Guide to Patient and Family Engagement
Table of Contents
Below, we discuss findings from the environmental scan. We have grouped findings according to the main categories identified in the preliminary conceptual framework (Exhibit 1):
- Individual Characteristics of the Target Audiences: Patients, Family Members, and Providers.
- Organizational Context.
- Strategies and Interventions.
- Key Informant Interview Recommendations for the Guide.
Individual Characteristics of the Target Audiences
In this section, we sought to identify how the individual characteristics, perspectives, and experiences of the target audiences for the Guide—patients, families, and health care professionals—affect their willingness and ability to adopt behaviors related to patient and family engagement. We have grouped findings into three main categories:
- Patients, family members, and health professionals' understanding and interpretation of health care quality, safety, and patient and family engagement.
- Patients, family members, and health professionals' experiences with and attitudes about behaviors and roles associated with patient and family engagement to improve quality and safety.
- Possible facilitators and barriers to those behaviors and roles.
Exhibit 6 highlights the strength of evidence for articles reviewed in this section. Generally, this evidence represents lessons learned from the larger body of literature on patients' and providers' perceptions of and their willingness and ability to engage in behaviors related to quality, safety, patient- and family-centered care, and shared decisionmaking. The evidence comes primarily from qualitative and quantitative research studies and systematic literature reviews.
The focus of the studies reviewed in this section is evenly divided between patients and providers. The bulk of the literature on patients relates to patients or consumers more broadly; few articles specifically address family members. Similarly, the provider literature generally focuses on physicians and nurses, although several articles highlight the role that other health care professionals, such as social workers1-3 or health educators,4 can play to bridge the gap between patients and providers. There was a dearth of information related to nonclinical hospital staff, although we did review one article describing an intervention to train housekeepers in patient safety issues and include them as part of the health care team.5
Understanding and Interpretation of Quality, Safety, and Engagement
In this section, we discuss findings from the literature about how patients and providers understand and interpret the concepts of health care quality, safety, and engagement.
Health Care Quality
Both patients and providers tend to think that the quality of care they receive or give is generally good, despite evidence that suggests this is not always the case.6-12 For example, in one qualitative study of consumers, participants could envision a health care provider making an occasional mistake, but they found it hard to believe that providers could deliver truly substandard care, particularly when it came to their own providers.13 Likewise, in a survey conducted by the Kaiser Family Foundation (KFF), fewer than half of individuals surveyed reported perceiving "big" differences in quality among different health-related providers.14
Although patients and providers have a shared belief that health care quality is generally good, they do not always share the same views of what constitutes quality or safety. Below, we discuss differences in these perceptions.
Patients and family members. A systematic review of the literature found that patients consider quality primarily in terms of a provider's ability to relieve their symptoms while keeping them safe from injuries.15 Moreover, patients often feel that they, not outside "experts," are in the best position to judge high-quality care. In a study that examined consumers' reactions to quality information and clinical quality standards, participants felt that quality information was less significant than their own perceptions of quality. In addition, consumers in one study noted that "bad" care was something they would recognize when they saw it and therefore felt that they did not need to rely on external arbiters of quality.13 When asked to identify poor-quality care, patients frequently pointed to lapses in service quality such as waits and delays, lack of sufficient information, and poor care coordination.16-19
Patients' and family members' perceptions of quality are also influenced to a large degree by their perceptions of a given provider. Because of this, patients often assess the quality of care primarily based on their interpersonal interactions with the provider, as opposed to the provider's specific clinical skills in treatment and diagnosis.13 Providers who are perceived to be responsive, empathetic, and attuned to patients' needs are judged by patients as being of higher quality than providers who are perceived to be less responsive and empathetic, even if the clinical care delivered is the same.13,15,20,21
Providers. In evaluating quality of care, providers tend to emphasize the clinical aspects of quality in terms of skills in diagnosing, treating, and obtaining positive clinical outcomes.21 Within a specific clinical encounter, a physician may focus more on biomedical aspects of the patient than other factors related to interpersonal communication, such as assessments of the patient's emotions, mood, expectations, or personal life.22 Providers do recognize the importance of communication as a component of quality, but they also tend to be overly positive in their perceptions of how effectively they communicate.23
As with quality, patients and providers differ at times in their understanding of patient safety. Moreover, they may not perceive safety in the same ways as researchers or other experts in the field.
