Forty years on a back burner

According to Makoul and Clayton [1], the concept of shared medical decision making originated over 40 years ago in a 1982 report published by the President’s Commission for the Study of Ethical Problems in Medicine And Biomedical And Behavioral Research titled Making Health Care Decisions. [2] Here are two pertinent quotations from the report:

It will usually consist of discussions between professional and patient that bring the knowledge, concerns, and perspective of each to the process of seeking agreement on a course of treatment. Simply put, this means that the physician or other health professional invites the patient to participate in a dialogue in which the professional seeks to help the patient understand the medical situation and available courses of action, and the patient conveys his or her concerns and wishes. This does not involve a mechanical recitation of abstruse medical information, but should include disclosures that give the patient an understanding of his or her condition and an appreciation of its consequences (p. 38). 

Shared decision making requires that a practitioner seek not only to understand each patient’s needs and develop reasonable alternatives to meet those needs, but also to present the alternatives in a way that enables patients to choose one they prefer. To participate in this process, patients must engage in a dialogue with the practitioner and make their views on well-being clear (p. 44).

Musings

Sadly, as far as I can tell, we haven’t yet learned how to incorporate these ethical imperatives into routine clinical practice. It’s time to return to the drawing board. I doubt much progress will be made until decision making becomes a core subject for clinicians.

References

1. Makoul G, Clayman ML. An integrative model of shared decision making in medical encounters. Patient education and counseling. 2006 Mar 1;60(3):301-12.

2. President’s Commission For The Study Of Ethical Problems In Medicine And Biomedical And Behavioral Research Making Health Care Decisions, October1982. Available at https://repository.library.georgetown.edu/bitstream/handle/10822/559354/making_health_care_decisions.pdf?sequence=1&isAllowed=y.

Interactive guideline decision models & dashboards

In last week’s post, I proposed that guideline creation panels start including decision models and dashboards in their guideline summaries and recommendations. I included an example decision model and dashboard to illustrate how this process could work. The dashboard I included has a couple of shortcomings. It is not interactive and does not show how uncertainties can be included in a dashboard display. The purpose of today’s post is to demonstrate the basic elements of an interactive decision dashboard.

Today’s technology makes it easy to build and disseminate complex, interactive dashboards and other information visualizations of all types. One of the best resources to learn about these capabilities is Tableau Public. Users of this site can explore information visualizations about many topics and learn how to create and post their own. Use of the Tableau Public site is free but all visualizations are open to public access. Licenses for the Tableau software can be purchased for proprietary use.

I used the Tableau Public site to create a simple interactive dashboard regarding the following decision:

Imagine you have a newly diagnosed chronic disease that is causing symptoms severe enough to limit your daily activities. Fortunately,  several treatment options are available. Information about how well they work, their risk of side effects, and monthly out-of-pocket cost is summarized in the table below:

The goal of a decision dashboard is to help you compare the three drugs and choose the one that you would pick to treat your symptoms. A dashboard using the information in the table that compares the 3 drugs is shown below. Note, if you want to choose which drugs to display, use the checkboxes on the upper right.

A screenshot of the dashboard is shown below. Click this link to access the working version.

The initial display compares how the three drugs compare over three different dimensions: disease control or effectiveness, risk of side effects, and out of pocket cost. The range of possible values are graphically illustrated in the different colored bars.

Some may find the default display sufficient to choose a preferred drug. For example, someone may not be able to afford more than $100 per month, which would eliminate Drug C and probably Drug B as well. They can quickly see, however, that Drug A is the safest drug and should be as nearly if not as effective as the other two, making it a good choice.

Others may choose to make a decision by first eliminating a less desirable drug and then reviewing how the remaining two match up. For example, someone might choose to eliminate Drug C due to its high cost and high risk of side effects compared to the other two drugs. They then uncheck the box next to Drug C on the interactive dashboard and concentrate on comparing the two remaining drugs across the three decision criteria as illustrated below:

Musings

Clinical guidelines summarize and evaluate existing data to recommend how it should be applied in clinical practice. Most guidelines as currently written summarize the research findings and list practice recommendations in great detail, sometimes across the span of two or three separate articles. This format does a good job of documenting the data supporting the recommendations but is not designed to make either the data or the recommendations readily usable to support decision making in clinical practice. A simple solution to this problem is to create a decision-ready summary of the guideline data and recommendations in the form of a decision model and interactive decision dashboard.

