Key concepts relevant to quality of complex and shared decision-making in health care: A literature review

Abstract

High-quality provider–patient decision-making is key to quality care for complex conditions. We performed an analysis of key elements relevant to quality and complex, shared medical decision-making. Based on a search of electronic databases, including Medline and the Cochrane Library, as well as relevant articles’ reference lists, reviews of tools, and annotated bibliographies, we developed a list of key concepts and applied them to a decision-making example. Key concepts identified included provider competence, trustworthiness, and cultural competence; communication with patients and families; information quality; patient/surrogate competence; and roles and involvement. We applied this concept list to a case example, shared decision-making for live donor kidney transplantation, and identified the likely most important concepts as provider and cultural competence, information quality, and communication with patients and families. This concept list may be useful for conceptualizing the quality of complex shared decision-making and in guiding research in this area.

Introduction

High-quality medical decision-making is critical to much of medical care, including differential diagnosis, test ordering and interpretation, prescribing, and skillful delivery of treatments and procedures. Shared or patient-centered decision-making, the process by which a health care provider communicates to the patient personalized information about options, outcomes, probabilities, and uncertainties of available options and a patient communicates values and the relative importance of benefits and harms (Foundation for Informed Medical Decision Making, 2006), is important for providing care consistent with patient preferences, and may improve satisfaction and adherence (Sepucha et al., 2004).

Most efforts to measure quality address relatively straightforward provider behaviors, such as compliance with practice guidelines (Blumenthal, 2004). These unidimensional quality indicators allow for complex decision-making only through exclusion of patients with certain characteristics, or risk adjustment for patients with complex illness. They cannot adequately address quality issues such as variation in use of preference-sensitive procedures; complicated situations, such as complex illnesses that require weighing risks and benefits of different treatments, or patients with multiple illnesses (Epstein & Hundert, 2002); or outcomes potentially related to poor-quality shared decision-making, such as nonadherence.

An important element in quality for these situations is determining how best to conceptualize quality of complex shared decision-making. Research in this area would benefit from a broader perspective of key relevant concepts in decision-making quality, shared decision-making, and related areas. We therefore conducted a qualitative literature review to develop a list of potential key concepts relevant to the quality of complex, shared medical decision-making.

Methods

Search strategy

We searched Medline, Cochrane Library, Evidence-Based Medicine Reviews, PsychInfo, and Sociological Abstracts, as well as key articles’ references, key authors’ publications, reviews of relevant tools, and Foundation for Informed Medical Decision Making (2006) annotated bibliographies. We adapted search terms from related reviews, including “quality” or “process” and “decision” or “decision-making”, “shared decision-making”, “participatory decision-making”, “partnership”, “patient-centered communication”, and “informed (patient) decision-making”. We did not limit the search by country of origin. We focused on systematic, structured, and conceptual reviews of relevant concepts and interventions, and abstracted key concepts relevant to shared decision-making and information on relevance. We incorporated the concepts into a coherent list. We applied the concepts to the context of structural, process, and external factors related to shared medical decision-making and outcomes. Finally, we developed a list of factors relevant to a case example, live donor kidney transplantation, and used these to evaluate which elements of the concept list might be most relevant to this example (Supplementary data).

Results

The concept list includes seven key concepts: provider competence, trustworthiness, and cultural competence; patient/surrogate competence; quality of information and communication; and decision-making roles and involvement (Table 1).

Table 1.

Concepts of complex shared decision-making.

Concepts and definitions	Key components
Provider competence Ability to conduct effective, appropriate shared decision-making	Clinical knowledge, reasoning, judgment Avoiding cognitive biases Knowledge of patient, preferences, values Respect, empathy Decision-making, shared decision-making skills Conflict resolution skills
Provider trustworthiness Being reliable, deserving of patients’ trust and confidence	Fidelity (caring and advocating for patient interests/welfare; avoiding conflicts of interest) Competence (good practice, interpersonal skills, making correct decisions, avoiding mistakes) Honesty Confidentiality (proper use of sensitive information)
Cultural competence Understanding social and cultural influences on patients’ health beliefs, behaviors, decision-making	Cross-cultural knowledge, sensitivity Use of culturally and linguistically appropriate health materials Use of language and interpreter services Humility (freedom from arrogance in mannerisms, interactions) Respect for patient preferences, values
Communication with patients and families Patient-centered exchange of information and meaning to create shared understanding and decisions	Understanding experience, expectations Eliciting values, preferences Presenting options, evidence, risks, benefits, recommendations Questioning preferred involvement in decision-making Checking for understanding, agreement Psychosocial communication, empathy Evaluating self-efficacy (ability to follow through with a plan)
Information quality Data and evidence affecting the decision	Appropriate in content, quantity, form, presentation, timing Comprehensive, evidence-based Understandable, useable Consistent with patient preferences, health literacy
Patient/surrogate competence Patient/surrogate decisional capacity	Ability to understand information, use in reasoning, apply to current decision Ability to express clear, consistent choice Freedom from conflicts of interest (surrogates) Ability to know and apply patient’s preferences (surrogates)
Roles and involvement Process and amount of patient/family involvement in shared decision-making	Degree of partnership appropriate for decision, consistent with patient preferences Patient empowerment, readiness, self-efficacy