Patients and family members. Consumers often have a narrow view of patient safety, seeing safety primarily in terms of medical errors. In one quantitative study where consumers ranked the importance of issues in health care, consumers did not equate the term "patient safety" with what they see as the more important issue: "medical errors."24 Likewise, during cognitive testing of AHRQ's Five Steps to Safer Health Care,c Medicare beneficiaries equated the term "patient safety" specifically with falls and accidents instead of recognizing more general aspects of care safety.25
Although consumers may lack a complete picture of patient safety, they are aware that medical errors occur; one in three people says that he or she or a family member has experienced a preventable medical error and report that these errors often result in serious health consequences, such as severe pain or long-term disability.26 However, while medical errors may be more salient than the general concept of patient safety, nearly half of consumers sampled in a 2006 Kaiser Family Foundation survey were still unfamiliar with the term.26
In general, patients and family members assume that most care is safe and that there are system checks to prevent errors.25,27When errors do occur, patients tend to think individual providers cause them, not the systems in which the providers work.26,28 In addition, consumers tend to underestimate the number of medical errors that occur. For example, about half (49 percent) of consumers sampled in a KFF survey thought that 5,000 or fewer deaths occur in hospitals each year as a result of medical errors.28 However, in 1999, the Institute of Medicine (IOM) estimated that this number is far greater—between 44,000 and 98,000 deaths each year.8 Also of note is that studies have shown that patients with limited English proficiency are at higher risk of adverse events, although there is little research to assess specifically, the perspectives of culturally diverse patients on safety.29
Providers. In general, providers conceptualize safety more broadly than do patients, recognizing safety as an important component of quality. However, providers—like consumers and patients—may also view errors as individual deficiencies that are within the control of the individual provider, as opposed to system deficiencies.30 Because providers (and particularly physicians) often think that errors are under their individual control and because they are worried about the level of effort required to address errors on a system level, they may not prioritize efforts to improve patient safety on a system level.30 Physicians also are wary that focusing on and creating transparency about medical errors will lead to an increased risk of litigation.30-32 For example, physicians in one study expressed concern that the term "medical errors" in AHRQ's "Five Steps to Safer Care" would signify liability and negligence.32
Within the general group of "providers," physicians and nurses may have different perspectives on patient safety. In a focus group study investigating perceptions of patient safety, nurses most frequently mentioned environmental, medical, and general safety issues as the main areas of concern. Most comments described environmental safety, such as the use of bed rails, alarms, patient falls, or restraints. In addition, nurses most frequently mentioned self-initiated "double-checking" as their main safety task.33 In the 2004-2005 Patient Safety Climate Survey administered to 92 U.S. hospitals, Singer and colleagues found that, compared with physicians, nurses perceived more problems with teamwork and were more negative toward their unit's support and recognition of safety efforts.34 In addition, nurses expressed less personal embarrassment with regard to medical errors than did physicians.34
Patient and Family Engagement
Very few studies address patients' and providers' views on "engagement" specifically. However, findings from the broader literature on patient participation/involvement, patient- and family-centered care, and shared decisionmaking shed light on these perspectives. Although engagement is a broader concept than participation or shared decisionmaking, taken together, this literature facilitates our understanding of how patients and providers view engagement.
General perceptions of patient participation. Conceptually, both patients and providers support patient and family involvement and participation in their own care and recognize that it can lead to better patient experiences and outcomes. For example, in one survey of physicians and nurses, participants reported the perception that involving patients in their own care can lead to better outcomes.35 Similarly, several studies of nursing staff and general practitioners have demonstrated positive feelings about ideas relating to patient involvement, to the degree that patients themselves feel comfortable.36-38
Shared decisionmaking—patient perceptions. Patient support for participation becomes more uncertain when patient involvement includes making diagnosis or treatment decisions. Analyses of multiple studies on shared decisionmaking indicates that most patient populations want to be involved in decisions about their health care but do not wish to make autonomous decisions or take on what they perceive as the provider's role.39,40 Likewise, a long-term cohort survey found that consumers as a whole preferred to share information with and receive information from their health care providers, but that individuals differed in their preferences for involvement in decisionmaking.41 However, patients also are amenable to increased involvement in the decisionmaking process. In one qualitative study of patients who were readmitted post-discharge, patients felt that readmission could have been prevented by enhanced patient education and involvement in the discharge decisionmaking process.42
Shared decisionmaking—provider perceptions. As with patients, most providers express general support for patient participation. For example, results from a systematic review of 38 studies examining health attitudes toward shared decisionmaking indicate that physicians perceive shared decisionmaking as leading to a positive impact on both patient outcomes and clinical practice.43 However, providers also may have reservations about the reality of involving patients in the decisionmaking process. The same systematic review noted that a frequently mentioned barrier to implementing shared decisionmaking was a lack of applicability in individual situations because of patient characteristics and the clinical situation.43 In addition, physicians most frequently cited time constraints as a barrier to implementing shared decisionmaking, suggesting that even if physicians support the idea in concept, there may be barriers to operationalizing it in practice.
