Error message

The page you requested does not exist. For your convenience, a search was performed using the query cbssm med umich edu people brian j zikmund fisher phd.

Page not found

You are here

What's in a Name? A Pregnancy Scenario (Nov-07)

Tell us how you'd respond to the results of a blood test for fetal chromosomal problems. And find out how your response compares with that of participants in a national survey.

Consider the following

Imagine that you are four months pregnant. You and your partner have talked with your doctor about prenatal screening tests for your fetus. Based on your family history and personal medical history, your doctor has told you that you're at low risk (2 in 1000) of having a fetus with chromosomal problems. Chromosomal problems include such conditions as Down Syndrome. In talking further with your doctor, you decide to have a routine blood test for chromosomal problems in your fetus. This test will help to give you a better estimate of the chance that your fetus would have a chromosomal problem.

Your doctor tells you that the results of this blood test have come back "abnormal." She clarifies that the blood test showed that your risk of fetal chromosomal problems is about 5 in 1000, which is higher than the number she had told you before the test. She next asks if you are interested in amniocentesis, a medical procedure in which a small amount of amniotic fluid is extracted from the amniotic sac surrounding the fetus. This procedure can tell you for sure whether or not the fetus has chromosomal problems. However, amniocentesis has its own risks. Your doctor explains that the risk of miscarriage as a result of amniocentesis may be as high as 5 in 1000.

In these circumstances would you be interested in having an amniocentesis performed?
  • Definitely No
  • Probably No
  • Probably Yes
  • Definitely Yes

How do your answers compare?

Many women decide to go ahead and have amniocentesis. There are two things in this scenario that could influence women's decisions about amniocentesis. First, the doctor described the test as "abnormal", a label that may increase worry about the possibility that the fetus would have a chromosomal problem. Second, the risk estimate of 5 in 1000 was higher than the original estimate of 2 in 1000, which also may increase concern.

CBDSM researchers, led by Brian Zikmund-Fisher, wanted to know how much influence labels such as "abnormal", "normal", "positive", or "negative" might have on people's decisions in situations like the one described above. To test this, they gave one group of women a scenario just like the one you read. In this scenario, the test results were described as either "abnormal" or "positive" before the risk estimate of 5 in 1000 was given. A second group of women read the same scenario, but in their scenario, the doctor presented only the numeric risk estimate, without any label.

Women whose test results were introduced using a qualitative label ("positive/abnormal") were significantly more worried - and significantly more likely to choose to have amniocentesis - than women who were told only the numeric risk estimate, without any label. Note that all of the women in this survey were told that they had the same final risk: 5 in 1000. The decision of the women in each group should have been the same, but adding that one qualitative label significantly changed what the women in the study decided to do.

Interestingly, the CBDSM researchers also found a reverse effect when test results were introduced with the labels "negative" or "normal." These labels tended to make women less worried and less likely to have amniocentesis than women in a comparison group. Again, these results show that adding a one-sentence introduction with a qualitative label could significantly change people's decisions.

Read the article:

Does labeling prenatal screening test results as negative or positive affect a woman's responses?
Zikmund-Fisher BJ, Fagerlin A, Keeton K, Ubel PA. American Journal of Obstetrics and Gynecology 2007;197(5):528.e1-528.e6.

Funded by National Institutes of Health.

Funding Years: 2013-2017.