Concepts

Provider competence

In a systematic review (Epstein & Hundert, 2002) on professional competence, focusing on physicians and trainees, concepts applicable to complex decision-making included decision-making skills; knowledge; clinical reasoning and judgment; presence of cognitive biases; values, respect, compassion, and professionalism; and skills in professional–patient relationships, teamwork, and handling conflict.

Concepts from studies on the ability to conduct effective and appropriate shared decision-making include skills in involving patients; exploring patient knowledge and preferences; and educating about options (Elwyn, Edwards, Kinnersley, & Grol, 2000). Physicians’ perceptions and values can affect how much they involve patients, and physicians must have the correct evidence and interpret it correctly. Lack of knowledge or misunderstanding of patient status, symptoms, or perspectives can negatively impact shared decision-making.

Key provider decision-making biases include the availability heuristic, where more recent or memorable events (e.g., a serious adverse event) weigh disproportionately in decision-making; and risk aversion, where providers weigh potential losses (e.g., fear of iatrogenesis) more strongly than potential gains. We found no reviews addressing issues with providers other than physicians.

Provider trustworthiness

The shared decision-making process is heavily influenced by patients’ perceptions of providers’ trustworthiness. Patients who perceive lack of trust are more likely to report shared decision-making problems (Ngo-Metzger et al., 2006). Patient trust (expectancy that a promise or statement can be relied upon) (Pearson & Raeke, 2000) may influence patients’ willingness to engage in decision-making. Interpersonal trust involves believing that individuals’ motives are genuine, and is influenced by direct experiences. Social trust involves the health system generally, and is influenced by both direct and second-hand experiences (Boulware, Cooper, Ratner, LaVeist, & Powe, 2003).

Provider trustworthiness is influenced by the length and nature of provider–patient relationships, continuity of care, patient perceptions and expectations, satisfaction with care, provider communication and collaboration, and provider–patient race and language concordance (Boulware et al., 2003). Provider trustworthiness and patient trust are also associated with quality of care and willingness to follow medical advice (Ngo-Metzger et al., 2006).

Cultural competence

Cultural competence includes behaviors and attitudes enabling providers’ effectiveness in cross-cultural situations. Cultural competence acknowledges the importance of social and cultural influences on patients’ health beliefs, behaviors, and decision-making, which are important for health equity for diverse patient populations. Cultural competence may reduce disparities in shared decision-making, particularly for underserved populations, and high-quality decision-making requires responsiveness to relevant cultural factors (Beach, Saha, & Cooper, 2006).

Culturally-competent decision-making involves encouraging patient participation while respecting cultural factors, such as language, beliefs, and values (Ngo-Metzger et al., 2006). Key components include cross-cultural knowledge and use of language and cultural interpreters and culturally and linguistically appropriate information. Other important factors include cultural sensitivity and respect for preferences, Providers should also be conscious that patients may have had negative experiences leading to distrust of health care providers, including past discrimination (Beach et al., 2006).

Communication with patients and families

Communication in shared decision-making can be defined as patient-centered exchange of information and meaning to create shared understanding and decisions. The goal is to help practitioners provide care concordant with patient values, needs and preferences, and allow patients to provide input and participate actively (Epstein et al., 2005). Communication of evidence involves building partnership and eliciting key patient/family information, including experience and expectations, values and preferences, and psychosocial context (Dunn, Nowrangi, Palmer, Jeste, & Saks, 2006). Presenting options and evidence includes balanced discussion of uncertainties and checking for understanding; presenting recommendations informed by clinical judgment and patient preferences; and evaluating agreement on decisions and next steps (Epstein, Alper, & Quill, 2004). Depending on the decision, the final step is often discussing patients’ self-efficacy, or ability to follow through with a plan (Makoul & Clayman, 2006). In ideal shared decision-making, this process should involve reciprocal information exchange, not just transfer from the provider. Other aspects of communication, such as assessing health literacy and responding to emotions, are also important.