Perceptions of the involvement of patient and family members on an organizational level. Few studies exist to provide information about whether patients and families want to participate in organization-level processes or decisionmaking with regard to quality or safety (e.g., as members of patient and family advisory councils). Information obtained from the key informant interviews conducted for this scan indicated that patients and family members are often motivated to get involved at an organizational level to "give back" because of positive experiences or to find ways to improve the system because of negative experiences.
Despite a willingness of certain patients to get involved at an organizational level, there are also significant perceived barriers to engaging in this way. In an Australian study of how consumers preferred to be involved in improving the quality of health care services in a medical center, participants identified the following barriers to participation: lack of time (44 percent); not interested (33 percent); costs involved and difficulty in transport (15 percent); poor health (14 percent); have no issues with medical center (6 percent); and lack of confidence (5 percent).44 Likewise, patients may be intimidated or unsure about roles and expectations for engagement at this level, concerned that their participation will not lead to changes or results, fearful of dealing with conflict, or concerned about behaviors from health care professionals that are not respectful.45
On an organizational level, providers generally perceive benefits of patient and family involvement. For example, in one qualitative study, health professionals felt that including patients in health care planning could help provide accountability, balance priorities, identify patients' needs, increase recognition of the role of patients, and allow mutual learning. However, providers were most comfortable with patients in the role of consultant (i.e., providing advice and feedback) and less comfortable with patients being involved in the organizational decisionmaking process.46
Patient Experience and Attitudes About Behaviors/Roles
In this section, we discuss how patients and family members feel about specific roles and behaviors related to patient and family engagement (including their baseline level of experience with these behaviors). We conclude with a brief discussion of patients' and providers' attitudes toward the involvement of family members in ensuring safety and quality.
Patient and Family Roles
Most of the literature on patient and family engagement roles focuses on what patients could do (or what researchers and policymakers want patients to do) instead of discussing what behaviors patients and family members currently engage in or would be willing to engage in. From this broader perspective of how patients and family members could be engaged in safety and quality, the literature focuses on the following:
- Being informed, asking questions about, and participating in their care. This includes making informed choices about providers, participating in reaching diagnoses, deciding on treatments, and participating in care coordination.15,47,48
- Helping prevent specific safety events/medical errors. This includes medication safety, infection control, and, to a lesser degree, surgical safety. In addition, patients and family members can observe and assess care practices for consistency, accuracy, and safety.48,49
- Reporting on safety events after the fact. This includes identifying adverse events that occur and reporting on these events.15,16,18,19,48
With regard to what patients themselves believe they can do, they generally believe that they have a role in health care safety, particularly with regard to the prevention of medical errors. For example, in one survey, 91 percent of patients agreed that patients could help prevent medical errors, and 98 percent agreed that the hospital should educate patients about error prevention.15,47
Patient and Family Experience in Quality and Safety
The literature on consumers' experience engaging in safety and quality generally reflects patients' and families' willingness to engage in potential behaviors, along with a few self-reports of actual behaviors. We found little observational research to provide evidence about behaviors that patients and families actively engage in within the hospital.