In addition to making the guideline information more readily usable, a recommendation-based decision model could also be used to collect the perspectives and judgments of diverse clinical decision makers. The model could be disseminated as a small, interactive file with questions regarding decision related trade offs and judgments, and a mechanism for modifying the basic model supplied by the guideline panel. Models adapted through clinical use could be saved and anonymously aggregated to provide important information that could be used to inform future iterations of the practice guidelines.

Interactive decision dashboards

Moving medical decision making into the 21st Century.

Information visualization is “the use of computer-supported, interactive, visual representations of abstract data to amplify cognition”. [1]

---

In last week’s Musing, I discussed the advantages of using information visualization as a way to quickly and easily enable busy clinical decision makers to engage in deliberative as well as intuitive decision making.

To be successful, clinical information visualizations have to be prepared in advance and easily accessible in the consultation room. With the advent of electronic medical records and ubiquity of smartphones and computer tablets, there is now a convenient, familiar delivery system available, if we choose to use it. The use of electronic tools also easily enables an additional important cognitive advantage of information visualization: the ability to interact with the data in a non-linear, self-directed manner.

For most clinical problems, several considerations will be of interest, such as the chances of a successful outcome, risks of side effects, and costs. A visual display showing a group of considerations and their respective data is called a dashboard.

Dashboards are collections of data visualizations, presented in a single-page view that imparts at-a-glance information on which users can act quickly. [2]

To illustrate, imagine a patient and her doctor are deciding which of four drugs is the best initial treatment for a new problem. Information about the treatment choices is summarized in the table below.

Instead of using this table, they choose to use an interactive dashboard displaying the same information running on a tablet computer. Note that the drug regimen data was first transformed to a numerical preference score running from 5 (most preferred) to 1 (least preferred) using input provided by the patient:

Working with the dashboard, the clinical decision makers first notice that Drug C is a dominated option, never the best on any criterion and often the worst. Therefore they eliminate it from consideration by simply checking a box on the dashboard display and concentrate on comparing the remaining three choices using the revised dashboard shown below:

At this point they note that, in terms of Effectiveness and Risk of Side Effects, the differences among the three options are small. The biggest differences are the costs, that range from $5 to $40 per month.

Drug D, the most expensive option, is only slightly easier to use than Drug B, which costs half as much. Drug D is also the option most likely to have a side effect and not the best in terms of effectiveness. They therefore decide to eliminate Drug D from consideration and focus their attention on Drugs A and B using the revised Dashboard shown below:

The tradeoffs between these two drugs are now clearly illustrated. Drug A is less expensive and safer but not quite as effective and more difficult to use. For this patient, they decide the effectiveness and side effects differences are too small to matter. The choice of treatment then comes down to whether cost or ease of use is more important. If cost is more important the better choice is Drug A. If ease of use is more important, the better choice is Drug B.

Musings

The dashboard I used for this illustration was built using Apple Numbers, a simple spreadsheet program. Adding and removing options only requires checking or unchecking a box. The process described in the illustration sounds tedious but flows quickly during a real application. In the first study I did with medical decision dashboards, the study subjects were able to analyze a dashboard comparing nine options over five criteria and choose a preferred treatment in an average of 4.6 minutes. [3]

Information visualization, particularly using modern, widely available interactive computer graphics, is a powerful tool for presenting information and using it to guide decisions. Because the most important skills involved are pre-attentive visual attributes that do not need to be learned, information visualization is an ideal way to help people, particularly those with limited numeracy skills, become better informed and engaged in choices affecting their health. Seems like a wonderful way to promote informed, patient-centered decision making in clinical practice settings. We should start using it.

References

1. Quotation from Card SK, Mackinlay JD, and Shneiderman, B. (1999) Readings in Information Visualization: Using Vision to Think. Academic Press included in Stephen Few. Now you see it. An introduction to visual data sensemaking. Second Edition. El Dorado Hills, CA: Analytics Press; 2020.

2. Dashboards: Making Charts and Graphs Easier to Understand [Internet]. [cited 2023 Feb 14]. Available from: https://www.nngroup.com/articles/dashboards-preattentive/

3. Dolan JG, Veazie PJ, Russ AJ. Development and initial evaluation of a treatment decision dashboard. BMC Med Inform Decis Mak. 2013 Dec;13(1):51.

Light and adapted to the context.

Incorporating deliberation in busy clinical contexts.

