- Email: [email protected]
Implementation of a health-literate patient decision aid for chest pain in the emergency department
G Model PEC 6447 No. of Pages 6
Patient Education and Counseling xxx (2019) xxx–xxx
Contents lists available at ScienceDirect
Patient Education and Counseling journal homepage: www.elsevier.com/locate/pateducou
Implementation of a health-literate patient decision aid for chest pain in the emergency department Kristie B. Haddena,* , Heather McLemoreb , Wesley Whiteb , Matthew H. Marksb , Jennifer M. Gana , Rawle A. Seupaulb a b
University of Arkansas for Medical Sciences, Center for Health Literacy, Little Rock, AR 72205-7199 USA University of Arkansas for Medical Sciences, Department of Emergency Medicine, Little Rock, AR 72205-7199 USA
A R T I C L E I N F O
A B S T R A C T
Article history: Received 4 September 2019 Received in revised form 22 October 2019 Accepted 11 November 2019
Objective: The aim of this study was to investigate the implementation of a new health-literacy-tested patient decision aid for chest pain in Emergency Department (ED) patients. Outcomes included disposition, knowledge, decisional conflict and satisfaction prior to discharge. Patient health literacy was explored as a factor that may explain disparities in sub-group analysis of all outcomes. Methods: A health-literacy adapted tool was deployed using a pre/post intervention design. Patients enrolled during the intervention period were given the adapted chest pain decision aid that was used in conversation with their emergency medicine physician to decide on their course of action prior to being discharged. Results: A total of 169 participants were surveyed and used in the final analysis. Patients in the usual care group were 2.6 times more likely to be admitted for chest pain than patients in the intervention group. Knowledge scores were higher in the intervention group, while no significant differences were observed in decisional conflict and patient satisfaction, or by patient health literacy level. Conclusion and practice implications: Using the adapted chest pain decision tool in emergency medicine may improve knowledge and reduce admissions, while addressing known barriers to understanding related to patient health literacy. © 2019 Elsevier B.V. All rights reserved.
Keywords: Health literacy Chest pain Decision aid Emergency department
1. Introduction Health literacy is “the degree to which individuals have the capacity to obtain, process and understand basic health information and services needed to make appropriate health decisions” [1]. Health numeracy, a component of overall health literacy [2], is “the degree to which individuals have the capacity to access, process, interpret, communicate and act on numerical, quantitative, graphical, biostatistical, and probabilistic health information needed to make effective health decisions” [3]. In accessing and navigating health care, the need for patients to use, understand, and work with numbers can be essential. Examples include choosing a health insurance plan, following medication dosing instructions, management of chronic diseases, and understanding risk and benefits for informed decision making [3–7]. Only 12 % of the American adult population has proficient health literacy, and individuals lack
* Corresponding author at: University of Arkansas for Medical Sciences, Center for Health Literacy, 4301 W. Markham Street, Slot 599A, Little Rock, AR 72205-7199 USA. E-mail addresses: [email protected] (K.B. Hadden), [email protected] (H. McLemore), [email protected] (W. White), [email protected] (M.H. Marks), [email protected] (J.M. Gan), [email protected] (R.A. Seupaul).
quantitative literacy skills even more than prose literacy [2,8]. Low health numeracy can be a barrier to shared decision making because many patients do not understand numerical concepts of risk and benefits [9], which can result in poor health decisions and increased costs [4]. Shared decision making involves collaborative input from both the patient and the health professional [10]. There are tools available to assist patients with making important health decisions; however, decision making tools are often difficult to understand and are rarely designed for patients with below average literacy skills [10]. Patients with low health literacy need more support to encourage participation in shared decision making [11]. Since health literacy affects the use of decision aids among patients when making difficult health decisions [12,13], it is recommended that decision aids are developed using health literacy best practices to facilitate understanding among all patients, particularly those with lower health literacy and numeracy skills [11,14]. Stakeholders across health care systems have recognized the important linkage between health literacy and health status, and are working to provide patient and consumer health materials, including decision aids, that are easy to understand and act upon, promote self- engagement in one’s own health, and lead to better health outcomes [15].
https://doi.org/10.1016/j.pec.2019.11.009 0738-3991/© 2019 Elsevier B.V. All rights reserved.