In general, the literature suggests that patients are more likely to engage when the goal relates to obtaining specific information about their care (e.g., asking questions to get information about their condition, treatment, functional activities, or discharge). However, patients are less likely to engage when faced with behaviors that seem new or confrontational (e.g., asking providers to wash their hands or mark surgical sites).24 For example:
- One survey found that patients were very comfortable asking questions about a medication's purpose (91 percent) and general medical questions (89 percent), but they were less comfortable asking medical professionals about hand washing (46 percent).47 This same survey found that when hospitalized, 85 percent of patients asked medical questions, and 75 percent asked about a medication's purpose, but only 17 percent marked their surgical site, and only 5 percent asked about hand washing.47
- In another survey, when asked about perceived likelihood or frequency of engaging in patient safety practices, more respondents were likely to ask for an explanation of something that they did not understand (91 percent); to question the reason for a procedure in the hospital (85 percent); or to question unfamiliar drugs in a hospital (84 percent). Fewer respondents were likely to ask for identity confirmation before a procedure (40 percent) or ask about hand washing (26 percent).50
- Finally, another study comprising 14 semi-structured patient interviews found that patients gathered information about the purpose of particular medications (i.e., information gathering) but did not confirm whether it was actually the right drug or correct dose, behaviors that are further outside the expected norm.27
The studies described above support what researchers found during the development of AHRQ's Five Steps to Safer Health Care: patients prefer messages that promote a collaborative patient-physician relationship as opposed to a confrontational one.25
Patient willingness to engage in perceived confrontational behaviors with providers may depend on the type of provider (i.e., physicians vs. nurses). For example, surgical patients were significantly more likely to ask physicians factual questions (e.g., When can I return to normal activities?) and were more likely to ask nurses challenging or confrontational questions (e.g., Did you wash your hands?).51
Patient and Provider Attitudes and Experiences with Family Involvement
The majority of the literature reviewed on family involvement focused on family involvement during rounds and invasive procedures. A brief discussion of findings related to family involvement follows.
Family member presence during rounds. Patients and family members indicate that the presence of family members during rounds can improve communication, facilitate information exchange and the provision of new information to medical staff, and generally improve care.52 Recent studies have found that patients and family members prefer to have at least one family member present during rounds.52-54 Results from a survey administered to staff, family members, and patients showed that family members believed that their presence would improve the medical staff's attitude toward the patient/family member and that it would improve communication with the medical staff.54 Another observational study showed that patient perceptions about more effective communication may be accurate, demonstrating that new information was discovered from family members 46 percent of the time.52
Family presence during resuscitation and invasive procedures. In one survey assessing patient and family wishes for family presence during invasive procedures, both patients and family members responded that family member presence was important. Patients preferred family member presence because it would be comforting to have someone with them. Further, family members responded that it was their right to participate in a resuscitation or an invasive procedure and that participation would allow them to better understand the patient's condition.55
Provider perceptions of family involvement. Evidence suggests that providers have mixed attitudes toward family involvement. On the one hand, providers report that family presence can be beneficial to patients and family members55 and useful in helping calm or comfort the patient.37,56 On the other hand, providers may also place certain limits on family involvement. For example, one survey found that although health care providers thought that family participation in simple, hands-on care was acceptable, they expressed concern about family member presence, particularly during more invasive procedures.37 Concerns associated with family member presence include potential interference with treatment, medical risk (e.g., exposure to infections), or the emotional response of the family member.55-57 These same studies also suggest that providers are concerned that the presence of a family member could increase the anxiety of the health care providers in the treatment room, resulting in the decreased success of a procedure.55-57
Facilitators and Barriers to Patient and Family Engagement in Safety and Quality
In this section, we explore facilitators and barriers related to patient and family engagement behaviors, particularly around safety and quality. We first examine the characteristics (e.g., demographic information, experience) that affect the ability of individuals to engage in those behaviors. We then discuss what hinders patients' engagement (barriers) and what helps patients engage (facilitators). We also discuss barriers and facilitators to providers' support of patient and family engagement. Additional information about how hospitals as organizations can support patients and providers in engagement behaviors is discussed in the findings section on organizational context.
Influences on Patient and Family Engagement
In this section, we present findings about how demographic characteristics and prior experiences may influence patient participation, desire for shared decisionmaking, and engagement in safety behaviors.