Every man hears only what he understands. ~Goethe

Research in cognitive psychology has proposed that humans make decisions using two distinct pathways: intuitive and deliberative. Intuitive decisions are fast, take place at a subconscious level and require little effort but are hard to explain to others and can be easily affected by cognitive biases and emotional states. Deliberative decisions require thought and effort. Consequently they are slower, but also more logical, explicit, and less susceptible to the adverse effects of cognitive biases and emotions. Because they are explicit, they are necessary for effective shared decision making when multiple people are involved in making a decision. [1,2]

As discussed in the October 4, 2022 Musing, one of the characteristics of decision support systems that have proven to be effective in busy applied settings is support for both intuitive and deliberative decision making processes.

The research regarding decisions made by soccer players reviewed in last week’s Musing, suggests that people can learn to use deliberative thinking methods to guide decisions in situations as dynamic and fluid as an ongoing soccer game.

Taken together, these findings raise the question of how to help providers and patients use an appropriate combination of intuitive and deliberative processes when making decisions in busy clinical settings.

As noted last week, the soccer research suggests that the answer is making the deliberative methods “…light and adapted to the constraints of the context…” [3] The simplest and fastest way to do this is probably through the creation of a decision model.

A decision model is a diagram that maps out the decision objective, the options, and the factors being used to judge which option to choose. By making the decision explicit, it is by definition a deliberative decision making procedure that fosters communication among decision makers and the integration and puts data and other decision-related information. I discussed how this process could work in the December 23, 2022 Musing.

Models can be created simply on a piece of paper, or after considerable study by a guidelines panel or similar group. A good general model to use for most clinical decisions is the decision quality chain, discussed in the December 2, 2022 Musing.

Musings

Decision models are basic elements of decision analysis – a proven set of tools to help people engage in deliberative decision making processes and, by so doing, make better choices. Why they are not a fundamental part of clinical, and especially shared decision making, is a mystery to me.

References

1. Stanovich KE, West RF. Individual differences in reasoning: Implications for the rationality debate? Behavioral and Brain Sciences. 2000 Oct;23(5):645–65.

2. Daniel Kahneman. Thinking, Fast and Slow [Internet]. Farrar, Straus and Giroux; 2011.

3. Petiot GH, Bagatin R, Aquino R, Raab M. Key characteristics of decision making in soccer and their implications. New Ideas in Psychology. 2021 Apr;61:100846.

Soccer decision making insights for clinical decision makers. A free kick.

There seems to be a specific adaptation of cognitive skills to sports-related decision making under pressure. [1]

In recent musings, I’ve been advocating teaching clinicians and patients how to be better decision makers to improve the overall quality of healthcare and foster shared decision making. In the October 4, 2022 Musing, I listed several useful takeaways derived from successful decision support interventions in other fields. Another place I’ve recently found ideas applicable to clinical decision making is the field of decision-making in sports.

In 2021, Gregory Petiot and colleagues published an article titled: Key characteristics of decision making in soccer and their implications. [1] The paper reviews literature that supports the premise that that during a game, good soccer players do not rely solely on intuitive decision making processes. Rather better players learn, and can be taught, to use a combination of rapid intuitive decision making and a streamlined deliberative process:

“Similarly, recognition, evaluation, and judgment seem to be processing mechanisms that promote better decision making as long as they are light and adapted to the constraints of the context of play and to the changing, uncertain nature of play.” [1]

Pettiot and colleagues also point out that soccer players usually have multiple options available to them. They suggest that the decisions soccer players make should not be classified as right or wrong but rather as whether or not they are coherent, i.e., decisions that support the overall team strategy for winning the game.

Like soccer players, clinical decision makers often must make decisions quickly, for example when a patient is unstable or due to the time pressures of a busy clinical setting. Clinical and soccer decisions are also similar in that, most of the time, several people are working together as a team to achieve a common objective. In healthcare, the primary teammates are the clinician and the patient; occasionally others will also be involved.