Please cite this article in press as: K.B. Hadden, et al., Implementation of a health-literate patient decision aid for chest pain in the emergency department, Patient Educ Couns (2019), https://doi.org/10.1016/j.pec.2019.11.009
G Model PEC 6447 No. of Pages 6
2
K.B. Hadden et al. / Patient Education and Counseling xxx (2019) xxx–xxx
1.1. Background and importance Patients who visit the emergency department (ED) may feel pressured to make quick decisions with little input from family or others whom they trust [16]. As a result, these patients typically rely on emergency medicine (EM) physicians to make critical decisions regarding their care or treatment options [16]. Research studies on use of shared decision making in emergency situations revealed that engaging patients in shared decision is helpful for reducing over-testing [16] and lowers health care costs and utilization [17,18]. Common obstacles associated with shared decision making in ED’s include: 1) lack of patient engagement in making important decisions regarding treatment options; 2) time constraints; and 3) lack of comfort or training in encouraging shared decision making [16]. Even low risk patients should have the option to decide between admitted to the hospital for observation or discharged with follow up. Since the risk of a major cardiac event is not 0 %, the patient has the option to be to be admitted for observation based on their informed tolerance of risk. Our leadership aimed to improve quality in our Emergency Department by 1) providing decision support for patients, 2) addressing patient health literacy and provider communication, and 3) reducing unnecessary admissions. To achieve these quality aims, we adapted an existing evidence-based shared decision making tool, created to educate and engage patients in decisions regarding their care. This “CHEST PAIN CHOICE” decision aid safely reduced admission rate for low risk study participants in a randomized control trial [17,19]. This shared decision making tool for low risk chest pain was evaluated for readability and health literacy/numeracy standards to provide a framework for the development of our refined tool. We used plain language best practices for all written portions of the refined tool. We field tested the refined tool with patients identified as having low health literacy using “stoplight coding” qualitative methods [20]. The adapted tool has a 3rd-4th grade readability, which is much lower than the previous tool with a 9th-10th grade readability. We also adapted the icon array using two distinct heart icons to reinforce the positive and negative risk frame which avoids precise percentages or more complex graphing that can be difficult for low numeracy or low graph literacy individuals. Finally, we used health literacy and plain language best practices and participant feedback to determine the adequacy of the tool for patients with low health literacy. 1.2. Goals of this investigation The specific aim of this study was to test the efficacy of a new health-literacy-tested patient decision aid for chest pain in a convenience sample of emergency department patients who used the decision aid compared to patients who were not exposed to the decision aid. Outcomes of interest include knowledge, decisional conflict and satisfaction prior to discharge. Health literacy was explored as a factor that may explain disparities in sub-group analysis of all outcomes between patients with inadequate, marginal, and adequate health literacy. Additionally, disposition outcomes were explored for the adequate/marginal/inadequate health literacy groups as well as the usual care/intervention arms. 2. Methods 2.1. Study design and procedures In the first phase of this project, the new decision aid was developed/adapted using health literacy standards and best practices for use in the Emergency Department (ED). The decision
aid was field-tested and changes were made to improve usability of the newly developed patient decision aid. This study then used a pre/ post intervention experimental design to test the impact of the newly created decision aid tool in clinical practice on outcomes of interest. Patients were recruited over a three-month period prior to intervention to establish the usual care group. Standard care was administered for eligible patients in this period. All ED providers received training on the content and delivery of the shared decisionmaking instrument prior to the fourteen-month intervention period. A training video using a standardized patient was developed to illustrate the practical application of the tool in a simulated clinical setting. In person sessions were also held with all ED providers to answer questions and/or provide clarifications on the adapted tool’s content and method of delivery. Integration of the adapted tool into the clinic workflow was well received by all ED providers. The UAMS emergency department is an urban, university based academic ED with a residency training program, with an annual census of 60,000 patients/year. Patients enrolled during the intervention period were given the chest pain decision aid (Fig. 1) that was used in conversation with their EM physician to decide on their course of action prior to being discharged from the ED. EM physicians asked patients to explain in their own words their risk and options prior to ending the encounter. Data collection procedures were the same for both arms of the study and completed prior to discharge by interview with an EM research assistant (RA). 2.2. Selection of participants Patients who presented with chest pain and were categorized as “low risk” for cardiac events in the ED during weekday, evening, and weekend recruitment periods were eligible to participate. For this study, “low risk” was defined as a HEART score of 3 or less [21,22]. Patients who presented with chest pain and were categorized as “low risk” based on a HEART score of 0–3 with unremarkable troponin were recruited [22,23]. Additional inclusion criteria included twenty-one years old or older and ability to see, hear, speak and understand spoken and written English. Patients who were unconscious or at moderate to high risk for cardiac events or other complications were excluded from participation. 2.3. Measurements Adapted knowledge questions [27] were used to assess the primary outcome of patient knowledge about key information presented in the decision support. A four item survey (3 True/False questions and 1 open response) was administered to assess participant knowledge about chest pain follow up care and likelihood of having a heart attack based on risk factors. These questions were adapted from the 7 item knowledge questions used in a previous trial [17]. Adapted Decisional Conflict Scale [26,27]. The Decisional Conflict Scale measures personal perceptions of uncertainty in choosing options, modifiable factors contributing to uncertainty such as feeling uninformed, unclear about personal values and unsupported in decision making; and effective decision making (in full version) such as feeling the choice is informed, values-based, likely to be implemented and expressing satisfaction with the choice [26]. Adapted questions from the Decisional Conflict Scale [26] were used for this study using a 5 item Likert (1-Strongly agree to 5-Strongly disagree). Adapted satisfaction questions [28]. To measure participant satisfaction, we asked two questions using a 5 item Likert scale (1Strongly agree to 5-Strongly disagree), with lower scores indicating higher satisfaction. The questions were “During this hospital visit, doctors answered all my questions to my satisfaction” and “During this hospital visit, doctors made sure I understood all the information I was given?”
Please cite this article in press as: K.B. Hadden, et al., Implementation of a health-literate patient decision aid for chest pain in the emergency department, Patient Educ Couns (2019), https://doi.org/10.1016/j.pec.2019.11.009
G Model PEC 6447 No. of Pages 6
K.B. Hadden et al. / Patient Education and Counseling xxx (2019) xxx–xxx
3
Fig. 1. Health-Literacy Based Patient Decision Aid for Chest Pain.