Demographic characteristics. Individuals may differ in their desire for engagement based on demographic characteristics. For example, a systematic literature review and a study analyzing the data from six independent trials on shared decisionmaking found that patients who are younger, female, and more highly educated are more likely to want to be involved in health care decisions.40,58 In addition, providers may be more likely to facilitate and reinforce engagement with certain types of patients. For example, one observational study found that specialists displayed more good communication behaviors with patients who were younger, more educated, and urban.22 In contrast, patients who are at a power disadvantage (e.g., because of lower levels of education, language, or literacy difficulties) may be less likely to engage. For example, one mixed-methods study found that patients who spoke through an interpreter made significantly fewer comments during medical encounters than English-speaking patients.59
Although there is a dearth of literature in this area, race may also influence a patient's willingness or ability to engage in patient safety behaviors. A survey pilot test of elderly individuals found that African Americans were significantly less likely than whites to report participation in patient safety behaviors.60 Another quantitative study with 2,765 respondents found that both Hispanics and African Americans were more likely than whites to prefer leaving decisions about medical care to the doctor.61 Moreover, cognitive testing of the Spanish version of AHRQ's Five Steps to Safer Care revealed that many Spanish-speaking consumers were uncomfortable asking any questions of their physicians and did not know they could ask providers for test results.25
Experience. Experience may affect an individual's skills in navigating the health care system, desire for participation, or engagement in safety behaviors. One systematic literature review found that patients with more experience in the health care setting were better able to navigate the interpersonal interaction between patient/family and provider.62 Similarly, experience over the course of a hospital stay may also affect the level of participation. In a mixed-methods study, family members were less likely to feel comfortable asking questions and more likely to find discussion confusing when interacting with providers during rounds on the first day of admission than before discharge.52 Likewise, another mixed methods study found that patient preference for involvement in care was greater at discharge than at preadmission.63
Individuals who have experience with a particular condition or have chronic health issues also may be more willing to engage in patient safety behaviors, particularly those that may seem confrontational. A qualitative study found that patients with chronic conditions were particularly careful about medication administration and were more vocal about advocating for safety.27 Likewise, in a small survey comparing patients who had a history of Methicillin-resistant Staphylococcus aureus (MRSA) with those who had no history of MRSA, those with a history of the infection were more likely to ask providers to wash their hands.64 However, there may be limits to this willingness. One quantitative study of shared decisionmaking found that acutely ill patients preferred to leave all decisions to the doctor or to let the doctor make the final decision after having considered the options.65
Finally, patients and families may be more willing to take a proactive role in their health care if they are familiar with the procedure or treatment. For example, one study of patients receiving regular chemotherapy treatments in an outpatient clinic found that they reported engaging in behaviors such as detecting procedural errors (e.g., unexpected events that occurred outside of treatment procedures) and coordinating tasks across multiple providers (e.g., not to get adhesive on the skin by a provider administering chemotherapy when undergoing radiation).49
Barriers: What can hinder patient and family engagement?
Barriers at the patient/family level can affect engagement in health care more generally, and engagement related to quality and safety more specifically.
Fear and uncertainty. Patients and family members can be unsure how to be involved and may feel overwhelmed and intimidated by the health care system. One systematic review of the literature found that patients and family members reported being unsure of what to expect in a medical encounter and feeling intimidated by health professionals.62 Beyond the individual encounter, a study based on focus groups and interviews with families and providers found that families felt overwhelmed by responsibilities for navigating the health care system and by the need to transmit highly technical information from one specialist to another.66
Health literacy. In one mixed-methods study, more than half (53 percent) of survey respondents agreed or strongly agreed with the statement "Most medical information is too hard for the average person to understand."67 The National Assessment of Adult Literacy reports that more than 90 million American adults have marginal literacy skills and are unable to read anything beyond very simple text. Additionally, more than 80 percent of American adults have less than "proficient" health literacy skills.68 In one qualitative study of elderly women who were participating in discharge planning conferences, patients noted difficulties taking part in the discussion because providers used professional language or because they felt overwhelmed by too much information.69
Provider reaction. Lack of provider support for engagement also can pose barriers to patient and family involvement. For example, one survey found that patients who felt as though they were not being given the information they needed from their providers (e.g., getting a prescription without explanation) were less likely to participate in their health care.70 In another study, patients reported significantly higher levels of satisfaction when they perceived physician facilitation of their involvement.71
Facilitators: What helps patients and family members engage?
Several factors can help patients and families engage in their health care.