The clinical management of patients with cancer does not entail a "one size fits all" approach. In fact, studies of the genomic landscape of human cancers have demonstrated that cancers can have a multitude of mutations, a subset of which may be "actionable" with current drugs. Thus, the personalization of therapy for cancer will require molecular characterization of unique and shared genetic aberrations. In particular, patients who have advanced / refractory cancer and are candidates for clinical trials could potentially benefit by identifying eligibility for "targeted" drugs based on the "actionable" genesin their specific tumor. Growing technological advances in genomic sequencing has now made it possible to consider the use of sequence data in a clinical setting. Thus, the translation of high throughput sequencing would support a "personalized" strategy for cancer. However, the translation of clinical sequencing bears unique challenges including identifying patients who could benefit, developing informed consent and human subjects protections, outlining measurable outcomes, interpreting what results should be reported and validated, and how results should be reported. This proposal brings together expertise at the University of Michigan including clinical oncology, cancer genetics, genomic science/bioinformatics, clinical pathology, social and behavioral sciences, and bioethics in order to implement this clinical cancer sequencing project. We have focused our clinical sequencing effort on sarcomas and other rare cancers as this is an area of clinical strength at Michigan. Three integrated Projects have the following themes: Project 1) "Clinical Genomic Study" will identify patients with advanced or refractory sarcoma or rare cancers who are eligible for clinical trials, consent them to the study obtain biospecimens (tumor tissue, germline tissue), store clinical data, and assemble a multi-disciplinary Sequencing Tumor Board to deliberate on return of actionable or incidental genomic results; Project 2) "Sequencing & Analysis" will process biospecimens and perform comprehensive sequencing and analysis of tumors to identify point mutations, copy number changes, rearrangements/gene fusions, and aberrant gene expression under CLIA/CAP guidelines; Project 3) "Ethics & Psychosocial Analysis" will evaluate the clinician and patient response to the informed consent process, delivery of genomic sequence results, and use of genomic results.

PI(s): Arul Chinnaiyan (Main Study PI), Scott Roberts (Project 3 PI)

Co-I(s): Ajjai Alva, Rashmi Chugh, Ray De Vries, Jeffrey Innis, Lakshmi Kunju, Rohit Mehra, Nallasivam Palanisamy, Dan Robinson, Moche Talpaz, Scott Schuetze, David Smith, Elena Stoffel, Scott Tomlins, Brian Zikmund-Fisher

PIHCD: Sarah Alvarez

Thu, November 05, 2015, 2:00pm
Location: 
B004E NCRC Building 16

Sarah Alvarez, a fellow at Stanford and formerly of Michigan Radio, will  present her work on creating a news product that can meet the information needs of low-income news consumers. Specifically her focus is on how to use data to discover which issues or systems information gaps exist for low-income news consumers and once the gaps are identified how the information should be presented to help people understand the information and use it to make decisions.

If you plan to attend this meeting please e-mail Nicole Exe at nexe@umich.edu by Monday November 2. If you decide to attend after that date you are still welcome and do not need to e-mail.

CBSSM Seminar: Aaron Scherer, PhD

Wed, February 03, 2016, 3:00pm to 4:00pm
Location: 
NCRC, Building 16, Room 266C

Aaron Scherer, PhD


CBSSM Postdoctoral Fellow

The Language of Medicine

Is the way we talk about health and medicine simply expressive or does the language we use actually change how we perceive and respond to health risks and medical interventions? Aaron Scherer will discuss a number of studies that explore how metaphors, labels, and explanations may shape our health-related perceptions and behavior.

Adult Ethics Committee

The Michigan Medicine Committee advisory groups are appointed by the Hospital's Office of Clinical Affairs. They review ethical or moral questions that may come up during an adult patient's care. The consultants facilitate communication among adult patients, their families and the treatment team to assist everyone in making appropriate choices when difficult decisions need to be made. The Committee's goal is to help everyone decide the right thing to do. The Michigan Medicine Adult Ethics Committee is a sub-committee of the Executive Committee on Clinical Affairs as determined by the Medical Staff Bylaws.

About Us

Sometimes patients, families and staff have very difficult choices and ethical questions they need to talk about. Discussions with the Ethics Committee can be helpful and reassuring when a difficult choice must be made (for example, questions on end-of-life care, or issues of confidentiality). The goal of the Committee is to facilitate communication among adult patients, their families and the treatment team to assist everyone in making appropriate choices, as well as to assist Michigan Medicine in complying with ethical regulatory standards, when difficult decisions need to be made. The Committee provides consultation to the treatment team, patients and families on ethical, moral or philosophical problems and issues encountered in the course of managing inpatient and outpatient care.

Committee members include physicians, residents, nurses and social workers, as well as medical students, an attorney/compliance officer, a chaplain, a medical ethics professor and members from the community.

The Adult Ethics Committee meets on the third Tuesday of the month, form 12-1:30pm, at University Hospital in dining room D, if you would like to attend as a guest, please contact Amy Lynn @ lynnam@med.umich.edu

What happens when a meeting with the Ethics Committee is requested?

The consultants on call review the patient's medical situation and treatment options. In addition, concerns and feelings of the patient, family members, and the health care team are discussed. Members of the committee may visit with patients, families and medical personnel to discuss these concerns.

