening DA for Women 75+International Patient Decision Aid Standards (IDPAS) Decision-making processes of older adults

42 Dual Process Theory System 1 (Experiential) System 2 (Deliberative)affective intuitive and holistic based on our experiences fast less than conscious “We are seized by our emotions” System 2 (Deliberative) deliberative analytical logical conscious Slower Fairly recent evolutionary history It is thought that decision making is processed using two different modes of thinking: experiential/affective and deliberative. Both modes of thought are important to informing decisions. There is mounting evidence that people use both modes of thought.

43 Age and Decision-MakingAs individuals age they process information less quickly Have less deliberative capacity and increasingly rely on affective processes Tend to comprehend numeric and information presented in tables and charts less well May be led astray by anecdotal information Tend to be more likely to remember positive outcomes (positivity bias) Numeracy, as we define it in our work, if the ability to understand and use basic probability and mathematical concepts. Many of the measures of numeracy focus on the use of probability and risk information that is often discussed in the medical setting. One should note that education is more of a protective factor for numeracy than it is for the fluid intelligence skills we just discussed. Predicts that older adults will inevitably make worse decisions

44 Age and Decision-MakingIs deliberative decline the whole story? Tend to comprehend numeric and information presented in tables and charts less well May be led astray by anecdotal information Tend to be more likely to remember positive outcomes (positivity bias) As individuals age they process information less quickly Have less deliberative capacity and increasingly rely on affective process Older adults selectively use their more limited deliberative capacity Accumulated experience can compensate Emotional focus as an adaptive and motivated process (Carstensen, Mather, Labouvie-Vief) Older adults selectively use their more limited deliberative capacity Compensate with accumulated experience Older adults may make better decisions in situations that require experience or attending to emotional content story. There is evidence that older adults use their deliberative capacity selectively, that accumulated experience can compensate in some tasks, and that some of the apparent poor decision making may be the result of a focus on information and experiences that improve their emotional state. Robust declines in deliberative capacity suggest that older adults will make worse decisions than younger adults in some situations. Deliberative decline is likely to be too simple a story though for three reasons. First, older adults appear to selectively use their deliberative capacity. Second, accumulated experience with compensates. And, third, emotional focus appears to increase with age. As a result, older adults will make better decisions than younger adults in some situations Predicts that older adults will inevitably make worse decisions

45 Age and Decision-MakingExpert Recommendations Tend to comprehend numeric and information presented in tables and charts less well May be led astray by anecdotal information Tend to be more likely to remember positive outcomes (positivity bias) As individuals age they process information less quickly Have less deliberative capacity and increasingly rely on affective process Older adults selectively use their more limited deliberative capacity Accumulated experience can compensate Emotional focus as an adaptive and motivated process (Carstensen, Mather, Labouvie-Vief) To use lists rather than paragraphs to present information To make the “gist” or evaluative meaning of any numeric information clear To emphasize the negative aspect of a decision to counteract the positivity bias To simplify the message to avoid information overload. story. There is evidence that older adults use their deliberative capacity selectively, that accumulated experience can compensate in some tasks, and that some of the apparent poor decision making may be the result of a focus on information and experiences that improve their emotional state. Robust declines in deliberative capacity suggest that older adults will make worse decisions than younger adults in some situations. Deliberative decline is likely to be too simple a story though for three reasons. First, older adults appear to selectively use their deliberative capacity. Second, accumulated experience with compensates. And, third, emotional focus appears to increase with age. As a result, older adults will make better decisions than younger adults in some situations Predicts that older adults will inevitably make worse decisions

46 Development of the DA Expert panel (internists, epidemiologists, geriatricians, and a psychologist) reviewed iterative versions Assessed clarity and acceptability with 15 women >75 years from our hospital’s primary care practices and 5 of their PCPs Tested this version in a pretest/posttest trial

47 %âãÏÓ

48 14 0 obj

49

50

51 23 0 obj

52

53

54 endobj

55 startxref

56

57 %%EOF

58 Pretest/Posttest Trial (Pilot)Eligible Women years English-speaking No history of dementia No history of invasive/non-invasive breast cancer Not screened in the past 9 months Screened in the past 3 years Refuses screening not documented Not first visit with PCP