The Newest Vital Sign (NVS) [24]. Health literacy was included in analysis models as a characteristic/predictor to determine if outcomes varied by patient health literacy level. The Newest Vital Sign assesses literacy skills for both numbers and words, has been validated against a previously validated measure of health literacy (the TOFHLA) [25], and has been shown to take approximately three minutes to administer. NVS scores are reported in a range from 0 to 6, with scores of 0–1 indicating limited health literacy, 2– 3 indicating possible limited health literacy, and 4–6 indicating adequate health literacy. Following an expedited review and approval by the Institutional Review Board (IRB), each EM RA was trained to verbally administer a secure online (LimeSurvey) survey on an iPad for data collection. The RA scanned the intake board for patients with chest pain and once identified, approached the patient’s physician to determine and record eligibility. If the patient was eligible, the RA proceeded to the “condition assignment” question in the survey and marked whether the patient was in the usual care group/time period or, if the patient had received the decision tool, the intervention group/ time period. Next, the RA initiated the informed consent process,
obtained verbal consent from all participants, and administered the survey. The survey included the adapted Decisional Conflict Scale, knowledge questions, and satisfaction questions. Following data collection, medical record numbers recorded by a RA were used for a retrospective data extraction of all participants, including demographics (age, race, sex, Medicaid yes/no) and disposition (whether or not the patient was admitted for chest pain after ED visit). Data from extraction was combined with the data collected in an excel spreadsheet for importing into analysis software. 2.4. Outcomes and analysis Both usual care and intervention group data included demographics, knowledge scores, Decisional Conflict Scores, satisfaction scores, and NVS scores, as well as disposition (admitted for chest pain = yes/no). The study sample size was estimated to detect a 0.5 standard deviation difference between the study group means at the 0.05 significance level for the primary outcome “knowledge” with 0.83 power. Data were cleaned and coded in MS Excel and analyzed to characterize the sample using descriptive statistics and
Please cite this article in press as: K.B. Hadden, et al., Implementation of a health-literate patient decision aid for chest pain in the emergency department, Patient Educ Couns (2019), https://doi.org/10.1016/j.pec.2019.11.009
G Model PEC 6447 No. of Pages 6
4
K.B. Hadden et al. / Patient Education and Counseling xxx (2019) xxx–xxx
Table 1 Participant characteristics. Demographics Age Category Sex
Race
Admitted for Chest Pain Health Literacy
Average (Years) Response Male Female Total Caucasian African American Hispanic Other Total Yes No Total Inadequate Marginal Adequate Total
Usual Care (n = 78)
Intervention (n = 91)
Total (N = 169)
43.7 n 31 47 78 34 41 3 0 78 14 64 78 23 24 31 78
42.2 n 34 57 91 35 51 3 2 91 7 84 91 27 28 36 91
42.9 n 65 104 169 69 92 6 2 169 21 148 169 50 52 67 169
% 39.7% 60.3% 100.0% 43.6% 52.6% 3.8% 0.0% 100.0% 17.9% 82.1% 100.0% 29.5% 30.8% 39.7% 100.0%
SAS 9.0. Linear regression models were fit to explore the degree to which health literacy predicted the outcome variables: knowledge, decisional conflict, satisfaction, and disposition. Simple regression models were fit with study group as predictor terms, followed by adjusted models that included NVS. 3. Results A total of 169 participants that were assessed in the ED for chest pain and met inclusion criteria were surveyed and used in the final analysis. We surveyed 78 participants prior to introduction of the intervention to establish a usual care group; 91 participants were surveyed after a trained physician introduced the shared decision making tool. We used the NVS to measure participants’ health literacy; the total sample included 67 participants with adequate health literacy, 52 with marginal health literacy, and 50 with inadequate health literacy. There was no statistically significant difference in health literacy between the usual care and the intervention group (P = 0.9996). 3.1. Characteristics of study subjects Table 1 presents the demographics and characteristics of the study participants that were identified using a retrospective chart review. The average participant age of the entire sample was 42.9 years. Similar proportions of Caucasians, African Americans, and Hispanics were surveyed in the usual care group compared to the intervention group. A higher percentage of females were surveyed (61.5 %) compared to males (38.5%); although similar percentages were represented in both the usual care group compared to intervention. Of the total surveyed 60.4 % had either marginal or inadequate health literacy. 3.2. Main results Table 2 details the analyses results for the main outcomes of interest. For the knowledge outcome, the mean total knowledge score for the usual care group was 0.54, while the mean total knowledge score for the intervention group was 1.7 (P < 0.0001). Further, 60 % of patients in the usual care group had 0 correct on the knowledge measure, compared to only 13 % in the intervention group. There was no statistically significant difference between the usual care group and the intervention group for the Decisional Conflict mean (P = 0.4460) or Satisfaction mean scores (P = 0.7976). A test of fixed effects revealed a statistically significant difference in the Admit for Chest Pain outcome between the usual care and
% 37.4% 62.6% 100.0% 38.5% 56.0% 3.3% 2.2% 100.0% 7.7% 92.3% 100.0% 29.7% 30.8% 39.6% 100.0%
% 38.5% 61.5% 100.0% 40.8% 54.4% 3.6% 1.2% 100.0% 12.4% 87.6% 100.0% 29.6% 30.8% 39.6% 100.0%
the intervention group when adjusting for race, sex, and age (P < 0.0001). Patients in the usual care group were 2.6 times more likely to be admitted for chest pain than patients in the intervention group (OR = 2.625). There was a 10 percentage point difference in this admission outcome; approximately 18 % of patients in the usual care group were admitted for chest pain, while only approximately 8 % of patients in the intervention group were admitted. Additional analyses explored the effect of health literacy between groups and for the total sample, but no statistically significant differences were revealed for any of the outcomes of interest when health literacy was included as a predictor variable in regression models, or as an interaction variable. 3.3. Limitations This study has notable limitations. While the sample size was adequate to detect significant differences in primary outcome variable between the two comparison groups, a larger sample may have contributed to more robust findings related to differences in health literacy among the sample participants. Additionally, once these best practices and the patient-centered tool were introduced, we proposed that standard practice would be affected in an irreversible way, and it would not be feasible to expect providers to not use the enhanced communication elements of the intervention with patients who would be randomized to a control condition. Because this study was implemented, at least in part, as a demonstration of best practice in the Emergency Department, it was important to observe standard practice before intervening with new best practices. This quality improvement design relied on collecting baseline data for a time period before the intervention was launched for all providers in the ED. While a randomized design with blinding would have eliminated potential differences in outcomes related to time, and perhaps Hawhorne effect, it was not feasible for this quality improvement study.
