Shared Decision Making in the Emergency Department Among Patients With Limited Health Literacy: Beyond Slower and Louder

Introduction

Health literacy is “the degree to which individuals can obtain, process and understand the basic health information needed to make appropriate health decisions and access health services needed to prevent or treat illness.”⁷ The prevalence of limited health literacy (LHL) among adult ED patients is as high as 88%.^8,9 Numeracy, or the “ability to understand and use numbers in daily life,” is a continuum of skills that ranges from basic recognition and computation of numbers to understanding and use of high level skills such as probability, risk, and randomization.^10,11 Levels of numeracy in the general population and the ED are relatively low and are only loosely related to education and health literacy.^8,9,12 Low numeracy skills are more common than LHL, with prevalence ranging from 44%⁸ to 84.8%⁹

LHL and low numeracy compound the baseline challenges to rapidly and safely caring for patients in the ED. LHL and low numeracy are often unrecognized by healthcare providers, making these communication barriers even more difficult to address. Low numeracy and health literacy have been associated with a range of adverse health effects, including worse self-management skills¹³; lower adherence to treatment and medication¹⁴; greater risk of hospitalization, ED, visits, and lack of preventative care; worse health status; and lower quality of life¹⁵. Among ED patients, LHL is associated with worse health status, higher healthcare utilization, and increased risk of death.^16–20 Barriers to clear communication among patients with LHL and low numeracy skills are only heightened in the setting of acute illness and cognitive impairment in this large population of vulnerable, high risk patients.²¹

SDM among those with limited health literacy

Though work has begun to focus on SDM among patients with LHL or low numeracy,²² to our knowledge this has not been studied among patients in the ED setting. To this end, the goal of this manuscript was to identify recommended best practices, gaps in evidence, and research needs related to SDM in the ED among patients with LHL or low numeracy.

We adopted and modified for use in the ED an approach outlined by McCaffery et al,²³ which applies the US Preventative Health Task Force (USPHTF) SDM framework to patients with LHL. Formal SDM in the USPHTF framework includes “5 stages that occur when an individual: 1) understands the nature of the disease or condition, 2) is aware of the clinical services available and likely consequences, 3) has considered his or her preferences as appropriate, 4) has participated in decision making at a personally desirable level; and 5) makes a decision consistent with her or her preferences and values, or chooses to defer a decision.”

Stage 1 of SDM in the ED: Understanding the condition and its management

The first stage of SDM is the most critical because it is the foundation upon which the other stages are built. Though a universal precautions approach for clear communication is generally recommended to accommodate different health literacy levels, whether a tailored approach is needed for SDM with patients with LHL has not been established. Information about a medical condition and its management is typically presented to patients via two routes: written health materials and verbal exchange. The literature to date largely focuses on these as two separate actions.

Patient Decision Aids (PtDAs) bridge the communication gap between written and verbal information exchange during SDM and are discussed elsewhere in this edition of Academic Emergency Medicine. PtDAs may be particularly helpful for patients with LHL, as they increase patient engagement, but we do not know which elements of PtDAs and approach are most effective for SDM with LHL populations. For example, some SDM interventions have been effective even though less than 50% of PtDA features were delivered.²⁴ IPDAS standards generally emphasize comprehensiveness of the PtDA^25,26 to guide verbal communication, help maintain consistency in terminology, and avoid problems with patient understanding that can result from extemporaneous speech. The so-called parsimonious approach to use of PtDAs focuses on their fitting “into the workflow of clinical encounters,” and serving more loosely as “scaffolding for conversations,” with clinicians filling in the gaps in knowledge or understanding.^25,27 Striking the optimal balance between a parsimonous and more comprehensive, structured approach in this patient population in the ED setting, while avoiding a two-tiered system of communication remains a challenge. Research focused on the design, use, and evaluation of patient decision aids should be a priority moving forward.

Below we review recommendations for delivering health information to patients with LHL, including in PtDA’s and highlight areas in need of further study.