59 Study Flow Complete “Before” Survey Read the Decision Aid PCP visitComplete “After” Survey Medical record review for follow-up

60 Outcomes Knowledge of pros/cons of mammography (10 items)Decisional conflict (16 items) Screening intentions (15 point scale) Documented discussion of the pros/cons of screening in the 5 years before/9 months after the intervention Receipt of screening 15 months after (at least 2 years since last mammogram)

61 Primary Care PhysiciansAcceptability of the Decision Aid help patients make more informed, value-laden decisions Balanced

62 Analyses We used the signed rank and McNemar’s test to compare pretest/posttest responses

63 Sample Population (n=49)Median age: 79 years (range 75-86) 70% were Non-Hispanic white 63% had attended some college 53% had <9 year life expectancy Patients of 26 PCPs

64 Results Before After P value Knowledge 6.3 (+/-1.2) 7.3 (+/-1.4%) 0.01Decisional conflict 20 (+/-14) 17 (+/-12) 0.03 Intend: yes 82% 59% Documented discussion 11% (up to 5 years before) 58% (9 months after) <0.01 Underwent screening 84% in 2 years before 58% 15 months after *52% with <9 year life expectancy were screened vs. 67% of those with >9 year life expectancy

65 Additional Results 94% would recommend the DA78% found the amount of information just right 94% preferred the DA to be on paper rather than computer 46% found the information balanced, 37% slanted towards not getting a mammogram

66 PCPs (19/26) 74% using the DA would result in their patients making more informed decisions 79% using the DA would result in their patients making more value laden decisions 65% found it balanced, 35% found it slanted towards not getting a mammogram

67 Limitations Limited generalizabililty- one site, small study

68 Conclusions Our before/after trial suggests that the decision aid helps older women make more informed, preference-sensitive decisions around mammography screening Next, we plan to test the effectiveness of the DA in a large randomized control trial

69 Additional Decision AidsLife expectancy calculators (www.eprognosis.org) Mobile application: eprognosis-cancer screening If it was still found to be effective in an RCT may be appropriate for implementation in academic detailing or Practice Facilitation as described by Dr. Sherri Sheinfeld Gorin, Ph.D.

70

71

72 It is not possible to predict the future for any single personIt is not possible to predict the future for any single person. Risk estimate tell us how many people will die and how many will survive, but we do not know who will die and who will survive. Given 100 people with similar answers to the index, 75 will die and 25 will survive over the next 9 years.

73 Mobile Application Focuses on providing information about life expectancy to help users decide whether or not to be screened for cancer; specifically breast or colon cancer User is asked 15 health related questions to estimate life expectancy Compares life expectancy to the lagtime to benefit for breast and colorectal cancer screening If it was still found to be effective in an RCT may be appropriate for implementation in academic detailing or Practice Facilitation as described by Dr. Sherri Sheinfeld Gorin, Ph.D.

74

75

76

77

78

79

80

81 Are Decision Aids the Answer?Strong pilot data of a mammography screening decision aid Enthusiasm ePrognosis ePrognosis: Cancer Screening Implementation

82 Next Speaker: Dr. Sherri Sheinfeld GorinShe will discuss the implementation of personalized medicine in primary care through academic detailing and practice facilitation

83 Implementing Personalized Medicine in Primary CareSherri Sheinfeld Gorin, Ph.D. NCI (SAIC) New York Physicians against Cancer (NYPAC) “Personalized cancer medicine is defined as medical care based on the particular biological characteristics of the disease process in individual patients. By using genomics and proteomics, individuals can be classified into subpopulations based on their susceptibility to a particular disease or response to a specific treatment. They may then be given preventive or therapeutic interventions that will be most effective given their particular characteristics.” (IOM, 2010). . Policy issues in the development of personalized medicine in oncology: Workshop summary. Washington, DC: The National Academies Press.