Musings

To me, this paper provides three key insights applicable to clinical decision making:

a) Decision making processes can and should be adjusted to be appropriate for the context in which they are being used.
In healthcare, variables that define the decision making context include the nature of the problem, the urgency of the situation, the stakes involved, the amount of information available, the patient-clinician relationship, and, as suggested in the January 27, 2023 Musing, the Cynefin framework decision making scenario.

b) Powerful deliberative decision making methods can be utilized in dynamic, time limited situations if they are appropriately modified.
Intuitive decision making processes are fast and require little thought, but are subject to cognitive biases and can be adversely affected by emotional states. Deliberative decision making is slower, less susceptible to bias, more likely to reflect current knowledge, and better able to address uncertainty. The soccer-related findings suggest that elements of the deliberative process can be successfully adapted for use in situations where decisions must be made even as quickly as those made by soccer players during the course of a game. It also suggests that the skills can be taught.

c) When more than one individual is involved, good decisions are choices expected to help achieve a shared goal.
In other words, all decisions involving patients should contribute to helping them achieve a healthcare goal. An important initial step in caring for a patient is establishing the goal being sought. The method the patient/clinician team uses to achieve the shared goal – the team strategy – should be adapted to fit the decision making context. From this perspective shared clinical decision making is not a separate entity; all clinical decisions should be shared. Actively engaging patients in shared decision making is a decision making strategy used to meet the demands of a specific decision making context. Fostering adoption of shared decision making in clinical practice therefore depends on improving the abilities of patients and clinicians to understand clinical decision making strategy and tactics.

Note:

The origin of the term soccer is fascinating, see references 2 and 3.

References

1. Petiot GH, Bagatin R, Aquino R, Raab M. Key characteristics of decision making in soccer and their implications. New Ideas in Psychology. 2021 Apr;61:100846.
2. Why Is Soccer Called “Soccer” Instead of Being Called “Football”? [Internet]. Soccermodo. 2021 [cited 2023 Feb 7]. Available from: https://soccermodo.com/why-is-soccer-called-soccer/
3. Why Do Some People Call Football “Soccer”? | Britannica [Internet]. [cited 2023 Feb 7]. Available from: https://www.britannica.com/story/why-do-some-people-call-football-soccer

Rethinking Decision Quality

Our motto could be the following: focus on identifying and solving problems with decision-making rather than moving the needle on decision-quality measures. [1]

In recent Musings, I have been exploring the concept of decision quality and ways to improve the quality of routine clinical decisions.

In the November 18, 2022 Musing, I suggested that finding a way to help clinicians and patients become better decision makers could speed implementation of shared decision making in clinical practice and improve the overall quality of medical decision making. In more recent Musings, I proposed using the Decision Quality Chain, developed by the Strategic Decisions Group, both to guide the clinical decision making process in real time and to measure the aggregate quality of clinical decisions.

I was, therefore, excited to find a newly published article by Peter Schwartz and Greg Sachs, titled Rethinking Decision Quality: Measures, Meaning, and Bioethics, published in the December 2022 Hastings Center Report. [1] Drs. Schwartz and Sachs are both Professors of Medicine at the Indiana University School of Medicine. In it, they analyze the concept of decision quality based on how it has been conceived and measured in the medical literature.

Schwartz and Sachs group existing decision quality metrics into three categories: Subjective measures that ask patients to reflect on a decision, Observational measures that evaluate decision quality by reviewing recorded conversations between patients and providers, and Informed-Concordance measures that objectively measure how well-informed a patient is about a decision and track whether they receive the option that best matches their preferences and values. They then discuss the advantages and disadvantages of each. Consistent with publication in a Bioethics journal, both the analysis and related discussion are focused on decision quality in terms of the bioethical goals of increasing patient autonomy and involvement in shared decision making.

The most well-known subjective measure is the Decisional Conflict Scale. [2,3] Although widely used, Schwartz and Sachs point out the shortcomings of relying on a subjective measure of how well informed a patient is and the fact that being more completely informed about a decision can decrease the measured quality of the decision by adversely affecting scores on the uncertainty and values clarification sub-scales. Measures on another commonly used subjective scale, the Decision Regret Scale [4] – which measures regret months after a decision is made – confounds measurement of the quality of decision by knowledge of the outcome. This is because, when decisions are made when the outcomes are uncertain, as the majority of healthcare decisions are, good decisions can result in bad outcomes and vice versa.

Examples of objective measures include the Informed Decision Making Scale and the OPTION scale. [5-7] While they are not subject to the limitations of the subjective measures, they are logistically difficult and run into concerns for patient privacy. Schwartz and Sachs also point out that they may not always accurately reflect a superior decision making process, particularly in the setting of an ongoing provider-patient relationship that involves multiple interactions in addition to the one being evaluated.

Problems with the Informed-Concordance measures such as the Multidimensional Measure of Informed Consent [8] include difficulties defining what knowledge is essential for the patient to know about the decision being considered and determining what the optimal choice is for any given patient.