Self-efficacy. In general, patients with higher self-efficacy are more likely to engage in patient safety behaviors. For example, a survey of patients found that patients who reported being "very comfortable" with specific behaviors were twice as likely as others to ask medical personnel about hand washing, mark their surgical location, have friends/family members watch for errors, ask general medical questions, report an error, and clarify a drug's purpose.47 In another study, patients with low self-efficacy and those who perceived their actions as less effective in preventing medical errors were 33 percent less likely than other patients to report a willingness to engage in preventive actions.24
Information. Providing patients with information may help them more fully engage in their care by increasing self-efficacy, confidence, and/or awareness. For example, one study found that patients who received less information about MRSA and hand hygiene upon hospital admission felt more anxious about asking providers to wash their hands.64 In another study, patients reported feeling overwhelmed and unprepared to take on an active role in their health care at discharge, noting that more information from providers would help.72 Information that patients indicated they needed to be able to assume an active role in their postoperative care included information about diet, postoperative functional activities, and care management. Interestingly, this study suggests that providers' perceptions of what information patients need may not match with patients' preferences. For example, nurses ranked wound care management as more important information to provide than the topics patients identified.63
Information is most helpful when it is tailored to patients' individual needs and concerns. For example, one study demonstrated that when patients received individualized care information, they were more likely to engage and participate in their health care.73
Invitation to participate. Extensive research has shown that although information is necessary, it is not sufficient to support behavioral change. Patients and families must also be motivated to participate in engagement behaviors through encouragement from others, such as providers, organizations, and systems and through positive feedback when they engage in these desired behaviors. One review of the literature in this area found that patient participation increased in interactions with those health care providers who responded positively to patients' needs and views.58 Likewise, physicians' instructions to patients about asking questions significantly increased patients' willingness to ask challenging questions of both physicians and nurses.51
How can providers support patient and family engagement?
As noted above, reactions by providers can be a barrier to patient and family engagement, and their active invitation for patients and family members to participate can be a facilitator. In general, the bulk of the literature on what providers can do to support engagement focuses on communication. In one observational study, the authors found that patients contributed more to the visit when physicians displayed "facilitating behavior" such as nonverbal or verbal encouragement, summarizing patients' comments, educated guesses, emotional reflections, or attentive silence.1 In addition, a patient survey found that increased satisfaction with elements of physician-patient communication (e.g., overall communication, explanations, use of understandable words, listening, patient involvement in decisions) was related to an increased sense of empowerment.74 Information from our key informant interviews noted that providers can make the "invitation" to participate more explicit by introducing themselves when entering the room, introducing all staff during rounds, sitting down, and asking about the patient's needs.
What hinders providers' ability to support engagement?
Even when providers see the need for better communication and patient-centered care, they may find it difficult to operationalize those skills in practice. For example, one survey found that even though providers recognized the importance of the elements of family-centered care, these elements were not always incorporated into care.75 Barriers to providers' support included professional norms and experiences and perceived level of effort.
Professional culture and practice norms traditionally have been based on individual autonomy, which can be a barrier to the teamwork and patient-centered practices that a patient safety culture requires.76 Also, providers lack experience with models that encourage collaboration with patients. Traditionally, professional schools have offered limited or no emphasis on patient/family engagement or patient- and family-centered care, and this applies not only to medical schools but also to academic programs that train health care leadership.77
Providers also may believe that patient and family engagement requires increased time and a greater level of effort. A systematic literature review and survey found that providers saw time constraints as a significant barrier to patient involvement.35,43 In a mixed methods study, 82 percent of medical residents perceived that rounds took longer when families were present.52 In fact, one article suggests that clinicians sometimes discourage patients from telling stories because they are afraid that it will take too much time.78 In addition, providers may be concerned about the time that it takes to educate patients. In a qualitative pilot study, nurses expressed concern that they do not have the time to spend educating patients about their health status, health care, or medication.38
Despite these concerns, some data do not support these perceptions of increased time and level of effort. A literature review and a mixed mode study found that patient-centered interviews do not take significantly more time (only 1 to 2.5 minutes more) than traditional interviews.52,79 Moreover, although a survey found that post-acute care unit (PACU) staff reported that a barrier to the new formal visitation policy was the staffing issues that arose, only 3 percent of PACU staff (not including physicians) reported that the presence of family members had hindered their performance.57
Individual Characteristics of the Target Audiences: Summary of Key Points
c. Five Steps to Safer Health Care. Rockville, MD: Agency for Healthcare Research and Quality; 2004. Available at http://www.ahrq.gov/patients-consumers/care-planning/errors/5steps/index.html
Page originally created September 2012