84 Disclosure of Financial RelationshipsMara Schonberg, MD, MPH Has no relationships with any proprietary entity producing health care goods or services consumed by or used on patients.

85 Funded by American Cancer Society Centers for Disease Control and Prevention Department of Defense National Cancer Institute “Personalized cancer medicine is defined as medical care based on the particular biological characteristics of the disease process in individual patients. By using genomics and proteomics, individuals can be classified into subpopulations based on their susceptibility to a particular disease or response to a specific treatment. They may then be given preventive or therapeutic interventions that will be most effective given their particular characteristics.” (IOM, 2010). . Policy issues in the development of personalized medicine in oncology: Workshop summary. Washington, DC: The National Academies Press.

86 “How do we integrate new evidence [from personalized medicine] into existing clinical practice?” (IOM, 2010) “Personalized cancer medicine is defined as medical care based on the particular biological characteristics of the disease process in individual patients. By using genomics and proteomics, individuals can be classified into subpopulations based on their susceptibility to a particular disease or response to a specific treatment. They may then be given preventive or therapeutic interventions that will be most effective given their particular characteristics.” (IOM, 2010). . Policy issues in the development of personalized medicine in oncology: Workshop summary. Washington, DC: The National Academies Press.

87 Personalized Medicine and Clinical PracticeDissemination of personalized medicine into clinical practice is not a passive process. It, too, is personalized, depending upon: the characteristics of a multi-level context (policy, provider, and patient), the attributes of the evidence (e.g., complexity), decision type (e.g. collective), communication channels (e.g., interpersonal), and the extent of the change agents’ promotion actions (Rogers, 2003). “Personalized cancer medicine is defined as medical care based on the particular biological characteristics of the disease process in individual patients. By using genomics and proteomics, individuals can be classified into subpopulations based on their susceptibility to a particular disease or response to a specific treatment. They may then be given preventive or therapeutic interventions that will be most effective given their particular characteristics.” (IOM, 2010). . Policy issues in the development of personalized medicine in oncology: Workshop summary. Washington, DC: The National Academies Press.

88 Aim Examples: Academic Detailing of: Colorectal Cancer ScreeningTo discuss the application of Academic Detailing (AD) and Practice Facilitation (PF) approaches to implementing personalized medicine among primary care practices serving older adults, using examples from cancer and other chronic behavior screening. Examples: Academic Detailing of: Colorectal Cancer Screening Informed Decision-making for Prostate Cancer Practice Facilitation of screening for cancer and other chronic behaviors.

89 Rationale Significant variability in MD screening practicesMD recommendation central to cancer screening Continuing medical education alone has limited success in changing MD behavior Academic Detailing: Face-to-face counseling of physicians, brief, focused, repeated frequently Effective in diverse individual studies (review in; Hulscher et al., 2002; O'Brien MA et al., Cochrane Database of Systematic Reviews 2007, 1998, 2010; multi-component, [reminders, feedback, education], CDC Task Force on Community Preventive Services, 2005). Significant effect on cancer screening behavior in subgroup of inner city primary care MD’s (> 3 visits) (Sheinfeld Gorin et al., 2000); on recommendations for mammography to woman age 40 and older among urban primary care physicians (Sheinfeld Gorin et al., 2001).

90 12/20/2017 Wagner et al., 1999 National Cancer Institute

91 Examples: Effect of academic detailing ondissemination of colorectal cancer screening to PCP’s CRC second leading cause of cancer-related mortality in US for men, third for women CRC largely can be prevented by the detection (via screening) and removal of precursor lesions (adenomatous polyps), and survival is significantly better when CRC is diagnosed while still localized. (Levin et al., 2008)

92 Communities Bronx Washington Heights/Inwood Central Harlem East HarlemUpper East Side Murray Hill Upper West Side Essex County, New Jersey – including Newark, Irvington, East Orange, and Orange, NJ