Based on their analysis, Schwartz and Sachs reach three conclusions:

1. While each type of measure captures something important, none of them captures all the important components of decision quality.

2. Decision quality is complicated and cannot be adequately captured in a precise quantity. Instead of trying to measure quality using a precise quantitative value, it is more appropriate to describe it using general terms such as good, satisfactory, or poor.

3. Efforts to improve clinical decisions should “… focus on identifying and addressing cases of poor or problematic decision-making, rather than on improving the mean score on a measure or combination of measures.”

Musings

This is a terrific contribution and everyone interested in shared decision making and the overall quality of clinical care should read the full article. However, it has a couple of limitations that are worth keeping in mind:

1. The analysis was limited to how decision quality has been conceived and characterized in the medical decision making literature. Healthcare is not the only field where good decision making is important and, as pointed out in prior Musings, there is a wealth of information about how to improve decision making in applied settings available that could easily be imported from other fields to address current problems with medical decision making. For example, the Decision Quality Chain is a well-developed crystallization of the elements that define a good decision that can be readily understood and used to guide and evaluate the quality of a decision process. (See the December 2, 2022 and January 13, 2023 Musings for details.) For this reason, I think the authors’ second conclusion, while correct given how decision quality has been conceived in medical settings to date, should not be generalized beyond the limits of the literature reviewed.

2. The authors don’t comment on a major shortcoming of all current measures: they involve an extra step to obtain the quality measurement, making them primarily research tools rather than methods to improve the ongoing quality of decisions in practice settings. To be truly effective, measures to improve clinical decision making , particularly those involving important bioethical goals such as involving patients in decisions about their healthcare, need to be effectively integrated into routine practice. Using a measure of decision quality, such as the Decision Quality Chain, that can be used to both guide on an ongoing decision making process and generate information useful in evaluating the aggregate quality of decisions makes sense to me.

Despite these limitations, I agree with the primary message of the article: it is time to to improve the quality of clinical decision making by instituting programs to diagnose and cure problematic clinical decision making practices at all levels of the healthcare system.

References

1. Schwartz PH, Sachs GA. Rethinking Decision Quality: Measures, Meaning, and Bioethics. Hastings Center Report. 2022;52(6):13–22.

2. A. M. O’Connor, “Validation of a Decisional Conflict Scale,” Medical Decision Making 15, no. 1 (1995): 25-30.

3. M. M. Garvelink et al., “Decisional Conflict Scale Use over 20 Years: The Anniversary Review,” Medical Decision Making 39, no. 4 (2019): 301-14.

4. J. C. Brehaut et al., “Validation of a Decision Regret Scale,” Medical Decision Making 23, no. 4 (2003): 281-92.

5. Braddock CH, Fihn DS, Levinson W, Jonsen AR, Pearlman RA. How doctors and patients discuss routine clinical decisions—informed decision making in the outpatient setting. J Gen Intern Med 1997;12:339–45.

6. Braddock CH, Edwards KA, Hasenberg NM, Laidley TL, Levinson W. Informed decision making in outpatient practice—time to get back to basics. J Am Med Assoc 1999;282:2313–20.

7. Elwyn G, Edwards A, Wensing M, Hood K, Atwell C, Grol R. Shared decision making: developing the OPTION scale for measuring patient involvement. BMJ Quality & Safety. 2003 Apr 1;12(2):93-9.

8. Michie S, Dormandy E, Marteau TM. The multi-dimensional measure of informed choice: a validation study. Patient education and counseling. 2002 Sep 1;48(1):87-91.

Using situational awareness to identify the first principles of high quality clinical decisions

The most common approach to tackling the question of how to foster shared decision making in routine clinical care is to identify barriers impeding its adoption. Commonly identified barriers include: health systems do not view shared decision making as the standard of care, insufficient time during patient visits, poor fit into the clinical workflow, paucity of appropriate information, and that clinicians find shared decision making difficult to accomplish. [1] Although these barriers have been known for some time, little progress has been made in overcoming them. [2]

In recent Musings, I suggested taking a First Principles Thinking approach to creating a new, more effective clinical decision making method that recognizes the importance of patient involvement in decisions about their health, i.e., shared decision making. First Principles Thinking is designed to encourage new approaches to resolving stubborn problems. The underlying premise is that starting from scratch by concentrating on what needs to be accomplished, what the basic elements involved are and how they can be effectively organized is more likely to succeed that continually trying to tweak the status quo.