Communication must be individualized to the patient and situation; using multiple methods may help provide balance (Moxey, O’Connell, McGettigan, & Henry, 2003). Information framing (e.g., descriptive vs. numerical presentation) can significantly impact understanding and decision-making. A systematic review found that evidence-based methods, such as visual depictions of risk or interactive exercises for eliciting preferences, enhance patient understanding and improve decision quality (Trevena, Davey, Barratt, Butow, & Caldwell, 2006). A systematic review of communicating risk information found that providing more information and in a more understandable format was associated with improved knowledge and decreased willingness to accept treatment (Edwards, Elwyn, Covey, Matthews, & Pill, 2001).

Information

Quality of information is important both for information about patients and caregivers that providers use in decision-making, and for information that providers give patients. Important patient information includes understanding of disease, prognosis, or treatment; quality of life, distress, or needs; caregiver issues; and expectations and perceptions of care. For providers communicating information to patients, the content (relevance, risks, benefits and uncertainty), form (verbal vs. written), quantity and level of detail, and timing should be appropriate to the decision and situation and to patient preferences (Epstein et al., 2004). Information should be evidence-based and balanced between clear presentation and accuracy and completeness, and patient and caregiver understanding and acceptance should be assessed using components such as patient self-report and reaction to and use of the information (Willis & Holmes-Rovner, 2003).

Patient/surrogate competence

Patient decisional capacity or competence includes the ability to understand information, appreciate its significance, and apply it by using it in decision-making and expressing a clear and consistent choice (Dunn et al., 2006). Cognitive abilities or psychosocial and psychiatric issues can affect capacity. Lack of capacity is very common in hospitals and often unrecognized. A systematic review on cognitively impaired elderly (Kim, Karlawish, & Caine, 2002) found that judgment of decisional competence depends on the decision and physician style; interventions can sometimes be helpful in improving decisional abilities.

Another systematic review found additional issues for surrogate decision-makers (health care proxies who make decisions for patients without capacity). Surrogates may not know or have difficulty ascertaining patient preferences for a particular situation, or have difficulty applying the patient’s preference rather than their own (Shah, Farrow, & Robinson, 2009). Family conflict, misunder-standings, and conflicts of interest can also affect decision-making. Important decision-making biases include omission bias, where omission (e.g., refusing vaccination) is preferred to commission (due to exaggerated importance of vaccination side effects). Other factors include emotions (e.g., negative mood causes overestimated likelihood of negative outcome) and role expectations (e.g., a daughter continues life support because she believes that is what a good daughter should do) (Arkes, 2004).

Roles and involvement

The process of and degree to which patients/families are involved in shared decision-making are critical, as well as involvement of other relevant providers. Involvement should be appropriate for the specific decision and consistent with patient preferences; the process of involvement may be as important as who eventually makes the decision (Edwards & Elwyn, 2006).

The concept of shared decision-making between patient and physician has been termed partnership, with three main models (Charles, Whelan, & Gafni, 1999). The paternalistic model assumes the doctor will make the best decision; there is no partnership. The shared model is interactive in all stages of the process. In the informed model, the physician communicates information to the patient, and the patient makes the decision. Research has found that decision characteristics affect patients’ preferences (e.g., preferences for physician control were significantly higher for potentially fatal scenarios than for quality of life issues (Deber, Kraetschmer, & Irvine, 1996)). Often, particularly for decisions requiring significant medical knowledge (e.g., intensive care), physician recommendations are appropriate. Patients usually want involvement in some form (Moxey et al., 2003); even when patients do not have the skills or are not ready, coaching or encouragement is often appropriate to increase involvement and improve the process.

Context of complex shared decision-making

In Fig. 1, we describe how these concepts may fit into the larger context of health care by illustrating potential relationships of key contextual structure and process variables in the decision-making encounter, external factors, and key related outcomes. The concepts of shared complex decision-making can be affected by structural elements, including access to decision-making resources (e.g., a skilled provider), setting (e.g., outpatient vs inpatient) and use and quality of relevant services (e.g., a social worker to help with family conflict). Relevant process variables include characteristics of the provider–patient relationship (e.g., duration), continuity (e.g., providers’ coordination of information), provider decision-making style, experience, and skills, and patient/family past experience with decisions (e.g., about end-of-life care for another family member).

Fig. 1.

Context factors related to quality of complex shared decision-making.

External factors may include illness characteristics such as severity (e.g., deciding whether to start insulin for diabetes vs. choosing a method of dialysis); access to care and insurance issues; patient beliefs (e.g., religious issues with blood transfusions) or family structure (e.g., an only child vs. many family members involved in decision-making); and provider specialty or knowledge (e.g., a back pain treatment decision with an occupational medicine physician vs. an orthopedic surgeon). Finally, in considering a particular decision, prioritizing key outcomes, such as appropriateness or timeliness of decision-making, adherence to treatment, health status, or satisfaction with the decision-making process or outcome, can help to determine which concepts are most applicable.