Table 2 Effect of Health-Literate Decision Aid on outcomes of interest. Usual Care N = 78 Intervention N = 91 P value Knowledge Mean Decisional Conflict Mean Satisfaction Mean Admit for Chest Pain (yes) *
0.54 10.42 3.33 n = 14 (18%)
1.7 10.10 3.26 n = 7 (8%)
< 0.0001* 0.4460 0.7976 0.0439*
p < 0.05.
Please cite this article in press as: K.B. Hadden, et al., Implementation of a health-literate patient decision aid for chest pain in the emergency department, Patient Educ Couns (2019), https://doi.org/10.1016/j.pec.2019.11.009
G Model PEC 6447 No. of Pages 6
K.B. Hadden et al. / Patient Education and Counseling xxx (2019) xxx–xxx
4. Discussion and conclusion 4.1. Discussion Similar to prior investigations [29], patients in the usual care group were more likely to be admitted for chest pain and have lower knowledge scores compared to the intervention group in both unadjusted and adjusted analytical models when shared decision-making was used to inform patients of their disposition options. The use of shared decision-making to discuss patient disposition options could result in meaningful reductions of unnecessary testing, radiation exposure, and adverse events related to hospitalization. Additionally, although not measured in this trial, reductions in admissions for a common complaint like chest pain could improve ED throughput, mitigate ED boarding, and reduce time on pre-hospital diversion. A recent prospective randomized trial concluded that shared decision-making in patients with cardiovascular conditions improved decision-making quality and safely matched resource utilization to patient needs and preferences [29]. There were no statistically significant differences between the two groups for decisional conflict or satisfaction. In larger trials, decisional conflict was reduced in the populations where a decision aid was used [29]. In those trials, the populations studied had a substantially higher level of health literacy as compared to ours. Perhaps those with lower health literacy are equally satisfied with independent informed decision-making as they are with physician led determinations. It is also possible that our small sample size limited our ability to detect a true difference if one actually exists. Interestingly, patient satisfaction has not been shown to be positively affected by use of a decision aid in our trial and others [29]. While it would seem intuitive that better informed engagement in personal health care decisions would improve satisfaction, the data does not support this assumption. This is likely due to the complex combination of factors that influence patient satisfaction. As noted in a recent meta-analysis, it is not entirely clear what the major determinants of patient satisfaction are [30]. It should also be noted that using a decision-aid has not been shown to negatively impact satisfaction. The differences between the groups and for the total sample for the significant variables (admission and knowledge) were consistent across health literacy level; there was no interaction effect between group and NVS score. While a main intervention effect was demonstrated, we expected there to be differences either between the groups or within the total sample. Based on previous research [31] we hypothesized that patients with higher health literacy in the total sample would have higher knowledge scores and would be less likely to be admitted. This lack of significant difference could be attributed to the lack of sensitivity in our outcome measures; the knowledge measure was very specific to chest pain and not sensitive to general health knowledge that has been shown to be a contributor to health literacy in conceptual models [32]. Additionally, the small sample size and low percentage of admissions could have contributed to the lack of significant difference in the admission outcome. We also hypothesized that we would observe a difference in the outcomes of interest in the usual care group related to health literacy, but not in the intervention group, though our results did not reveal this scenario. Because we made deliberate efforts to address known health literacy barriers to understanding and using the chest pain decision tool in its adaptation, we anticipated that the health literacy-adapted tool would mitigate outcome disparities in the intervention group related to health literacy. However, when the analyses were repeated within each arm, and included health literacy level, no interaction was observed in either arm; the differences observed were consistent across all health literacy
5
levels. Even though health literacy was not a significant predictor of knowledge or other outcomes in either arm, the success of the intervention and the added value of ensuring patient understanding through health literacy best practices and field testing with patients are all important contributions of this study. Our project demonstrated that using health literacy best practices and decision support can address quality issues related to provider communication, patient knowledge, and admission rates for low risk patients in the ED. 4.2. Conclusion Using the adapted chest pain decision tool in emergency medicine may improve knowledge and reduce admissions, while addressing known barriers to understanding related to patient health literacy. