Incorporating Patient Preferences

Key Points

• There has been a cultural shift from paternalistic to shared decision making.

• The degree of shared decision making depends on risk, benefit, and a patient's preferences.

• Decision aids can help patients understand complex probabilities.

Until the 1980s, medical decision making was physician driven. Such decision making required and perhaps engendered patients' trust in their physicians. With such paternalistic care, patients effectively ceded decisional authority to their physician through implied consent. Physicians made the decisions, patients accepted them, and most patients preferred it that way. Many patients still prefer this type of interaction (Goldfarb, 2004). How many times do family physicians explain the risks and benefits of a procedure, only to have the patient respond, "So Doc, what would you do?"

In 1980 a small, influential group was formed: the Society for Medical Decision-Making. Over the years, the society's work has spawned two other movements: evidence-based medicine and shared decision making. Evidence-based medicine (EBM) seeks to provide physicians with the empirical data necessary to make wise clinical decisions through critical appraisal of the literature. Shared decision making occurs when physicians communicate the evidence to patients with sufficient clarity that they can become fully informed partners in their own health care decisions. Probabilistic evidence


About WHI Study Findings

Study updates

Frequently asked Questions

WHI in the news

WHI quilts

Related sites

Senior resources

Recent Findings

Estrogen with and without Progestin and urinary incontinence

Updated results from the WHI showing the effect of Estrogen with and without Progestin on the incidence of urinary incontinence were published in the February 23, 2005 issue of the Journal of the American Medical Association (JAMA)

The researchers concluded that postmenopausal hormone therapy with either conjugated equine estrogen combined with progestin or alone should riot be prescribed to women to prevent or treat urinary incontinence.

• Summary of these findings for WHI participants » Abstract of scientific paper in JAMA

Estrogen plus Progestin and venous thrombosis

Updated results from the WHI showing the effect of Estrogen plus Progestin on the incidence of venous thrombosis were publsihed in the October 6 issue of the Journal of the American Medical Association (JAMA).

These results show that after an average of 5.6 years, 18 additional women per year developed VT per year among 10,000 women using E + P

» Summary of these findings for WH participants

• Abstract of scientific paper in JAMA

In the months and years ahead there will be a great deal more WHI research made available to the public.

In order to provide WHI particiapnts a way of obtaining information about research results directly from the study rather than through news reports, we ask the scientists to summarize their findings especially for WHI participants.

These summaries are posted on this site on an ongoing basis as soon as they are available.

Figure 10-2 Women's Health Initiative (WHI) site.

alone should not guide decision making; patient preferences and values must be considered for a decision to be shared. As a result, shared decision making has been called an "ethical" process. The physician not only informs the patient before making the decision, but also seeks to empower the patient to become an active participant in decision making.

"Doctor," Mrs. Smith interjects as you finish clicking through the guideline pages. "I know you're concerned about me being on these hormones. But, I really am doing well, and I don't want to stop them unless I have to."

The cultural change toward active patient participation in medical decision making occurred rapidly. One survey of patients seeing their physician for chronic disease management between 1986 and 1990 found that 69% preferred "to leave decisions about my medical care up to my doctor." In contrast, a study only 10 years later found that 84% of women with node-negative breast cancer preferred a shared role in decision making regarding adjuvant chemotherapy. One week after using a decision aid to explore the risks and benefits, an additional 10% said they preferred a shared role in the decision (Arora and McHorney, 2000). This cultural shift toward greater patient involvement in clinical decision making may stem from the increased availability of medical knowledge in the public domain and an increasingly con-sumerist society (Guyatt et al., 2004).

Physician reactions to this cultural change range from self-interest in avoiding legal liability to selfless beneficence in seeking to include patient preferences (McGuire et al., 2005). Increased patient participation in medical decision making is associated with increased trust, higher patient satisfaction, and greater compliance with treatment, particularly lifestyle changes. In some cases, this leads to an improvement in disease outcomes (Epstein et al., 2004; O'Conner et al., 2003; Trachtenberg et al., 2005).

Shared decision making is substantially different from informed consent, a legal doctrine focused on disclosure of risk and benefit to the patient or the surrogate rather than on partnership in medical decision making. Shared decision making also differs from medical consumerism, in which patients obtain information from various media sources before actively interviewing their physicians and making their own decisions about their health care. Shared decision making involves the creation of a physician-patient partnership, one of the six principal aims of health care in the 21st century, as identified by the Institute of Medicine's 2000 report (Sheridan et al., 2004).

A busy family medicine practice involves straightforward and complex medical decision making. The degree to which decision making is shared with the patient varies according to the risk of the intervention, certainty of the benefit, and the patient's desire for autonomy. Figure 10-3 illustrates how these three dimensions of risk, certainty of benefit, and patient preferences interact to determine whether shared decision making is necessary.

For decisions involving low risk and a single best course of action, shared decision making is unlikely to be helpful. Physicians are free to make the decision as long as they explain what needs to be done and why. The TV image of the wise family doctor making a diagnosis of strep throat and then sitting back in his chair to elicit patient involvement in the decision about whether to treat with penicillin is laughable to most patients, of little benefit, and a sure way to put the physician behind schedule.