With regard to fostering the adoption of shared decision making in clinical practice, I think the goal is to improve the overall quality of the clinical decision making process. This goal includes routine use of shared decision making when appropriate and also seeks to improve the quality of all aspects of the clinical decision making process. The First Principle irreducible elements can be divided into two big categories: those related to the decision making process and those related to the clinical environment.

From this perspective, the concepts of Situational Awareness, as described by Endsley, provide a useful guide to the both of these categories. [3] I presented an adapted Situational Awareness model for clinical shared decision making in the November 4, 2022 Musing. For this discussion, I’ve left the shared decision making/situational awareness components the same but added a new set of system factors and several new links showing how the clinical practice environment affects elements of the decision making process:

This model identifies several First Principle elements and the interactions among them within a clinical decision making system. Both provider and patient need to achieve an appropriate level of situational awareness about the decision under consideration. This process involves recognizing that a decision needs to be made, achieving a good understanding of the decision parameters (the goal, the options, and the factors important in choosing a preferred option), and considering the likely outcomes of each option. For both parties, how well they can achieve an appropriate level of situational awareness depends on what they are trying to accomplish and their experience, how well they can access needed information and effectively process it, and decision-making skill. They also need to be able to work together to effectively create a shared situational awareness that will result in making a high quality decision. (I described decision quality using the decision quality chain in the November 18, 2022 Musing.)

The ability of the clinical decision makers to achieve individual and shared situational awareness is also a function of the environment they are working in, i.e., the local clinical microsystem as well as the more widespread healthcare macrosystem. Systems with an underlying goal of improving the quality of clinical care would ensure that patients and providers have the information, time, and training necessary to achieve adequate levels of individual and shared situational awareness and use it to make a high quality decision.

Musings

My proposed model is definitely a work in progress. Nevertheless, I think it presents a good overview of how a better clinical decision making system could be created. I pose it to stimulate further thought and discussion. Note that a system like the one proposed would alleviate the most commonly cited barriers to shared decision making, but depends on a general system-wide approach very different from one based primarily on tweaking the way clinicians discuss care decisions with their patients.

Is a system like this feasible? I believe it is. There is a large trove of resources available that could be used to implement it that have not been introduced into clinical care. It’s time we started doing so. I discussed some of these approaches in the December 2, 2022 Musing.

I will present more details about the use of the decision quality chain as a metric for assessing decision quality in a future Musing.

References

1. Elwyn G, Durand MA, Song J, Aarts J, Barr PJ, Berger Z, et al. A three-talk model for shared decision making: multistage consultation process. BMJ. 2017;359:j4891.

2. Finset A, Street RL. Shared decision making in medicine – improving but still a work in progress. Patient Education and Counseling. 2022 May 1;105(5):1055–6.

3. Endsley MR, Jones DG. Designing for Situation Awareness: An Approach to User-Centered Design, Second Edition. 2nd ed. 2012;396.

The lessons of First Principle Thinking

Implications for improving clinical decisions

Lesson One: —

Beware of what you inherit. When someone presents something to you, the tendency is to make it slightly better than before rather than to ensure it’s the best solution in the first place.

Don’t do that.

When it comes to decision making, this means never letting anyone else define the problem for you. [1]

---

In recent decades, all medical decisions, even common, everyday ones, have become increasingly complex. Advancing technology has led to the creation of many new drugs, therapeutic modalities, and diagnostic tests. Multiple management options, none of which is clearly the most appropriate option for every patient, are now available for many clinical problems. Consequently there is an increased need for ensuring that options chosen are consistent with the preference and values of individual patients, the ones who ultimately will need to live with the outcomes of the choices made.

Despite a clear moral imperative to actively engage patients in making decisions about important choices affecting their health, current research studies have repeatedly shown that shared decision making is not widely practiced. As I’ve reviewed in previous Musings, currently proposed solutions to this problem have primarily consisted of modifications to the standard clinical dialog between patient and provider. Multiple modifications have been suggested and, so far, have had little impact. (2) It will be interesting to see if the additional suggestions included in the recent NHS Advice (discussed in the August 12, 2022) will make a difference.

In the October 10, 2022 Musing, I suggested using First Principles Thinking to facilitate uptake of shared decision making in clinical practice. In a recent Facebook post, posted on November 23, 2022, the Farnum Street Blog presented three lessons derived from First Principles Thinking which they defined as: “You break things down into the core parts and reassemble them in a more effective way.” [1] In the post, they liken first principles to LEGO blocks, items that cannot be reduced further that will be used to create a desired outcome.