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Declaration of Competing Interest None. References [1] U.S.Department of Health and HumanServices, Healthy People 2010 Available at:, (2000) . https://www.healthypeople.gov/sites/default/files/ HP2020_brochure_with_LHI_508_FNL.pdf. [2] M. Kutner, E. Greenberg, Y. Jin, C. Paulsen, S. White, Health Literacy of America’s Adults: Results From the 2003 National Assessment of Adult Literacy Available at:, (2006) . www.nces.gov. [3] A.L. Golbeck, C.R. Ahlers-Schmidt, A.M. Paschal, S.E. Dismuke, A definition and operational framework for health numeracy, Am. J. Prev. Med. 29 (4) (2005) 375–376. [4] Institute of Medicine, Health Literacy and Numeracy: Workshop Summary, (2014) . [5] R. Garcia-Retamero, E.T. Cokely, Communicating health risks with visual aids, Curr. Dir. Psychol. Sci. 22 (5) (2013) 392–399. [6] E. Garrison, C. Graves, R. Gregory, et al., Numeracy and literacy in pregant women with pregestational diabetes, Am. J. Obstet. Gynecol. 206 (1) (2012) S123. [7] H. Huang, Y. Chan, D. Feng, Health numeracy confidence among racial/ethnic minorities in hints 2007: sociodemographic, attitudinal, and knowledge correlates, Lit. Numer. Stud. 20 (2) (2012) 3–16. [8] M. Goodman, R. Finnegan, L. Mohadjer, T. Krenzke, J. Hogan, Literacy, Numeracy, and Problem Solving in Technology-rich Environments Among U.S. Adults: Results From the Program for the International Assessment of Adult Competencies 2012: First Look (NCES 2014-008), (2013) . [9] G. Elwyn, D. Frosch, R. Thomson, et al., Shared decision making: a model for clinical practice, J. Gen. Intern. Med. 27 (10) (2012) 1361–1367. [10] D.M. Muscat, H.L. Shepherd, S. Morony, et al., Can adults with low literacy understand shared decision making questions? A qualitative investigation, Patient Educ. Couns. (2016). [11] K.J. McCaffery, M. Holmes-Rovner, S.K. Smith, et al., Addressing health literacy in patient decision aids, BMC Med. Inform. Decis. Mak. 13 (Suppl 2) (2013) S10. [12] A. Delanoe, J. Lepine, M.E. Leiva Portocarrero, et al., Health literacy in pregnant women facing prenatal screening may explain their intention to use a patient decision aid: a short report, BMC Res. Notes 9 (2016) 339. [13] L.J. Malloy-Weir, C. Charles, A. Gafni, V.A. Entwistle, Empirical relationships between health literacy and treatment decision making: a scoping review of the literature, Patient Educ. Couns. 98 (3) (2015) 296–309. [14] Agency for Healthcare Research and Quality, Hospital guide to reducing medicaid readmissions, (Prepared by Collaborative Healthcare Strategies, Inc., and John Snow, Inc., Under Contract No. HHSA290201000034I). AHRQ Publication No. 14-0050-EF, Agency for Healthcare Research and Quality, Rockville, MD, 2014. [15] D.A. DeWalt, L.F. Callahan, V.H. Hawk, et al., Health Literacy Universal Precautions Toolkit, 1st ed., Agency for Healthcare Research and Quality, Rockville, MF, 2010. [16] H.K. Kanzaria, R.H. Brook, M.A. Probst, D. Harris, S.H. Berry, J.R. Hoffman, Emergency physician perceptions of shared decision-making, Acad. Emerg. Med. 22 (4) (2015) 399–405. [17] E.P. Hess, M.A. Knoedler, N.D. Shah, et al., The chest pain choice decision aid: a randomized trial, Circ. Cardiovasc. Qual. Outcomes 5 (3) (2012) 251–259.
Please cite this article in press as: K.B. Hadden, et al., Implementation of a health-literate patient decision aid for chest pain in the emergency department, Patient Educ Couns (2019), https://doi.org/10.1016/j.pec.2019.11.009
G Model PEC 6447 No. of Pages 6
6
K.B. Hadden et al. / Patient Education and Counseling xxx (2019) xxx–xxx
[18] D. Stacey, F. Legare, N.F. Col, et al., Decision aids for people facing health treatment or screening decisions, Cochrane Database Syst. Rev. 1 (2014)CD001431. [19] M.A. Pierce, E.P. Hess, J.A. Kline, et al., The chest pain choice trial: a pilot randomized trial of a decision aid for patients with chest pain in the emergency department, Trials 11 (2010) 57. [20] K. Hadden, The stoplight method: a qualitative approach for health literacy research, HLRP Health Lit. Res. Pract. 1 (2) (2017) e18–e22. [21] S.A. Mahler, R.F. Riley, B.C. Hiestand, et al., The HEART Pathway randomized trial: identifying emergency department patients with acute chest pain for early discharge, Circ. Cardiovasc. Qual. Outcomes 8 (2) (2015) 195–203. [22] J. Laureano-Phillips, R.D. Robinson, S. Aryal, et al., HEART score risk stratification of low-risk chest pain patients in the emergency department: a systematic review and meta-analysis, Ann. Emerg. Med. (2019). [23] B.E. Backus, A.J. Six, J.C. Kelder, et al., A prospective validation of the HEART score for chest pain patients at the emergency department, Int. J. Cardiol. 168 (3) (2013) 2153–2158. [24] Pfizer, The Newest Vital Sign Available at:, (2011) . http://www.pfizer.com/ files/health/2016_nvs_flipbook_english_final.pdf. [25] J.R. Nurss, R.M. Parker, M. Williams, D.W. Baker, Test of Functional Health Literacy in Adults (TOFHLA), (2001) .
[26] A.M. O'Connor, Decisional Conflict Scale - User Manual, The Ottawa Hospital Research Institute, 2010 Available at: https://decisionaid.ohri.ca/docs/ develop/User_Manuals/UM_Decisional_Conflict.pdf. Accessed April 5, 2017. [27] A.M. O’Connor, Validation of a decisional conflict scale, Med. Decis. Making 15 (1) (1995) 25–30. [28] Agency for Healthcare Research and Quality, CAHPS Health Literacy Item Sets, U S Department of Health and Human Services, 2012 Available at: https:// www.ahrq.gov/cahps/surveys-guidance/item-sets/literacy/index.html. Accessed January 4, 2017. [29] E.P. Hess, J.E. Hollander, J.T. Schaffer, et al., Shared decision making in patients with low risk chest pain: prospective randomized pragmatic trial, BMJ 355 (2016) i6165. [30] E. Batbaatar, J. Dorjdagva, A. Luvsannyam, M.M. Savino, P. Amenta, Determinants of patient satisfaction: a systematic review, Perspect. Public Health 137 (2) (2017) 89–101. [31] N.D. Berkman, S.L. Sheridan, K.E. Donahue, et al., Health literacy interventions and outcomes: an updated systematic review, Evid. Rep. Technol. Assess (Full Rep) 199 (2011) 1–941. [32] M.K. Paasche-Orlow, M.S. Wolf, The causal pathways linking health literacy to health outcomes, Am. J. Health Behav. 31 (Suppl 1) (2007) S19–S26.