Ethics Committee members discuss the information which has been gathered. The Ethics Committee makes suggestions about the best course of action. Often there are a number of options available in the course of a patient's care. Final decisions are made by the patient, family and the health care team.

Request a Consult

Monday-Friday
8:00 a.m. - 5:00 p.m. Call 734-615-1379
After normal business hours, please call 936-6267 and ask for the clinical ethicist on call to be paged.

Resources

Financial Assistance

Non-Beneficial Treatment

Advance Directives

Committee Bylaws

 

For upcoming Bioethics Grand Rounds see Events

Angela Fagerlin, PhD

Alumni

Dr. Fagerlin served as Co-Director of CBSSM from 2010-2015. She is currently Chair of the Department of Population Health Sciences at University of Utah School of Medicine and Research Scientist, Salt Lake City VA Center for Informatics Decision Enhancement and Surveillance (IDEAS)

Last Name: 
Fagerlin
Research Projects: 

Pictographs/Icon Arrays

Pictographs and icon arrays are two names for a type of risk communication graphic that CBSSM investigators have developed and extensively tested. Because pictographs are made up of a matrix of unique elements representing individual units (people) within the at-risk population, they accurately communicate exact percentages while simultaneously conveying “gist” impressions derived from the relative proportion of colored vs. uncolored area in the graph. Click here to learn more and create your own downloadable pictograph images.

 

Pictographs combine some of the best elements of alternate communication formats such as tables or bar charts. A pictograph is made up of unique icons representing individual units (people) within the at-risk population. As a result, it accurately communicates exact percentages the way a table does. However, pictographs also convey “gist” impressions derived from the relative proportion of colored vs. uncolored area in the graph. As such, they are similar in effectiveness to bar graphs and other area or height-based graphics. Furthermore, pictographs are like pie charts in that they represent the entire risk denominator visually, unlike bar charts which focus attention primarily on the risk numerator.

CBSSM researchers have shown that using pictographs in risk communication contexts can be used to effectively communicate the incremental benefit of risk reducing treatments (Zikmund-Fisher, 2008) and the risk of developing side effects from medications, especially when multiple colors are used to distinguish the incremental risk caused by treatment (Zikmund-Fisher, 2008). Pictographs can also limit the biases induced by the presence of powerful anecdotal narratives of former patients (Fagerlin, 2005) and incremental risk formats (Zikmund-Fisher, 2008). In a study that directly compared graphical formats, pictographs were also the only graphical format that supported acquisition of both verbatim and all-important “gist” knowledge (Hawley, 2008). Another study (2010) showed that simpler pictographs (ones that showed a single risk) appeared to be more effective than more visually complex pictographs that used multiple colors to show different risks simultaneously. In a similar vein, two studies (2011, 2012) have found advantages of using static pictographs instead of more complex animated or interactive versions (perhaps because these elements distract attention from the part-whole relationship that represents the risk being communicated).

CBSSM researchers are not alone in our use of pictographs. Other researchers have shown that image matricies of this type are easier to interpret quickly and accurately than other formats (Feldman-Stewart, 2007), are sometimes preferred by patients (Schapira, 2006), and may reduce side effect aversion in treatment decision-making (Waters, 2007). More recent work has shown that icon arrays overcome some of the barriers to comprehension caused by low numeracy (e.g., Galesic & Garcia-Retamero, 2009 & 2010; Garcia-Retamero & Galesic, 2009). In fact, it appears that high numeracy and low numeracy people use pictographs in different ways (Hess, et al, 2011).

To encourage broader use of pictographs in risk communication and medical decision-making in general, CBSSM has collaborated with the UM Risk Center to develop Iconarray.com, a web-based application that enables people to develop and download their own tailored icon array graphics. A companion site, clinician.iconarray.com, enables clinicians (or anyone else) to make side-by-side icon array displays for use in consultations in less than 1 minute.

 

Supporting information for: 2017 CBSSM Research Colloquium and Bishop Lecture (Norman Daniels, PhD)


"Setting priorities for Medicaid: The views of minority and underserved communities"
Presenter: Susan Goold, MD, MHSA, MA


Co-authors: Lisa Szymecko, JD, PhD; H. Myra Kim, ScD; Cengiz Salman, MA; A. Mark Fendrick, MD; Edith Kieffer, MPH, PhD; Marion Danis, MD, Zachary Rowe, BBA


Setting priorities for state Medicaid programs challenges policy makers. Engaging beneficiaries affected by tradeoffs could make allocations more just and more sensitive to their needs. 

