REALIZING SHARED DECISION-MAKING IN PRACTICE

Shared decision-making (SDM) is integral to clinical practice. In medical training, clinicians are encouraged to engage patients in SDM so that clinical care is consistent with patients’ values and preferences. Professional societies and those generating screening and treatment guidelines specifically recommend SDM. Recently, reimbursement from Medicare for lung cancer screening is contingent on SDM.¹

Shared-Decision Making in Clinical Practice

While particular definitions and articulations of SDM vary, the core attributes of explaining different clinical options and taking explicit steps to elicit patients’ values and preferences rest on solid ethical grounds. SDM is intended to respect patients’ autonomous preferences; and is also supported by the ethical principle of beneficence as it is likely to increase patients’ adherence to treatment plans. As such, SDM is poised to help protect the rights and welfare of patients.

Nevertheless, data repeatedly show that SDM is rarely achieved in practice;² the reasons for this are likely multifactorial. One possible reason might be that measuring SDM using observed dialogue underestimates the degree to which SDM is achieved. Studies show that patients tend to think they have been involved in making decisions when direct observation suggests they have not. This may be because patients are unaware that a decision was made, the measurement standards for observed behavior are too dogmatic, or both. In terms of measurement, some coding schemes require explicit articulation of actions that are often conveyed implicitly. For example, ensuring that patients understand there is a decision to be made may be achieved implicitly through discussion of the pros and cons of various choices. Explaining to patients that they have a role in decision-making or assessing their desired role in making decisions, which are familiar criteria for evaluating SDM, may be unnecessary when it clear that patients understand this by asking questions and expressing their ideas. Although explicit communication about these issues may have additional value, the potential added benefit is unclear and clinicians may sense a marginal return on the required investment of time.

Practical Suggestions for Enhancing Shared Decision-making

Setting aside technical issues over how SDM ought to be measured, and even granting that the actual degree to which SDM occurs might be higher than studies suggest, it still seems clear there is less SDM than there should be. When SDM does not occur, diagnostic and treatment courses may be inconsistent with what patients would want if they had been informed and involved. This problem deserves attention. Among a complex set of factors influencing the uptake of SDM, two main limitations stand out. First, clinicians are under a great deal of time pressure and meaningfully involving patients in decisions requires time.³ Second, clinicians may be unaware that some, even routine, decisions are sensitive to a patient’s preferences. For example, some clinical guidelines make unequivocal recommendations about screening or treatment interventions, such as routine mammography or lipid-lowering treatments, for certain patients. These decisions thus seems straightforward to clinicians, who may overestimate the benefits and not fully understand the tradeoffs involved. Given these issues, five potential solutions may enhance patients engagement in their care.

Professional guidelines that call for SDM should include specificity of the task.

Calls for SDM in guidelines and recommendations, which appears to becoming more common, alone are insufficient. To be meaningful, a specific recommendation for SDM should outline clearly the particular values, as well as the risks, benefits and consequences of different decisions for patients. Such an approach would better position clinicians to understand the rationale for SDM and help prioritize it along with other issues and concerns. It would further be helpful if professional organizations that generate recommendations for SDM also identified or created patient decision aids that can guide clinicians and patients through the process.

Use decision aids.

Increasing the availability and routine use of patient decision aids would help patients engage more meaningfully in SDM. Studies demonstrate that the risks presented by clinicians in patient encounters are rarely comprehensive.⁴ An unstructured conversation without well-developed, comprehensive written materials is not the best format for conveying the complexity of information that a patient needs to absorb to be able to participate meaningfully. One systematic review found that the use of decision aids nearly always improved the quality of SDM concluding: “Therefore, it seems unrealistic to ask healthcare providers to bear the responsibility of involving their patients in health-care decisions single-handed – the patients themselves and communication tools are also a big part of the solution.”²

Prioritize decisions that require SDM.

It is impractical to engage explicitly in SDM for every clinical decision. A standard for SDM that all decisions should be shared, even routine and obvious ones (such as treating cellulitis with antibiotics), seems impossible to meet and would likely frustrate clinicians (and potentially patients) aiming to deliver good care. There is limited time in each encounter. As Schneider noted, “Every time a doctor listens to a heart, palpates a liver, or reads an EEG, a decision follows about whether there is a problem worth pursuing. Patients cannot make an informed choice about each such issue.”⁵ For every moment spent reviewing the risks and benefits of each basic decision, there is a moment not spent doing something potentially more valuable. Braddock et al. described a graduated standard of communication behaviors for SDM with increasing decision complexity.⁶ Whitney et al. suggested that SDM be reserved for decisions where there is clinical uncertainty or equipoise,⁷ and others have proposed using SDM for ‘preference-sensitive’ decisions. More work needs to be directed at when and how to make these tradeoffs. In the meantime, it would seem reasonable for clinicians to prioritize explicit SDM efforts for clinical decisions that have substantial consequences for patients and are likely to be preference sensitive.

Create an interpersonal environment that facilitates engagement.

The ideal environment for SDM is egalitarian, respectful and warm, and persists throughout the medical encounter. When multiple decisions are likely to be made over time, including those that are routine or basic, an explicit acknowledgment about how the clinician plans to approach decision-making with the patient, along with an open invitation for the patient to engage, might be useful. For example, “Whenever I think that there are more than one reasonable option, I’ll tell you about those options and you can weigh in on what you think is best for you. Whenever I think a situation is straightforward, I will make a recommendation for what I consider standard care. Even then, if I suggest something that doesn’t seem right for you, you should let me know and we can talk about all the other possible options. How does that sound?” Thereafter, the clinician could proceed without engaging in SDM for every simple decision, but patients would likely feel more in control and free to ask questions and disagree when they wanted to do so.

Give recommendations with prudence.

Some communication models regarding SDM seem asynchronous with the moral intuitions of clinicians, which may contribute to underutilization of SDM. For example, strict SDM standards do not include behaviors that provide emotional support to help patients make decisions. Some standards for SDM even discourage clinicians from making a recommendation at all. However, patients often want a recommendation, and failure to provide one could create substantial emotional distress for the patient as well as conflict for the clinician concerned about unduly influencing patients by imposing their own values. There are, of course, communication strategies that help the patient reason through options, such as, “if you’re the type of person who doesn’t like the idea of taking a medication every day, then you might want to…” or “if you’re afraid of anesthesia, then it would also be reasonable to…” or “if you’re not able to take the time off work right now, then maybe…”

Conclusions

SDM is a means to an end. The principal goals of SDM are to respect patients as individuals and to deliver care consistent with their values and preferences. Achieving these goals will sometimes involve explicitly engaging patients in decisions. But decision-making can be emotionally demanding, and imposing a standard by which patients are expected to engage in all (or even most) decisions is not only unrealistic and inefficient, but also potentially burdensome to patients. SDM might be better realized in practice by: including specificity for particular decisions in professional guidelines; using decision aids for consequential choices; prioritizing decisions that require SDM; creating interpersonal environments that facilitate engagement; and giving clinical recommendations with prudence.