I’ve quoted the first lesson in its entirety above. The second and third lessons advise breaking up a troublesome problem into it’s smallest components based on what is 100% true and then starting over. The First Principles hypothesis is that by taking a fresh approach, you are more likely to create a methodology that will achieve the desired outcome than repeatedly trying to tweak an existing approach that is not working.

Musing

It seems to me that all of the currently proposed solutions have failed to heed Farnum Street’s Lesson One of First Principles Thinking: they are trying to modify the traditional clinical interaction instead of first considering whether it is capable of adding shared decision making into an already complex decision making scenario. The lack of progress using this approach suggests something else needs to be done. An approach based on First Principles thinking seems well worth trying.

Reference

1. https://www.facebook.com/FarnamStreet/posts/pfbid017HFW5YtEGCzjSrv8JvSjgTeLoRE1Vk8Nbk7qaYJcNWeuXNBe58buXxHRpuzydLvl

2. Stiggelbout AM, Pieterse AH, De Haes JC. Shared decision making: concepts, evidence, and practice. Patient Educ. Couns. 2015 Oct 1;98(10):1172-9.

Helping clinicians and patients become better decision makers: A proposal

Do what you can, with what you have, where you are. ~ Theodore Roosevelt

---

In the November 28, 2022 post, I suggested that finding a way to help clinicians and patients become better decision makers could foster implementation of shared decision making in clinical practice. In addition I believe it would also increase the overall quality of healthcare.

A number of strategies have been proposed to help people make better choices when faced with a complex decision. All advocate simplifying the decision-making process by breaking it down into a series of smaller, simpler steps. [1] The decision quality chain, developed by the Strategic Decisions Group, is one of the best developed of these strategies. [2,3] A slightly modified version, changed to show a common clinical decision sequence, is shown in the figure below:

At this point it is worth noting that clinical encounters typically involve many decisions. To be successful, a clinical decision support method also has to include a way to identify and focus on decisions that require particular attention because they have potentially significant impacts on patient well-being, risks, or costs.

Developing a clinical decision-making support system

Now let’s examine how we could convert the decision quality chain into a practical clinical decision support method.

I’ve previously reviewed the article by Helen Wu and colleagues that identified the three decision making principles, six design features, and two implementation strategies listed below that are shared by decision support systems successfully implemented in non-medical settings. [4] (See the October 4, 2022 post.)

Decision making principles

1) The system should draw on the strengths of both rational-analytic and naturalistic-intuitive decision making styles while minimizing their respective weaknesses.

2) The system should use a robust, flexible approach that addresses multiple criteria and possibilities.

3) The system should provide an “appropriate level of trust” in its methods and recommendations.

Design features

1. The system should provide broad, top level perspectives to help users understand the full extent and scope of the decision being considered.

2. The system should be readily adaptable to meet the needs of the intended users in different settings.

3. The underlying methods used to analyze data and generate recommendations must be transparent and deemed trustworthy by intended users.

4. Data and other system-related information should be effectively organized and clearly presented to avoid information overload.

5. Systems should allow users to examine multiple scenarios and options simultaneously to facilitate the optimal use of both rational-analytic and naturalistic-intuitive decision making processes.

6. Systems should promote rapid collaboration among decision makers and facilitate access to an appropriate range of expertise to support good decision making.

Implementation strategies

1) Systems should be regularly evaluated and improved.

2) Effective user training is imperative.

Proposed principles for creating a successful clinical decision support system

Combining the findings by Wu and colleagues with the decision making process described in the decision quality chain results in a set of principles for creating a practical clinical decision support system designed to promote good clinical decisions:

A. Clinical Decision Quality Assessment:

• Helpful Frame: A mechanism to clearly define and map the nature and scope of a clinical decision regarding diagnosis or treatment of a patient’s problem.
• Clear Values: Inclusion of all important decision-related objectives.
• Alternatives: Identification of a full set of options that could be used to manage the patient’s problem.
• Useful information: Ready access to an accurate, trustworthy summary of current evidence in a format designed to support clinical decision making.
• Sound reasoning: A process designed to compare alternatives relative to decision objectives, minimize the adverse effects of cognitive biases, cope with decision-related uncertainties, facilitate needed tradeoffs between decision objectives, and allow patients and clinicians (and other appropriate parties) to discuss each other’s decision-related preferences and priorities.
• Commitment to follow through: Willingness of both patient and provider to implement the chosen alternative.