Reading materials and visual aids

Plain language is writing that is clear, concise and allows readers to understand the content at first read-through.²⁸ Using plain language, minimizing content to align with working memory constraints, maintaining adequate white space, using sans-serif fonts, and using graphics or pictograms are some of the commonly used strategies to improve the delivery of health information in print materials to foster patient comprehension (Table 1).²⁹ Several of the strategies noted here relate to presenting risk and numerical concepts, which are discussed further in Stage 2 and 3 below. Although these strategies have been shown to improve understanding among patients with LHL, they have not been evaluated in the context of SDM or in the ED setting.²⁹ More research is needed to determine how to design PtDAs for SDM specifically for patients with LHL in the ED.

Table 1:

Recommended Design Features for Print Materials

General Recommendations

1. Focus on one or two main messages

2. Present essential information by itself or first

3. Use analogies that are familiar and culturally appropriate for the target     audience

4. Add video to verbal narratives, if appropriate

5. Print in sans-serif font

6. Maintain white space

7. Left-justified format

Principles for Text

1. Use short sentences

2. Use conversational style in writing

3. Limit medical jargon

4. Avoid ambiguous words

5. Use familiar words and use them consistently

6. Avoid symbols and question marks

Principles for Presenting Numerical Information

1. Present numerical information in tables or pictographs rather than text

2. Present numerical information so that the higher number is better

3. Present numerical information with the same denominator

4. Use natural frequencies rather than relative risks

5. Where possible when presenting statistics, use words a like “half,” “one-third

Adapted from McCaffery et al29

According to the International Patient Decision Aid Standards (IPDAS) checklist, PtDA materials should be understood by patients with an 8^th grade level or lower^30,31 and should be tailored to the target patient population.²⁹ Although clearly the intent of ptDA’s is to ensure patient understanding in general, health literacy and the recommended design tenets for addressing LHL are rarely mentioned explicitly in ptDA studies.²⁹

Verbal communication

In contrast to the large body of literature focused on print health literacy, the literature on verbal communication with patients with low literacy is scarce. This may be due to the lack of tools for easily quantifying the health literacy level of the verbal exchange.³² The amount of information understood by the patient is influenced by health literacy-listening skills and the spoken message (wording, sentence structure, delivery), which can be difficult to measure. We are not aware of any publications that guide scripting health care decision conversations for patients with LHL.

Doak, Doak, and Root published recommendations for creating audio-recorded health materials for patients with low literacy that mirror many of the best practices for creating print materials.³³ The principles are: limit the objectives, be interactive, and state the key points first and/or last. They also note that “the message is best understood if presented at a normal speaking rate”, even for patients with poor reading skills, and that speaking slowly may make the listener focus on one word at a time rather than the whole message.³³ Other recommendations include avoiding jargon, being culturally respectful, focusing on actions, and confirming understanding.³⁴

Confirmation of Comprehension

Confirmation of patient understanding of the medical evaluation and treatment options is fundamental to SDM. ^29,35,36 Self-reported comprehension among patients with LHL is unreliable. Several strategies to assess comprehension (usually for discharge instructions) have been recommended, including the “teach-back” or “show-me” method, but these have not been studied extensively in the ED setting.³⁷ The communication required in SDM is much more complex than what is required for discharge instructions, so further investigation is needed to test their effectiveness.²⁵

Stage 2 of SDM in the ED: Understanding the Consequences: Risks, Limitations, Benefits, and Uncertainties

Numeracy plays a particularly important role in how clinicians present and patients understand risks, limitations, benefits, and uncertainties for diagnostic and treatment options. A single communication method or format is not likely to be successful for every healthcare provider, patient population, numeracy level, or health literacy level.^38,39 More research is needed to develop and evaluate methods for rapidly and accurately communicating and discussing risk across a range of literacy levels and learning preferences. At present, in order to understand and communicate risk and uncertainty to their patients, clinicians must themselves have high-level numeracy skills. Lower numeracy, common even among clinicians,^40–42 influences the ways that providers present information to their patients, which further impacts patient perceptions, decisions, and satisfaction with the decision process.^43–49 Provider numeracy skills and potential biases in their perceptions of risk and uncertainty unrelated to numeracy must be addressed in approaches to SDM.^50–53 Risk perception is also influenced by the method used to present the information.⁵⁴ For example, use of vague language describing probabilities rather than written, discrete numbers influences understanding and interpretation of information.⁵⁵

Techniques are available to assist patients in understanding risk, but research is needed to determine whether these principles are generalizable to patients with LHL and/or low numeracy and to develop methods for ensuring that the information is clearly communicated by healthcare providers. This may require better understanding of the current state of health communication in the ED settings, which may in turn lead to development of interventions addressing areas in need of improvement or change. Further, numeracy skills of providers are also variable and physicians are often subject to various inherent biases in estimating and presenting risk. How clinician estimation and presentation of risk and uncertainties are important areas of focus in need of further work.