Please cite this article in press as: K.B. Hadden, et al., Implementation of a health-literate patient decision aid for chest pain in the emergency department, Patient Educ Couns (2019), https://doi.org/10.1016/j.pec.2019.11.009
Patient Education and Counseling xxx (2019) xxx–xxx
Contents lists available at ScienceDirect
Patient Education and Counseling journal homepage: www.elsevier.com/locate/pateducou
Implementation of a health-literate patient decision aid for chest pain in the emergency department Kristie B. Haddena,* , Heather McLemoreb , Wesley Whiteb , Matthew H. Marksb , Jennifer M. Gana , Rawle A. Seupaulb a b
University of Arkansas for Medical Sciences, Center for Health Literacy, Little Rock, AR 72205-7199 USA University of Arkansas for Medical Sciences, Department of Emergency Medicine, Little Rock, AR 72205-7199 USA
A R T I C L E I N F O
A B S T R A C T
Article history: Received 4 September 2019 Received in revised form 22 October 2019 Accepted 11 November 2019
Objective: The aim of this study was to investigate the implementation of a new health-literacy-tested patient decision aid for chest pain in Emergency Department (ED) patients. Outcomes included disposition, knowledge, decisional conflict and satisfaction prior to discharge. Patient health literacy was explored as a factor that may explain disparities in sub-group analysis of all outcomes. Methods: A health-literacy adapted tool was deployed using a pre/post intervention design. Patients enrolled during the intervention period were given the adapted chest pain decision aid that was used in conversation with their emergency medicine physician to decide on their course of action prior to being discharged. Results: A total of 169 participants were surveyed and used in the final analysis. Patients in the usual care group were 2.6 times more likely to be admitted for chest pain than patients in the intervention group. Knowledge scores were higher in the intervention group, while no significant differences were observed in decisional conflict and patient satisfaction, or by patient health literacy level. Conclusion and practice implications: Using the adapted chest pain decision tool in emergency medicine may improve knowledge and reduce admissions, while addressing known barriers to understanding related to patient health literacy. © 2019 Elsevier B.V. All rights reserved.
Keywords: Health literacy Chest pain Decision aid Emergency department
1. Introduction Health literacy is “the degree to which individuals have the capacity to obtain, process and understand basic health information and services needed to make appropriate health decisions” [1]. Health numeracy, a component of overall health literacy [2], is “the degree to which individuals have the capacity to access, process, interpret, communicate and act on numerical, quantitative, graphical, biostatistical, and probabilistic health information needed to make effective health decisions” [3]. In accessing and navigating health care, the need for patients to use, understand, and work with numbers can be essential. Examples include choosing a health insurance plan, following medication dosing instructions, management of chronic diseases, and understanding risk and benefits for informed decision making [3–7]. Only 12 % of the American adult population has proficient health literacy, and individuals lack
* Corresponding author at: University of Arkansas for Medical Sciences, Center for Health Literacy, 4301 W. Markham Street, Slot 599A, Little Rock, AR 72205-7199 USA. E-mail addresses: [email protected] (K.B. Hadden), [email protected] (H. McLemore), [email protected] (W. White), [email protected] (M.H. Marks), [email protected] (J.M. Gan), [email protected] (R.A. Seupaul).
quantitative literacy skills even more than prose literacy [2,8]. Low health numeracy can be a barrier to shared decision making because many patients do not understand numerical concepts of risk and benefits [9], which can result in poor health decisions and increased costs [4]. Shared decision making involves collaborative input from both the patient and the health professional [10]. There are tools available to assist patients with making important health decisions; however, decision making tools are often difficult to understand and are rarely designed for patients with below average literacy skills [10]. Patients with low health literacy need more support to encourage participation in shared decision making [11]. Since health literacy affects the use of decision aids among patients when making difficult health decisions [12,13], it is recommended that decision aids are developed using health literacy best practices to facilitate understanding among all patients, particularly those with lower health literacy and numeracy skills [11,14]. Stakeholders across health care systems have recognized the important linkage between health literacy and health status, and are working to provide patient and consumer health materials, including decision aids, that are easy to understand and act upon, promote self- engagement in one’s own health, and lead to better health outcomes [15].
https://doi.org/10.1016/j.pec.2019.11.009 0738-3991/© 2019 Elsevier B.V. All rights reserved.