93 Sheinfeld Gorin & Heck, J Nat Med Association, 2007

94 Physician Socio-demographics: Model of PCP Colorectal Cancer Screening Recommendations Background Factors Mediating Factors Behavioral Intention Physician Socio-demographics: Age, Gender, Race/Ethnicity (Non-Hisp White) Negative Behavioral Beliefs Stage of Change to Recommend Colonoscopy Perceived Behav Control Self-efficacy in Prevent Counseling Patient Mix Theory of planned behavior, social cognitive theory, stages of change (precontemplation, contemplation, preparation stages of change. A number of studies have found that forward transitions from precontemplation, contemplation, and preparation stages of change can be interpreted in terms of increasing strength of respective behavioral intentions Normative belief about the effectiveness of colonoscopy: 1. According to published studies, regular colonoscopy screening is effective in the detection of colorectal cancer. 2. Given the demands of a medical practice, colonoscopy can effectively identify asymptomatic patients with adenomatous polyps. 3. Given the demands of a medical practice, colonoscopy can effectively identify asymptomatic patients with early colorectal cancer. Perceived behavioral control over uncertainty about patient care 1. Not being sure of what is best for a patient is one of the most stressful parts of being a physician. 2. I usually feel anxious when I am not sure of a diagnosis. 3. I fear being held accountable for limits of my knowledge. Negative behavioral beliefs to recommend colonoscopy 1. Colonoscopy takes too much time to explain. 2. Colonoscopy is too difficult for the patients in this practice to understand. 3. Colonoscopy is too difficult to explain. . Practice barriers to preventive care 1. Lack of systems for tracking and promoting preventive care. 2. Lack of consensus on what preventive services to provide. 3. Lack of effective patient education materials. Theory is little used in the prediction of physician cancer screening stage of change. Structural equation modeling was used to evaluate the theoretical predictors of stage of change to recommend colonoscopy among 235 urban physicians. Constructs from the theory of planned behavior, social– cognitive theory, and the transtheoretical model were systematically tested. As predicted, contextual factors, such as the physicians’ ages, their race– ethnicities, patient race– ethnicity, and office-related barriers to preventive care were associated with stage of change through self-efficacy, normative beliefs, and negative behavioral beliefs. The findings demonstrate the relevance of these models to studying the behavior of physicians and support the development of interventions that are tailored to normative beliefs and specific physician cognitions for colonoscopy recommendation. CRC Screening Recommendation Practice Barriers to Preventive Care Normative Beliefs TPB, SCT, SOC Honda & Sheinfeld Gorin, 2006

95 Digital Detailing

96 Decision point: At a brief “decision point,” the physician is asked to select an answer that reflects his or her realistic response to the archetypal patient.

97 Expansion of the decision pointExpansion of the decision point. With a “click,” the physician may expand the decision point to obtain additional information on the patient, to assist with his/her decision-making. Access on-line resources (for physician or patient and families) to assist with decision-making.

98 Response evaluation: The physician’s response to the archetypal patient is also rated on how informative, responsive, and appropriate it is.

99 Stratification of communities by geography, socio-demographicsStudy Design Identification of primary care physician offices in North, East and West Manhattan, Bronx, and Essex County, NJ Stratification of communities by geography, socio-demographics N. Manhattan, Bronx, Essex Co. MD offices (N=140) Upper East/Westside, Murray Hill (N=140)

100 Study Design (continued)Upper East & West sides, Murray Hill (N=140) North Manhattan, Bronx and Essex County, NJ (N=140) R R I,N=70 C,N=70 I,N=70 C,N=70 Baseline with MD 6-month follow-up with MD 12-month follow-up with MD Chart Audit (N=1950) Pt. Survey (N=350)

101 Summary of GEE Logistic Model Estimates Outcome is Patient Colonoscopy Screening Status as of 60 Days After Final Detail (N=1340) Variable Description Odds P-value Ratio Colonoscopy Screening at Baseline Community Arm % of Uninsured Patients % of Patients Covered by Medicare Doctor screens in office Patient Age Years Doctor has Practiced Doctor in Private Practice Office Practices Arm*Private Practice Arm*Years Practiced Notes: Excludes patients with any prior positive test results. Sheinfeld Gorin et al., 2009

