The influence of values in shared (medical) decision making

A systematic meta-study based on 22 qualitative papers revealed three value categories that patients consider to be important (Bastemeijer et al. 2017): values related to the individual (I—patient), values regarding the physicians’ behaviour (II—professional), and values related to the interaction between patients and physicians (III—interaction). Further, specific values could be identified, which are broadly assignable to these three realms but are also overlapping because they are interrelated. Seven key values were the result of the review: uniqueness and autonomy (I), compassion, professionalism, and responsiveness (II), and partnership and empowerment (III). These universal values indicate that patients wish to be viewed as unique and autonomous individuals, who are treated professionally with respect and compassion and can at least decide if they want to take part in the decision-making process (Bastemeijer et al. 2017).

Another study concerned with patients’ values asked cancer patients what their relatives, friends, and physicians should know about them and what mattered most to them (Naik et al. 2016). For example, if the cancer recurred, what should others know about their treatment goals? The study supplied evidence for five health-related values: self-sufficiency, life enjoyment, connectedness and legacy, balancing quality and length of life, and engagement in care (Naik et al. 2016). These results show that patients have individual values, which they will voice when asked about them. It is the physician’s responsibility to incorporate them into the SDM process (Légaré and Witteman 2013).

As the following studies demonstrate, physicians also have their own values that affect their medical decisions. Van Kleffens et al. (2004) found that the amount of pressure physicians exert on cancer patients to persuade them to undergo a particular treatment depends on the treatment goal. If there is a curative treatment goal, physicians try harder to convince patients to go along with their recommended treatment, while they respect the patient’s autonomy more when a non-curative treatment is refused. This is a clear indication of the influence of physicians’ values—in this case “as long as a medical procedure could cure your disease you should at least try it”—on the medical decision making process. Another study, which highlights this influence, found that physicians are more reluctant to engage in SDM when the patient prefers a treatment that is not recommended by the clinical guidelines (Sanders et al. 2017).

Furthermore, Hermann et al. (2015) conducted a study with 637 physicians and asked them about their values using the Portrait Value Questionnaire Revised (Schwartz et al. 2012). They divided them into two groups: one group highly valued achievement, power-resource, face (security and power by maintaining one’s public image and avoiding humiliation), and conformity to interpersonal standards. The other group scored high on the values of hedonism, conformity to rules, and universalism-concern. Results uncovered that the first group used different criteria depending on the riskiness of a certain situation (the higher the risk, the stricter the assessment standards). The second group disregarded the risk and evaluated all situations with the same standards. Despite the small number of studies in this field, their results are indicative, and suggest that physicians’ values exert influence on medical decisions.

While the influence of physicians’ values has to be critically scrutinized (Wiesing 2017), the influence of patient’s values on medical decisions is welcome, since it is crucial that the chosen option reflects the patient’s values—especially in oncology, where different treatment schemata are available. Patients have the right to an autonomous decision (Faden et al. 1986). Birnbacher (2012), however, argues that this right does not include an obligation for the patient to make an autonomous decision and that it is not permissible to educate a patient towards autonomy, except if these conditions apply: Patient autonomy ensures a better decision and it is wanted by the patient. In SDM, patients should decide together with their physicians, because only patients themselves know which option is in accordance with their values. Hence, patient involvement would lead to a better decision. Therefore, I argue that it is vital that physicians inform patients about the importance of their values, so that they also want to decide. Naturally, one cannot force patients to reflect upon their values and take part in the decision making process, but physicians should make an effort to convince them of the benefits. When it comes to value-decisions, it is perfectly understandable that patients ask physicians for their expert view, but it should not substitute their own reflection.

In the SDM model and in recommendations on how to implement it in the medical practice, only the patient’s values receive attention and should be elicited during the decision making process (Elwyn et al. 2012; Légaré and Witteman 2013). However, as shown above, the physician’s values also exert influence on medical decisions and thus, ought to be taken into account in deliberations on SDM processes. The studies also showed that there are two ways in which physicians’ values influence medical decisions:

In the first case, the physicians’ values already influenced the options that the physician is going to present to the patient. This means that the patient does not get all viable options to choose from, which clearly constitutes a violation of the patient’s right to an autonomous decision (Beauchamp and Childress 2009) and should therefore be avoided. This raises the question, how physicians can be prevented from letting their values influence their selection of treatment options. Hermann et al.’s (2015) study also showed that almost one-quarter of the physicians stated in the questionnaire that their values did not influence their medical decisions at all. Hence, they are not aware of the influence.