Please cite this article in press as: K.B. Hadden, et al., Implementation of a health-literate patient decision aid for chest pain in the emergency department, Patient Educ Couns (2019), https://doi.org/10.1016/j.pec.2019.11.009
G Model PEC 6447 No. of Pages 6
2
K.B. Hadden et al. / Patient Education and Counseling xxx (2019) xxx–xxx
1.1. Background and importance Patients who visit the emergency department (ED) may feel pressured to make quick decisions with little input from family or others whom they trust [16]. As a result, these patients typically rely on emergency medicine (EM) physicians to make critical decisions regarding their care or treatment options [16]. Research studies on use of shared decision making in emergency situations revealed that engaging patients in shared decision is helpful for reducing over-testing [16] and lowers health care costs and utilization [17,18]. Common obstacles associated with shared decision making in ED’s include: 1) lack of patient engagement in making important decisions regarding treatment options; 2) time constraints; and 3) lack of comfort or training in encouraging shared decision making [16]. Even low risk patients should have the option to decide between admitted to the hospital for observation or discharged with follow up. Since the risk of a major cardiac event is not 0 %, the patient has the option to be to be admitted for observation based on their informed tolerance of risk. Our leadership aimed to improve quality in our Emergency Department by 1) providing decision support for patients, 2) addressing patient health literacy and provider communication, and 3) reducing unnecessary admissions. To achieve these quality aims, we adapted an existing evidence-based shared decision making tool, created to educate and engage patients in decisions regarding their care. This “CHEST PAIN CHOICE” decision aid safely reduced admission rate for low risk study participants in a randomized control trial [17,19]. This shared decision making tool for low risk chest pain was evaluated for readability and health literacy/numeracy standards to provide a framework for the development of our refined tool. We used plain language best practices for all written portions of the refined tool. We field tested the refined tool with patients identified as having low health literacy using “stoplight coding” qualitative methods [20]. The adapted tool has a 3rd-4th grade readability, which is much lower than the previous tool with a 9th-10th grade readability. We also adapted the icon array using two distinct heart icons to reinforce the positive and negative risk frame which avoids precise percentages or more complex graphing that can be difficult for low numeracy or low graph literacy individuals. Finally, we used health literacy and plain language best practices and participant feedback to determine the adequacy of the tool for patients with low health literacy. 1.2. Goals of this investigation The specific aim of this study was to test the efficacy of a new health-literacy-tested patient decision aid for chest pain in a convenience sample of emergency department patients who used the decision aid compared to patients who were not exposed to the decision aid. Outcomes of interest include knowledge, decisional conflict and satisfaction prior to discharge. Health literacy was explored as a factor that may explain disparities in sub-group analysis of all outcomes between patients with inadequate, marginal, and adequate health literacy. Additionally, disposition outcomes were explored for the adequate/marginal/inadequate health literacy groups as well as the usual care/intervention arms. 2. Methods 2.1. Study design and procedures In the first phase of this project, the new decision aid was developed/adapted using health literacy standards and best practices for use in the Emergency Department (ED). The decision
aid was field-tested and changes were made to improve usability of the newly developed patient decision aid. This study then used a pre/ post intervention experimental design to test the impact of the newly created decision aid tool in clinical practice on outcomes of interest. Patients were recruited over a three-month period prior to intervention to establish the usual care group. Standard care was administered for eligible patients in this period. All ED providers received training on the content and delivery of the shared decisionmaking instrument prior to the fourteen-month intervention period. A training video using a standardized patient was developed to illustrate the practical application of the tool in a simulated clinical setting. In person sessions were also held with all ED providers to answer questions and/or provide clarifications on the adapted tool’s content and method of delivery. Integration of the adapted tool into the clinic workflow was well received by all ED providers. The UAMS emergency department is an urban, university based academic ED with a residency training program, with an annual census of 60,000 patients/year. Patients enrolled during the intervention period were given the chest pain decision aid (Fig. 1) that was used in conversation with their EM physician to decide on their course of action prior to being discharged from the ED. EM physicians asked patients to explain in their own words their risk and options prior to ending the encounter. Data collection procedures were the same for both arms of the study and completed prior to discharge by interview with an EM research assistant (RA). 2.2. Selection of participants Patients who presented with chest pain and were categorized as “low risk” for cardiac events in the ED during weekday, evening, and weekend recruitment periods were eligible to participate. For this study, “low risk” was defined as a HEART score of 3 or less [21,22]. Patients who presented with chest pain and were categorized as “low risk” based on a HEART score of 0–3 with unremarkable troponin were recruited [22,23]. Additional inclusion criteria included twenty-one years old or older and ability to see, hear, speak and understand spoken and written English. Patients who were unconscious or at moderate to high risk for cardiac events or other complications were excluded from participation. 2.3. Measurements Adapted knowledge questions [27] were used to assess the primary outcome of patient knowledge about key information presented in the decision support. A four item survey (3 True/False questions and 1 open response) was administered to assess participant knowledge about chest pain follow up care and likelihood of having a heart attack based on risk factors. These questions were adapted from the 7 item knowledge questions used in a previous trial [17]. Adapted Decisional Conflict Scale [26,27]. The Decisional Conflict Scale measures personal perceptions of uncertainty in choosing options, modifiable factors contributing to uncertainty such as feeling uninformed, unclear about personal values and unsupported in decision making; and effective decision making (in full version) such as feeling the choice is informed, values-based, likely to be implemented and expressing satisfaction with the choice [26]. Adapted questions from the Decisional Conflict Scale [26] were used for this study using a 5 item Likert (1-Strongly agree to 5-Strongly disagree). Adapted satisfaction questions [28]. To measure participant satisfaction, we asked two questions using a 5 item Likert scale (1Strongly agree to 5-Strongly disagree), with lower scores indicating higher satisfaction. The questions were “During this hospital visit, doctors answered all my questions to my satisfaction” and “During this hospital visit, doctors made sure I understood all the information I was given?”
Please cite this article in press as: K.B. Hadden, et al., Implementation of a health-literate patient decision aid for chest pain in the emergency department, Patient Educ Couns (2019), https://doi.org/10.1016/j.pec.2019.11.009
G Model PEC 6447 No. of Pages 6
K.B. Hadden et al. / Patient Education and Counseling xxx (2019) xxx–xxx
3
Fig. 1. Health-Literacy Based Patient Decision Aid for Chest Pain.