102 Sustainability of academic detailing over time (N=852 medical audits, 20% sample)Percent of Patients Receiving Colonoscopy Screening By Month Post-Intervention 2 4 6 8 10 0-6 mos 6-12 mos 12-18 mos 18-24 mos Intervention Control p=.003 P=.003 p=.003 p=.03 Sheinfeld Gorin et al., 2005

103 Conclusions Intervention-related 7% increase in colonoscopy screening recommendations for individuals age 50 and older No differences in recommendations for other CRC screening tests as a result of the intervention Consistent findings across self-report data, medical audit findings Per-case cost comparable to other more intensive approaches (M, $721.77; minimum cost: $641.22, maximum cost: $849.51). The incremental cost effectiveness ratio was $21,124 per percentage point increase in CRC screening. Some economies of scale as number of physicians increase. Less cost effective than other, less intensive, intervention approaches with providers (e.g., tailored mailed materials). Not compared with practice facilitation approaches, for example. ICER: The ICER was calculated on the basis of the total cost of the intervention and the relative change in CRC screening rates between control and intervention groups across both communities. The clinical outcomes used in the calculation of ICER were the completion of colonoscopy, FOBT, and flexible sigmoidoscopy. The ICER is reported as dollars per additional percentage point increase in CRC screening rate. As in prior studies of the cost effectiveness of CRC screening promotion efforts, sensitivity analysis estimates were based on a 10% variation around the point estimates for the various cost inputs.10,11 An additional sensitivity analysis evaluated the ICER for differing numbers of physicians in the practice and, accordingly, varying intervention delivery costs among the practices (Table 3; Fig 2). The average practice setting included four physicians. For this sensitivity analysis, the intervention delivery costs varied, whereas other costs remained constant. Intervention delivery costs were doubled if the practices included an average of two practitioners and were halved for those with eight practitioners. Academic detailing was associated with a 7% increase in CRC screening with colonoscopy. The total intervention cost was $147,865, and the ICER was $21,124 per percentage point increase in CRC screening rate. Sensitivity analyses that varied the costs of the intervention and the average medical practice size were associated with ICERs ranging from $13,631 to $36,109 per percentage point increase in CRC screening rates.

104 Not compared with practice facilitation approaches, for example.ICER: The ICER was calculated on the basis of the total cost of the intervention and the relative change in CRC screening rates between control and intervention groups across both communities. The clinical outcomes used in the calculation of ICER were the completion of colonoscopy, FOBT, and flexible sigmoidoscopy. The ICER is reported as dollars per additional percentage point increase in CRC screening rate. As in prior studies of the cost effectiveness of CRC screening promotion efforts, sensitivity analysis estimates were based on a 10% variation around the point estimates for the various cost inputs.10,11 An additional sensitivity analysis evaluated the ICER for differing numbers of physicians in the practice and, accordingly, varying intervention delivery costs among the practices (Table 3; Fig 2). The average practice setting included four physicians. For this sensitivity analysis, the intervention delivery costs varied, whereas other costs remained constant. Intervention delivery costs were doubled if the practices included an average of two practitioners and were halved for those with eight practitioners. Academic detailing was associated with a 7% increase in CRC screening with colonoscopy. The total intervention cost was $147,865, and the ICER was $21,124 per percentage point increase in CRC screening rate. Sensitivity analyses that varied the costs of the intervention and the average medical practice size were associated with ICERs ranging from $13,631 to $36,109 per percentage point increase in CRC screening rates.

105 Dissemination of prostate cancer informed decision-makingStudy designed to demonstrate the feasibility of disseminating the American Cancer Society guidelines for prostate cancer informed decision-making among primary care practitioners using digital detailing. Prostate cancer is the primary cause of cancer-related mortality among men in U.S. Nearly two thirds are diagnosed in men aged 65 or older. Uncertain benefit in mortality reduction from screening (Schroder et al., 2009; Andriole et al., 2009).