Stage 3 of SDM in the ED: Identifying Salient Preferences and Combining Utilities with Probabilities

In this stage of SDM, patients identify their ‘preferences’ for the possible outcomes of a decision and combine these with information about the likelihood of each outcome occurring. Formal approaches to this task developed in the Decision Analysis literature typically involve making the problem and its objectives explicit, listing alternative actions and how these alter future events using probabilities, values, and trade-offs, and finally, synthesizing the balance of benefits and harms of each alternative.⁵⁶ Values clarification exercises are intended to help identify and weigh specific preferences and may attempt to quantify a patient’s utilities and values for a given health state. For example, consider two similar ED patients with similar neurologic deficits from an acute ischemic stroke and similar scores on measures of quality of life, pain and distress. One patient may value a shorter lifespan without disability the same as a longer lifespan with disability, while the second patient may have the opposite preference. In the case of stroke, this exercise is made more complicated by the risk of intracranial hemorrhage that could result from treatment with thrombolytics. Such trade-offs are conceptually challenging and are unintuitive ways to make real-life and real-time decisions, particularly among patients with LHL and/or low numeracy skills. Patients with LHL have difficulty imagining future health states and difficulty understanding and articulating their own health values and may not recognize this difficulty until after a medical decision has been made.^57,58–63 Though patients with LHL are more likely to have decision uncertainty and decision regret after a medical decision,²⁹ they also have the greatest gains in content knowledge from the SDM process.⁶⁴

Though a formalized approach to eliciting patient preferences is more robust and provides quantifiable information, it is time consuming, requires training to conduct, and may not be feasible in the ED or appropriate for patients with LHL. In light of this, other simplified approaches to values clarification have been proposed, such as the Institute for Healthcare Improvement’s concept “What Matters to You?,” which asks this question of patients as a means to guide decision making.^65,66 A similar approach is used in The Informed Medical Decisions Foundation’s “Six Steps of Shared Decision Making” which includes “assisting patients in evaluating options based on their goals and concerns.” This provides sample language such as, “Just as people are different, no one decision is right for everyone. As you think about your options, what’s important to you?”⁶⁷. These simple approaches for assessing patient values and goals may help facilitate clear communication. However, they do not rise to the same level of rigor in values clarification and risk superficiality. Perhaps similar open-ended questions targeted around a specific question might provide a middle ground. This is an area that requires further study in the acute setting.

Stage 4 of SDM in the ED: Participating in the Decision

Stage 4 of the SDM process requires first that patients understand that their “involvement and choice are possible”; they must have confidence to “express preference, articulate choice, ask questions, and negotiate with health professionals.” ²³ Patients with LHL often initially decline to participate in or report aversion to SDM,^60,68,69,64 though they may gain the most from its use. This aversion may arise from a greater perceived power imbalance between the physician and the LHL patient.^{23, 71} Patients with LHL may also view SDM as a decision to consent to a doctor’s recommendation, and that the patient bears responsibility for the ultimate decision made.^70,23,71 Clinicians’ perceptions that communication with LHL patients is challenging is also an obstacle to implementing SDM.⁷¹

Participation in SDM requires three levels of health literacy skills: 1) functional health literacy or patient comprehension of the medical encounter; 2) interactive health literacy including social and communicative skills necessary to extract, elucidate, and discuss information with others; and 3) communicative and critical literacy, including cognitive and social skills to evaluate, interpret, and ultimately, make informed decisions.” ^29,35,36 Both level (high compared to low/limited) and type of health literacy (functional, communicative, or critical) influence question asking, which is critical to participation in SDM. When they do ask questions, patients with LHL ask more clarification questions and fewer medical and lifestyle questions, suggesting attempts to gain a basic understanding of the information presented, i.e. functional health literacy in its most fundamental sense.^72,73

During this stage of SDM, the key question is how to foster the interactive, communicative, and critical literacy processes required for successful SDM in the ED. Standardized question prompt lists, patient cue sheets, patient agendas, and patient decision aids have been proposed to increase patient question asking, participation in the SDM process, and patient satisfaction.²³ Patient decision aids have been among the most promising of these interventions. As noted, more research is needed to better understand the development, use, and effectiveness of these aids in limited health literacy populations.