Conversely, when multiple alternatives of almost equal benefit exist, the decision necessarily involves patient preferences. For example, in a young healthy patient with elevated lipid levels, the patient should share in the decision regarding initiation of a statin versus continued aggressive lifestyle changes. Such involvement is more likely to yield patient compliance with the mutually agreed course of action.

When the choice involves significant risk or moral beliefs and the certainty of benefit is low, the patient should be encouraged to make the decision, with the physician acting as an informant rather than a co-participant (Whitney et al., 2004). An excellent example occurs with first-trimester screening for Down syndrome and trisomy 18.

Physicians must be flexible, adjusting to the patient's desire for involvement in decision making. Some patients want collaborative decision making, whereas others feel more comfortable with traditional physician-directed decision making (McGuire et al., 2005). This is shown as the third dimension in Figure 10-3.



Clear be


Shared decision-making

(PAP smear in elderly)

Physician decision-making

(Penicillin for strep throat)

Patient decision-making

(First trimester quad screen)

Shared decision-making

(Warfarin in afib)

Shared decisionmaking


Shared decisionmaking

Shared decision-making

Figure 10-3 Shared decision-making model.

The desire to communicate data effectively to patients has resulted in development of many decision support tools, such as pamphlets, audio/videotapes (CDs), personalized counseling, smartphone apps, and interactive computer programs on the Internet. These tools inform patients about complex risk/benefit balances and help them determine their personal preferences and reach a decision by weighing the possible outcomes. An excellent example is the Atrial Fibrillation Treatment Decision-Making Aid offered by the University of Ottawa.* This tool informs patients about atrial fibrillation and strokes, estimates risks and benefits, and helps rank the patient's outcome preferences. Based on these data, a worksheet guides the physician and patient in a shared decision-making process (Ottawa Health Decision Centre, 2006,

Because working through such decision aids can be time-consuming and impede workflow in a busy clinic, they are often best assigned as homework, with a recheck visit at a later date to engage in shared decision making.

Mrs. Smith needs more patient-oriented evidence to help her share in the decision to continue or stop the HRT. Following the links on the Ottawa Health Decision Centre website, you bring up the decision aid, "Osteoporosis Decision Aid: Should I take hormone replacement therapy?" You print it, hand Mrs. Smith a pen, and head off to see your next patient while she ponders the information.

Decision making often is not as complex as Mrs. Smith's case, and no written decision aid may exist. A family physician may only have 2 or 3 minutes during a 15-minute patient encounter to explain the complex probabilities of risks and benefits for a particular intervention. Various statistical constructs have been proposed to communicate these probabilities to patients in an understandable manner.

The ideal format depends on the patient's condition and ability to understand probabilities. The relative risk ratio (RRR), absolute risk ratio (ARR), and number needed to treat (NNT) are reasonable constructs that can communicate the expected magnitude of benefit. Although the NNT has been proposed as an intuitive concept, a Danish trial showed that patients have difficulty understanding the magnitude of osteoporosis treatment benefit when presented in terms of NNT. Conversely, when explained as postponement of hip fracture, their treatment decisions reflected a better sense of the actual benefit (Christensen et al., 2003). Sheridan and colleagues (2003) similarly found that patients frequently misinterpreted the NNT but found the ARR and RRR more intuitive. The Evidence-Based Medicine Working Group has proposed more clearly communicating treatment benefit by using the construct of "likelihood of being helped versus harmed" (McAlister et al., 2000).

When you return to her examination room 15 minutes later, a satisfied Mrs. Smith declares, "I appreciate you giving me all the information, but what do you think I should do?" You review the decision aid with her, discussing the risk/benefit ratios; you note that she has identified the benefits of decreased fractures and menopausal symptoms as more important than the risk of harm from blood clots, heart disease, stroke, or breast cancer. She states, "Doc, I know that my diabetes raises my risk of heart disease, but I just don't want to fall and break my hip and have to spend the rest of my life in a nursing home. Besides, I really enjoy not having the hot flashes."

You reflect for a moment on the Heart and Estrogen/ progestin Replacement Study (HERS) study results (Hulley et al., 1998) that showed a decrease in cardiovascular risk with HRT in years 3 to 5 of the study. You have heard some of the controversy surrounding the WHI results (2002) and the possibility that lower rates of cardiovascular disease might have been seen with longer exposures to estrogens if the trial had been continued.

You inform Mrs. Smith that you cannot really be sure, but because she has been receiving HRT so long, you do not believe her risk of heart attack is increased. You also make sure she understands the absolute risks of breast cancer and blood clots. She replies, "Well, Doc, I've been on estrogen for 25 years, and I never had a problem with my heart or my mammograms. I just don't want to wind up like my friends who've gone off estrogen and have hot flashes every night—you know what I mean?"

Clicking the pen, you say, "Well, Mrs. Smith, you seem to have a good understanding of your situation. Shall we continue that Premarin?"

Alternative Medicine

Alternative Medicine

Discover How to Heal Yourself Naturally! Safe, All-Natural Alternatives to Taking High-Cost, Side-Effect-Laden Prescription Drugs! If you are tired of paying the high cost of prescription drugs if you are tired of worrying about prescription drug side affects or if you just dont want to risk becoming dependent on prescription drugs then have I got great news for you!

Get My Free Ebook

Post a comment