106 Digital Detailing

107

108

109 Usability Evaluation Planning: Locate existing prototypesConduct Focus Groups among age-eligible men (2) Develop Electronic Prototype (3) Evaluate Usability: Heuristic (Expert) Evaluation Community Stakeholder Review

110

111 Summary of the findings for IDM around Prostate Cancer ScreeningWhile most PCP’s in this urban sample are testing for prostate cancer using the PSA, most also stop by age 65. The physician is a central cue for CaP screening. Most physicians viewed the interactive digital detailing program very briefly, across few cases, with materials on cultural competence their central focus.

112 Practice Facilitation“Practice facilitators are specially trained individuals who work with primary care practices “to make meaningful changes designed to improve patients’ outcomes. [Practice Facilitators] help physicians and improvement teams develop the skills they need to adapt clinical evidence to the specific circumstance of their practice environment” --DeWalt, Powell, Mainwaring, et al. 2010

113 Evidence for the effectiveness of practice facilitation in primary care changeA  recent  meta-analysis of studies of PF within primary care settings concluded that primary care practices are almost three times as likely to adopt evidence-based guidelines through PF compared with no-intervention control group practices (Baskerville et al., 2012) Practice facilitators increased the delivery rates of preventive services and also improved relationships and communication between providers, improved chronic disease management, and facilitated system-level improvements (Nagykaldi, Mold, and Aspy. 2005). The National Demonstration Project looked at the impact of PF on practice redesign among early adopter practices and found that facilitation increased the practices’ capability to make and sustain change and increased organizational capacity to adopt new methods and service models, compared to non-facilitated practices (Nutting et al., 2010).

114 What do Practice Facilitators do?Assess practice and give feedback to practice Support introduction of new processes related to PCMH: team based care, population management, improved access, self-management support Develop QI teams Identify and spread “exemplary” practice Develop reporting systems to track practice progress Facilitate local learning collaboratives Training on QI, new processes (empanelment, team based care, etc) Identify and engage resources for practice Executive coaching on change management Project management HIT optimization

115 “Personalizing” Approaches to Practice ChangeProposed Effects of QI Interventions on Change Elements “Personalizing” Approaches to Practice Change Priority Change Capacity Change Process Content Performance Feedback Academic Detailing Practice Facilitation Local Learning Collaboratives HIT Support Solberg, 2007; James Mold, 2012 (unpublished)

116 Policy changes are necessary to implement personalized medicine in clinical practicesAccess to and support of knowledge and skill development in pharmacogenomics;. Access to and usability of clinical decision support tools, such as electronic medical records, to enable the interpretation and clinical use of quality testing. Redesign of and increased access to commercial software packages that rapidly and reliably analyze and interpret the immense amount of data generated with genome-wide typing or sequencing to reduce clinical data overload. Clinicians have limited time spend with their patients trying to support and sort through complicated bodies of information. CER not yet conducted comparing performance feedback, practice facilitation, local learning collaboratives, and Health Information Technology support and ACADEMIC DETAILING. Secetary’s Advisory Committee on Genetic Testing (SACGT)

117 Conclusions Promise of Academic Detailing to increase breast, colorectal cancer screening in accord with ACS guidelines. Evidence is thus far insufficient regarding cervical cancer. Potential influence on reducing cancer-related racial/ethnic disparities Emerging evidence for impact of Practice Facilitation on the provision of primary care, including cancer screening. Questions remain about what approach works best, in what kinds of policy and patient contexts, practices, among which kinds of providers and patients, with which outcomes: to personalize medicine at the practice level.

118 Prevention Practice in Sherri Sheinfeld GorinFor more information: Prevention Practice in Primary Care Sherri Sheinfeld Gorin NY: Oxford University Press forthcoming, November,2013