The degree to which patients with LHL and/or low numeracy skills in the ED setting want to participate in the SDM process has not been examined in depth. These patients in particular may first decline the opportunity to participate in SDM for their clinical care, but in many cases initial deference to clinician recommendation changes after explanation or experience with SDM. The resources needed to adequately prepare this vulnerable population to effectively and equitably participate in the SDM process in outpatient studies has been considerable, requiring six hours, in one study.⁷⁴ These resources and time are not typically available in the rapid-paced ED setting, and in the setting of acute illness or cognitive impairment would likely require even more resources and time. The question of how best to involve our patients to the degree that they are able and prefer is an important one and underpins all the other SDM research foci in this patient population.

Stage 5 of SDM in the ED: Implementing a Shared Decision

In Stage 5 the patient and clinician must implement an agreed-upon choice. This stage assumes successful completion of stages 1–4. Greater evidence is needed to guide the rapid and reliable assessment of patient understanding at each SDM stage. Comprehension tools may vary with each stage of the SDM process, e.g., understanding of the disease may be measured by a separate tool that measures patient understanding of the risks and uncertainties of treatment options; these tools may also vary with each clinical condition. A core set of tools designed to accomplish this task that may be modified without sacrificing validity will be helpful in broad evaluation and implement of SDM among patients with LHL and/or low numeracy in the ED setting. Similarly, the impact of patient literacy and numeracy skills on their ability to clarify values and preferences is not well studied. Approaches are needed that avoid unnecessary complexity while balancing oversimplification in this critical SDM stage. Like questions about the best use of decision aids, the best approach and level of rigor applied to answering these questions may vary with the clinical importance of each quest or clinical scenario.

Implementation is also a shared act which places responsibilities on both parties and may require resources to accomplish. In the ED, many decisions to test and treat are both started and completed during a single visit. Clear communication techniques and mindfulness in avoiding complex discharge instructions, medication regimens, and follow up plans are starting points to help improve implementation of SDM plans. However implementation of decisions that require patient adherence, such as timely follow up with a provider may be challenging in the absence of resources such as transportation, etc., particularly for patients with LHL. Research is needed to determine how EPs can better partner with patients and caregivers and coordinate with other specialties to improve implementation of SDM in the ED.

SDM in the ED: Summary

Although conducting SDM in the ED setting presents a number of challenges that are magnified for patients with LHL and/or low numeracy skills, this vulnerable patient population may gain the most benefit from the SDM process.²⁹ Given the prevalence of LHL in ED patients, evidence guiding the effective and appropriate development and testing of the use and impact of SDM in the ED setting among this population is therefore highly important. In light of the challenges ahead, an initial focus on patients with adequate health literacy may be a reasonable starting point as a proof of concept. Incorporation of principals and best practices as part of universal precautions will benefit all patients, while examination of tailored approaches may be pursued in future research.

Priority questions for research in SDM among patients with LHL and/or low numeracy skills in the ED include:

• 1) What are the best approaches to the design, use, and evaluation of patient decision aids, attempting to balance the competing needs for parsimony and completeness?

• 2) How can we ensure comprehensive and transparent communication of risk and uncertainties between providers and patients?

• 3) What methods will maximize patient engagement and optimize the ability to implement a shared decision, including assessment of patient preferences and values, in the SDM process?

• 4) How can we best confirm patient understanding at each stage of SDM in a reliable, feasible and objective fashion?

Conclusion:

Though data suggest that patients with limited health literacy and/or low numeracy skills stand to gain the most from SDM, the impact of these conditions on the effective implementation of SDM in the ED setting is not well understood. We identified four consensus-driven research priorities to address important gaps in knowledge to help guide future work to improve SDM for this patient population in the ED.