In psychology, numerous studies revealed that humans often do not decide rationally because they struggle with the perception, processing and memory of information (Kirchler 2011). Unknowingly, people, who have to make a decision, draw on heuristics, which are rules of thumb for decision making that only use limited information. These heuristics often lead to right decisions, but can also result in systematic misjudgements (Kahneman and Tversky 1973; Tversky and Kahneman 1974). The only way to avoid these kinds of misjudgements is to be aware of the processes that lead to them. In medicine, there is often no simple right-or-wrong-solution to a certain issue, which is why it is often hard to determine, if the decision making was somehow biased. Nevertheless, a comparison between using heuristics for “simple” right-or-wrong-decisions and being influenced by values in medical decisions can be drawn. Only the awareness of one’s own values allows for actively controlling the extent to which different values are going to be a part of the decision making process, which is similar to being aware of heuristics and not relying on them to make a decision.

Therefore, I argue that these insights should also be part of the physicians’ training to prevent them from exerting their values unreflecting in the decision making process. Ideally, the regular reflection of values becomes part of the physician’s fundamental attitude. Another strategy for making physicians aware that they are sometimes biased by their values is to integrate this knowledge into an SDM training (where the education about how to elicit patients’ values could be extended to physicians’ values). Plenty of SDM-related trainings were developed in the last years to help physicians to apply SDM-promoting techniques more efficiently (Diouf et al. 2016). The only problem with this approach is that while these trainings do result in more SDM behaviour and significantly decrease paternalistic decision making (Sanders et al. 2017) they are, however, more likely to change behaviours than attitudes (Bombeke et al. 2012). Hence, I think the implementation of a value education would be more effective in the training of physicians when they are about to develop their attitude as physicians rather than trying to tell practising physicians what to do (or even what they do wrong).

In the second case, the physicians’ values come into play after they proposed viable options to the patient. Thereby, a distinction has to be drawn between situations, in which the patient asked for a recommendation and in situations, where this was not the case. If the patient did not ask for a recommendation, the physician should not give one. Physicians are required to share information, which is medically relevant for the decision (Légaré and Witteman 2013); however, it is not part of their job to try to influence patients and tell them their views unsolicited.

Admittedly, patients often ask for the physician’s opinion when confronted with medical decisions (Elstein et al. 2005), which raises the question if the physician should or should not offer recommendations in this case. Birnbacher (2012), on the one hand, argues that the autonomy of a decision depends on how independent it is from the influence of others. On the other hand, Dove et al. (2017, p. 162) state, “[…] autonomy is most useful as an ethical norm when we recognize that it does not mean simply being left alone to decide”. Therefore, it seems reasonable to say that physicians should not leave the patient alone with the decision, but should instead help them with the decision making process, especially when they are explicitly asked for such help. Ideally, physicians decide with the patient and not for the patient while exerting as little influence as possible.

To respect patient autonomy, I argue that it is necessary that physicians, who make a personal recommendation, highlight that this is their personal opinion. A patient’s decision will most likely always be influenced by a physician’s recommendation, but for patients to be able to assess the situation accordingly, physicians have to differentiate between what is medically indicated and what is their personal advice. For example, if there are multiple therapies available, but there is one most likely to be successful, physicians clearly should recommend that therapy (medically indicated recommendation, e.g., “Studies have shown that in cases like yours, this therapy works best.”). If there are multiple therapies with different advantages and disadvantages, but the physicians favour one, all therapies have to be presented as viable options. If in this case, nevertheless, a patient asks for a recommendation and the physician has a personal opinion about it, it can be shared but it has to be stated that this opinion is personal (personal recommendation, e.g., “In my experience, this treatment is the most tolerable one. But from an evidence based medicine point of view, they are all possible options for you to choose from.”). Otherwise, the physician would not respect the patient’s right to an autonomous decision.

All students shall think of (any kind of) values and write each of them onto a piece of paper. All values are then collected on a blackboard or flipchart and the students shall try to categorize them into terminal and instrumental values (Rokeach 1973). The values can also be grouped thematically. This exercise shall give the students an impression of the multitude of values.

In a next step, each student shall pick five values that best represent his or her own believes and rank the values in terms of importance. When they are done with the ranking, they shall explain it to another student and vice versa. This exercise gives the students opportunity to get in touch and reflect their own values. They will also discover that it can be difficult to put one value above another if they are all important to them (which patients often have to do when deciding in a medical matter). Furthermore, the students will see that if others value the same things as them, the ranking can still be different.

The next exercise stipulates the discussion of example cases, favourably ones from the students’ own experiences. If they do not have any practical experience it is also possible to provide own examples or fictional ones (see “Case example”). It is important that the case examples involve some kind of physician-patient-interaction and a medical decision.