119 12/20/2017 Collaborators Sherri Sheinfeld Gorin, Ph.D., Principal Investigator, studies conducted at Columbia University (CU), Herbert Irving Comprehensive Cancer Center (HICCC), currently: NCI, ORB (SAIC), Director, NYPAC Alfred Ashford, M.D., Harlem P& S, CU, HICCC Andrew Dannenberg, Cornell Weill Medical College, HICCC Donald Gemson, M.D.* Rafael Lantigua, M.D., CU, HICCC Alfred Neugut, MD, PhD, CU, HICCC Biostatisticians, Economists, Bioinformaticians Bin Cheng, Ph.D., biostatistician, CU Yalini Sen, Ph.D., bioinformatics, SUNY Downstate Andrea Troxel, PhD, biostatistician, University of Pennsylvania Josh Graff Zivin, Ph.D., economist, UCSD New Jersey PI’s Ana Natale Pereira, M.D., UMDNJ Diane DeCosimo, M.D., UMDNJ Intervention Team Rebeca Franco, M.P.H. Project Director N. Muhammet Kanibar, M.D. Community Health Educator Sharon Guilfoyle, M.P.H.* Farida Hajiani, M.B.B.S., M.P.H. Community Health Educator Caroline Omotoso, M.B.B.S., Recruiter Medical Audit and Analysis Team Savita Gopal , (medical student) Medical Chart Auditor Julia Heck, Ph.D., Data analyst Rhadjena Hilliard, M.D., Medical Chart Auditor Keiko Honda, Ph.D., M.P.H., Post-doctoral fellow Gylynthia Trotman, M.D., M.P.H., Medical Chart Auditor * deceased National Cancer Institute

120

121 Next Speaker: Dr. Erica BreslauShe will discuss the implementation of personalized medicine in primary care through academic detailing and practice facilitation

122 What does the Future Hold?Erica S Breslau, PhD, MPH National Cancer Institute Division of Cancer Control and Population Sciences Process of Care Research Branch Bethesda, MD

123 We heard… There is a need to expand research that informs what it means to screen with considerations for: Factors beyond age Integration of decision aids Application of academic detailing

124 We also need to consider…Addressing over- and underscreening Follow-up to abnormal results

125 Centerpiece of ACA A centerpiece of the Patient Protection and Affordable Care Act (ACA) of 2010 is the focus on preventive services as a way to foster optimal health and well-being.

126 Cancer screening as opportunity“Even modest efforts to implement known tactics for cancer prevention and early detection could result in up to a 29 percent drop in cancer deaths in about 20 years.”  Institute of Medicine. National Cancer Policy Board, 2003

127 Screening as a process Screening is often viewed as a single test or event, rather than a multidisciplinary care process.  The process begins when an individual considers screening, then undergoes testing, receives a diagnosis, and follows up on results and/or is referred to treatment.

128 Multilevel Model: Personalized Cancer Screening with Older AdultsLife expectancy Functional status Comorbidity Age Biologic/physiological Cognitive status Health literacy General health perceptions Training/expertise Interpersonal skills Knowledge of benefits/harms of screening Leadership Integrated EHR Culture of patient engagement Benefit of screening Harm of screening Values & preferences Care communication Elicit patient goals Team support Performance measures/incentives Feedback to providers High Quality Decision Patient understands risks/benefits Decision concordant with patient preferences According to estimated benefits and harms According to national guidelines According to compliance with health system protocols Optimal Care Improved survival Early detection at curable phase Improved Quality of Life Avoid unnecessary testing Sub-optimal Care Overdiagnosis Missed diagnosis Reduced Quality of Life Increased mortality Inputs Factors Related to Optimal Care Use/non-use of cancer screening Health Outcomes Proximal Outcomes Intermediate Outcomes Primary Health Outcomes Input Patient Levels of Influence Provider Health System

129 Goal: High quality decisions and outcomesHigh Quality Screening Decision Making Process Patient’s health considered Patient informed of risks and benefits Patient preferences incorporated Guidelines and evidence basis High Quality Screening Outcomes Early detection Avoided overdiagnosis Improved or maintained QoL Low Quality Screening Decision Making Process Patient’s age is sole factor affecting physician recommendation Patient not informed of risks and benefits Patient’s preferences are not incorporated Low Quality Screening Outcomes Over or underdiagnosis Patient experiences physical, psychological, financial harm Mortality

130 Sherri Sheinfeld-GorinErica S. Breslau Mara Schonberg Sherri Sheinfeld-Gorin Louise C. Walter