Academic-community partnerships adapted the simulation exercise CHAT (CHoosing All Together) to engage community members in deliberations about Medicaid spending priorities.  After an informational video about Medicaid, individuals and deliberating groups choose from a menu of spending options constrained by limited resources. We randomly assigned participants from low-income communities throughout Michigan to participate in CHAT with (n=209) or without group deliberations (n=181) in English, Spanish or Arabic. Data collection included pre- and post-CHAT individual priorities and group priorities.

Low-income participants ranged from 18 to 81 years old (Mean 48.3); 61.6% were women. Over half (56.7%) self-identified as white, 30.8% African-American, 17.3% Hispanic, 9.2% Native American, and 12.1% Arab, Arab-American or Chaldean. Most (65.9%) had a chronic condition and 30.3% reported poor or fair health.

Before CHAT, most participants prioritized eligibility consistent with Medicaid expansion. They also prioritized coverage for a broad range of services. Most accepted daily copays for elective hospitalization (71.6% deliberators, 67.9% controls) and restricted access to specialists (60.2% deliberators, 57.4% controls). Deliberators were more likely than controls to increase, after deliberations, what they allocated to mental health care (between arm difference in allocation=0.22, p=.03) and eligibility (between arm difference in allocation=0.18, p=.04). Deliberating groups also prioritized eligibility; only 3 of 22 chose pre-expansion eligibility criteria, and 9 of 22 chose to expand eligibility further.

Members of underserved communities in Michigan put a high priority on Medicaid expansion and broad coverage. When given the opportunity to deliberate about priorities,  participants increased the priority given to expanded eligibility and coverage for mental health services.


"How Acceptable Is Paternalism? A Survey-Based Study of Clinician and Non-clinician Opinions on Decision Making After Life Threatening Stroke"
Presenter: Kunal Bailoor, MD Candidate


Co-authors: Chithra Perumalswami, MD, MSc; Andrew Shuman, MD; Raymond De Vries, PhD; Darin Zahuranec, MD, MS


Complex medical scenarios may benefit from a more paternalistic model of decision making. Yet, clinicians are taught to value patient autonomy, especially at the end-of-life. Little empirical data exist exploring opinions on paternalism.

Methods: A vignette-based survey exploring surrogate decision making after hemorrhagic stroke was administered to clinicians (faculty, residents, and nurses) at an academic health center, and non-clinicians recruited through a university research volunteer website. The cases involved an urgent decision about brain surgery, and a non-urgent decision about continuation of life support one week after stroke. Respondents rated the acceptability of paternalistic decision making, including clinicians not offering or making an explicit recommendation against the treatment, on a 4 point Likert scale.

Results: Of 924 eligible individuals, 818 (649 non-clinicians, 169 clinicians) completed the survey (completion rate 89%).  A minority of respondents (15.3%) found it acceptable not to offer surgery. Most believed it was acceptable to make an explicit recommendation that would likely result in death (73% for avoiding surgery, 69% for stopping the ventilator). Clinicians were more likely than non-clinicians to consider not offering surgery acceptable (30% vs 11%, p<0.0001). Clinicians were more likely to consider recommendations against surgery acceptable (82% vs 71%, p=0.003) and to consider recommendations to discontinue the ventilator acceptable (77% vs 67%, p=0.02). There were no differences between the nurse and physician acceptability ratings (p=0.92).

Conclusions: Clinicians and the lay public differ on the acceptability of paternalistic decision making. Understanding these differences are vital to improving communication between clinicians, patients, and families.


"Ethical Challenges Faced by Providers in Pediatric Death: A Qualitative Thematic Analysis"
Presenter: Stephanie Kukora, MD


Co-authors: Janice Firn, PhD, MSW; Patricia Keefer, MD; Naomi Laventhal, MD, MA
 

Background: Care providers of critically ill patients encounter ethically complex and morally distressing situations in practice. Though ethics committees guide ethical decision-making when conflicts arise in challenging cases, they rarely address routine needs of individual providers. Without ethics education, providers may lack skills necessary to resolve these conflicts or insight to recognize these dilemmas.