Application of the concept list

The potential relevance of these concepts for decision-making for the case of live donor kidney transplantation (LDKT) for patients with end-stage renal disease (ESRD) is in Table 2. LDKT, where patients receive a kidney from a living relative, friend, or other altruistic donor, is the optimal treatment for ESRD and provides higher quality of life and longer life expectancy than dialysis.

Table 2.

Case: live donor kidney transplantation (LDKT).

	Concepts
Decision-related factors	Provider competence	Provider trustworthiness	Cultural competence	Information quality	Patient/Surrogate competence	Communication with patients/families	Roles and involvement
Structural factors
1. Availability of cultural, language and literacy-appropriate patient information			×	×		×
Provider factors
2. Provider LDKT and transplant evaluation process knowledge	×			×
3. Provider perceptions of patient LDKT suitability	×		×	×	×	×
4. Provider-patient timely discussion	×					×	×
Patient/family
5. Trust in providers’ LDKT recommendation	×	×	×	×		×
6. Cultural differences in decision-making, support for LDKT			×			×	×
7. Knowledge about LDKT benefits	×	×	×	×	×	×
8. Concerns about donor risks	×	×	×	×		×
9. Concerns about recipient risks, transplant failure	×	×	×	×	×	×
10. Religious views about organ donation				×		×

We identified various structural, provider, and patient-related factors associated with the decision in a US-based context (Table 2). Provider, patient, and family education and knowledge about LDKT are needed to initiate provider–patient decision-making; the availability of culturally sensitive, and language- and health literacy-appropriate information about LDKT are vital. These factors may also impact providers’ evaluation of patient suitability for LDKT, as well as timely referrals for transplant evaluation and decision-making (Waterman et al., 2006). Patient and family factors also impact LDKT decision-making, including perceptions of provider trustworthiness, cultural and religious influences, and concerns about potential clinical risks for the donor and recipient (Boulware et al., 2002).

In evaluating the relevance of the concept list to each of these factors, the concepts of cultural competence, information quality, provider competence, and communication with patients and families were the most relevant concepts for this case example.

Conclusions

This analysis of key concepts relevant to quality of complex, shared medical decision-making identified provider competence, trustworthiness, and cultural competence; patient/surrogate competence; quality of information and communication; and roles and involvement. Many of these concepts are interrelated. For example, patient preferences for participation can be affected by aspects such as trust or information (Kraetschmer, Sharpe, Urowitz, & Deber, 2004). The concept shared decision-making may include aspects such as physician knowledge and information exchange (Epstein et al., 2004). Concepts also overlap: for example, information quality is included as a separate concept, but is also integral to high-quality communication.

Many other concepts could be considered for this list. Other concepts were not included because they have not been well-developed or researched in the context of decision-making, such as other aspects of provider–patient relationships (e.g., respect) (Epstein et al., 2004); or for shared decision-making, such as decision-making style or process steps (e.g., agenda-setting, negotiation). The impact of components on outcomes may over-lap, including patient understanding, concordance between patient preferences and decisions, treatment adherence, and satisfaction. Outcomes for measuring impact include health care utilization and variation; patient/surrogate understanding, expectations, emotions, and involvement; decisional conflict (degree of certainty) and what decisions were made; satisfaction; and health status (O’Connor et al., 2009). Intermediate outcome measures include whether a patient understands adequately, can use information in decision-making, and can make a good choice (Hibbard, 2004). Ideally, implementing decision quality measures would require patient-reported data, targeted to specific populations or procedures.

Evaluating decisions by the choice selected or outcomes has drawbacks, since decisions are based on likelihoods and outcomes depend on many factors. Defining control groups is challenging, and outcomes of many complex decisions are too multifaceted or long-term to measure adequately. These decisions are partly based on values and other personal attributes, which affect choices’ appropriateness. These concepts also vary depending on who is involved and who is asked: the patient, the physician, other physicians, or society.

Our review has several limitations, including our concentration on systematic and structured reviews and on individual concepts, rather than interrelations between concepts. The list also has several limitations. These elements may only apply when there is a clear decision, which frequently does not occur (e.g., avoiding decisions about end-of-life care). Evidence is limited because shared decision-making often is not considered, is not applicable due to time constraints, situational issues, or patient characteristics (Gravel, Légaré, & Graham, 2006), and is difficult to research because much is not observable. Finally, although the application of the list to the case example helped to illuminate which concepts might be most relevant for further research, this application would require empirical testing.

In conclusion, this list may be useful for evaluations and interventions for complex shared decision-making. Future studies should explore the concepts’ use in research, relative importance for different decisions and settings, and interactions between them, as well as the role of other potentially important concepts.