B: Integrating a clinical decision support system into routine clinical practice:

• Guidelines for determining significant decisions requiring careful deliberation.
• A simple but effective procedure to identify important considerations from the patient’s point of view.
• A simple but effective procedure patients and providers can use to assess and compare their decision priorities.
• A simple but effective procedure patients and providers can use to evaluate the decision options and compare the evaluations.
• A procedure that is accepted by clinicians as feasible, worthwhile, and trustworthy.
• A procedure that is accepted by patients as worthwhile and trustworthy.
• A mechanism for recording the decision making process, documenting it in the medical record, and making it available to revisit in the future if desired.
• Inclusion of assessment of decision quality in routine quality assurance activities.
• Initial and ongoing decision making education for providers and patients.

Musings

1. Is developing this system feasible? Yes. the necessary methods and resources are currently available.
2. If implemented, would this proposal work? I don’t know. But there is a good chance it might since it is based on methods that have proven to be effective in other areas.
3. Is this business as usual? No. Adopting this proposal means accepting that the decision sciences have a place at the table of medical sciences along with anatomy, pathology, physiology, pharmacology and all the rest. However, since one of the basic acts of medical practice is making decisions, I think it is time to make better use of the knowledge that is available that can improve the quality of healthcare.
4. Clearly this is just a general outline. I will try to fill in some of the gaps in future posts. Comments, suggestions, and questions are welcome.

References

1. Decision-making. In: Wikipedia [Internet]. 2022 [cited 2022 Nov 16]. Available from: https://en.wikipedia.org/w/index.php?title=Decision-making&oldid=1122078473

2. Decision Quality Defined | Strategic Decisions Group [Internet]. SDG. [cited 2022 Dec 1]. Available from: https://sdg.com/thought-leadership/decision-quality-defined/

3. Decision Quality: Value Creation from Better Business Decisions | Wiley [Internet]. Wiley.com. [cited 2022 Dec 1]. Available from: https://www.wiley.com/en-us/Decision+Quality%3A+Value+Creation+from+Better+Business+Decisions-p-9781119144670

4. Wu HW, Davis PK, Bell DS. Advancing clinical decision support using lessons from outside of healthcare: an interdisciplinary systematic review. BMC medical informatics and decision making. 2012;12(1):1–10.

Reverse salients and shared decision making

Time to train the monkey?

James Dolan

The reverse salient … refers to the sub-system that has strayed behind the advancing performance frontier of the system due to its lack of sufficient performance. In turn, the reverse salient hampers the progress or prevents the fulfillment of potential development of the collective system. … Because reverse salients limit system development, the further development of the system lies in the correction of the reverse salient, where correction is attained through incremental or radical innovations. [1]

---

The concept of reverse salients was discussed by Ethan Mollick of the Wharton School in a recent Twitter tweet. He defined a reverse salient as “the technology or process that is holding back development of the whole system … Solving the salient unlocks change.” As an example, Mollick notes that the development of adequate car batteries made electric cars viable.

The term reverse salient, originally a military term, to describe technological changes was introduced by Thomas Hughes in a 1983 book about the early years of electrification where the development of alternating current solved the problem with “low voltage transmission distance” of Thomas Edison’s direct current system. [1]

In her August 16, 2022 newsletter, Annie Duke proposed that considering reverse salients is an example of a mental model called “monkey’s and pedestals”. [2] The basic idea is if you want to train a monkey to recite Shakespeare (or juggle flaming torches) on a pedestal in a town square, you should start the more difficult task: training the monkey. Even though building the pedestal would be much easier and a quicker way to show progress is being made, it won’t accomplish the goal until the monkey is trained. In this case training the monkey is the reverse salient that is holding the whole project back.

Musings

I’ve written a lot in this space about the difficulties introducing routine shared clinical decision making into practice and identified a series of models describing conversation-based solutions to the problem. None have made much of a difference.

Perhaps we have focused too much on the pedestal of just changing the script of a standard clinical consultation and not enough on the monkey of how to enable clinicians and patients to become better decision makers.

References

1. Wikipedia, Reverse Salient: https://en.wikipedia.org/wiki/Reverse_salient

2. Duke, Annie. NEWSLETTER: MONKEYS AND PEDESTALS: FIND THE BOTTLENECK AND SOLVE FOR THAT FIRST. https://www.annieduke.com/newsletter-monkeys-and-pedestals-find-the-bottleneck-and-solve-for-that-first/