Objective: We sought to identify whether providers remark on ethical dilemmas/moral distress without being specifically prompted, when asked to comment on a recent in-hospital pediatric death. We also sought to characterize the nature of dilemmas or distress if found.

Methods: Providers involved in a deceased child’s care in the 24 hours prior to death were electronically surveyed. Questions included demographic information and free-text response. Free-text responses were thematically analyzed in Dedoose.

Results: There were 307 (35%) free-text responses in 879 completed surveys (33% total response rate), regarding the deaths of 138 patients (81% of in-hospital pediatric deaths) from November 2014 to May 2016. Multidisciplinary care team members from diverse hospital units were represented. 52 respondents described ethical challenges and/or moral distress. Disagreement/regret was a major theme, with subthemes of futility, suffering, and “wrong” medical choice made. Failure of shared decision-making was also a major theme, with subthemes of autonomy and best interest, false hope, denial, and misunderstanding/disagreement between the family and medical team. Some providers revealed personal ethical struggles pertaining to their role, including medication provision for pain at the end of life, struggling to be “truthful” while not divulging information inappropriate for their role, and determining when providing comfort care is ethically permissible.

Discussion/Conclusion: Providers experience ethical conflicts with pediatric end-of-life care but may be unwilling or unable to share them candidly. Education assisting staff in identifying and resolving these dilemmas may be helpful. Further support for providers to debrief safely, without criticism or repercussions, may be warranted.


"Capacity for Preferences:  An overlooked criterion for resolving ethical dilemmas with incapacitated patients"
Presenters: Jason Adam Wasserman, PhD; Mark Navin, PhD
 

Clinical bioethics traditionally recognizes a hierarchy of procedural standards for determining a patient’s best plan of care. In broad terms, priority is given first to autonomous patients themselves and then to surrogates who utilize substituted judgments to choose as they believe the patient would have chosen. In the absence of good information about what the patient would have wanted, clinical ethicists typically retreat to the “best interest” standard, which represents a relatively objective assessment designed to maximize benefits and/or minimize harms.  In this paper, we argue that “capacity for preferences” is a conceptually distinct and morally salient procedural standard for determining a patient’s best plan of care.  We build our argument on the grounds that 1) that many patients who lack decisional capacity can nevertheless reliably express preferences (an empirical claim); 2) these preferences are distinct from best interest and not reducible to best interest considerations; 3) that capacity for preferences, at a minimum, has moral valence for situations in which best interest is undetermined (and we argue this happens more frequently than commonly recognized); and, finally, 4) that capacity for preferences in incapacitated patients lacking reliable or valid surrogates might even subvert a best interest course of action in some cases.  Some precedent for our analysis can be found in the concept of pediatric assent. However, the idea that patient preferences matter morally has broad application for adult patients, including for those with advanced dementia and other mental illnesses that preclude capacity for decision-making.

CBSSM Seminar: Jacob Solomon, PhD

Thu, November 19, 2015, 3:00pm to 4:00pm
Location: 
NCRC, Building 16, Room 266C

Jacob Solomon, PhD


CBSSM Postodoctoral Fellow

Title:

Designing the information cockpit: The impact of customizable algorithms on computer-supported decision making

Abstract:

Intelligent systems that provide decision support necessitate interaction between a human decision maker and powerful yet complex and often opaque algorithms. I will discuss my research on end-user control of these algorithms and show that designing highly customizable decision aids can make it difficult for decision makers to identify when the system is giving poor advice.

CBSSM Seminar: Reshma Jagsi, MD, PhD

Wed, May 18, 2016, 3:00pm to 4:00pm
Location: 
NCRC, Building 16, Room 266C

Reshma Jagsi, MD, PhD
Associate Professor, Radiation Oncology

"Stewardship and Value:  Are we choosing wisely in managing breast cancer?"

Abstract: This lecture will begin with a brief discussion of the moral foundations of physicians' obligations to serve society, in addition to the patients they directly serve.  It will then consider analogies between financial stewardship and antibiotic stewardship, and it will conclude by focusing on several examples of opportunities for better physician stewardship in breast cancer, including slow uptake of short courses of breast radiation and rapid increases in the use of bilateral mastectomy for unilateral disease.

Pages