The Newest Vital Sign (NVS) [24]. Health literacy was included in analysis models as a characteristic/predictor to determine if outcomes varied by patient health literacy level. The Newest Vital Sign assesses literacy skills for both numbers and words, has been validated against a previously validated measure of health literacy (the TOFHLA) [25], and has been shown to take approximately three minutes to administer. NVS scores are reported in a range from 0 to 6, with scores of 0–1 indicating limited health literacy, 2– 3 indicating possible limited health literacy, and 4–6 indicating adequate health literacy. Following an expedited review and approval by the Institutional Review Board (IRB), each EM RA was trained to verbally administer a secure online (LimeSurvey) survey on an iPad for data collection. The RA scanned the intake board for patients with chest pain and once identified, approached the patient’s physician to determine and record eligibility. If the patient was eligible, the RA proceeded to the “condition assignment” question in the survey and marked whether the patient was in the usual care group/time period or, if the patient had received the decision tool, the intervention group/ time period. Next, the RA initiated the informed consent process,
obtained verbal consent from all participants, and administered the survey. The survey included the adapted Decisional Conflict Scale, knowledge questions, and satisfaction questions. Following data collection, medical record numbers recorded by a RA were used for a retrospective data extraction of all participants, including demographics (age, race, sex, Medicaid yes/no) and disposition (whether or not the patient was admitted for chest pain after ED visit). Data from extraction was combined with the data collected in an excel spreadsheet for importing into analysis software. 2.4. Outcomes and analysis Both usual care and intervention group data included demographics, knowledge scores, Decisional Conflict Scores, satisfaction scores, and NVS scores, as well as disposition (admitted for chest pain = yes/no). The study sample size was estimated to detect a 0.5 standard deviation difference between the study group means at the 0.05 significance level for the primary outcome “knowledge” with 0.83 power. Data were cleaned and coded in MS Excel and analyzed to characterize the sample using descriptive statistics and
Please cite this article in press as: K.B. Hadden, et al., Implementation of a health-literate patient decision aid for chest pain in the emergency department, Patient Educ Couns (2019), https://doi.org/10.1016/j.pec.2019.11.009
G Model PEC 6447 No. of Pages 6
4
K.B. Hadden et al. / Patient Education and Counseling xxx (2019) xxx–xxx
Table 1 Participant characteristics. Demographics Age Category Sex
Race
Admitted for Chest Pain Health Literacy
Average (Years) Response Male Female Total Caucasian African American Hispanic Other Total Yes No Total Inadequate Marginal Adequate Total
Usual Care (n = 78)
Intervention (n = 91)
Total (N = 169)
43.7 n 31 47 78 34 41 3 0 78 14 64 78 23 24 31 78
42.2 n 34 57 91 35 51 3 2 91 7 84 91 27 28 36 91
42.9 n 65 104 169 69 92 6 2 169 21 148 169 50 52 67 169
% 39.7% 60.3% 100.0% 43.6% 52.6% 3.8% 0.0% 100.0% 17.9% 82.1% 100.0% 29.5% 30.8% 39.7% 100.0%
SAS 9.0. Linear regression models were fit to explore the degree to which health literacy predicted the outcome variables: knowledge, decisional conflict, satisfaction, and disposition. Simple regression models were fit with study group as predictor terms, followed by adjusted models that included NVS. 3. Results A total of 169 participants that were assessed in the ED for chest pain and met inclusion criteria were surveyed and used in the final analysis. We surveyed 78 participants prior to introduction of the intervention to establish a usual care group; 91 participants were surveyed after a trained physician introduced the shared decision making tool. We used the NVS to measure participants’ health literacy; the total sample included 67 participants with adequate health literacy, 52 with marginal health literacy, and 50 with inadequate health literacy. There was no statistically significant difference in health literacy between the usual care and the intervention group (P = 0.9996). 3.1. Characteristics of study subjects Table 1 presents the demographics and characteristics of the study participants that were identified using a retrospective chart review. The average participant age of the entire sample was 42.9 years. Similar proportions of Caucasians, African Americans, and Hispanics were surveyed in the usual care group compared to the intervention group. A higher percentage of females were surveyed (61.5 %) compared to males (38.5%); although similar percentages were represented in both the usual care group compared to intervention. Of the total surveyed 60.4 % had either marginal or inadequate health literacy. 3.2. Main results Table 2 details the analyses results for the main outcomes of interest. For the knowledge outcome, the mean total knowledge score for the usual care group was 0.54, while the mean total knowledge score for the intervention group was 1.7 (P < 0.0001). Further, 60 % of patients in the usual care group had 0 correct on the knowledge measure, compared to only 13 % in the intervention group. There was no statistically significant difference between the usual care group and the intervention group for the Decisional Conflict mean (P = 0.4460) or Satisfaction mean scores (P = 0.7976). A test of fixed effects revealed a statistically significant difference in the Admit for Chest Pain outcome between the usual care and
% 37.4% 62.6% 100.0% 38.5% 56.0% 3.3% 2.2% 100.0% 7.7% 92.3% 100.0% 29.7% 30.8% 39.6% 100.0%
% 38.5% 61.5% 100.0% 40.8% 54.4% 3.6% 1.2% 100.0% 12.4% 87.6% 100.0% 29.6% 30.8% 39.6% 100.0%
the intervention group when adjusting for race, sex, and age (P < 0.0001). Patients in the usual care group were 2.6 times more likely to be admitted for chest pain than patients in the intervention group (OR = 2.625). There was a 10 percentage point difference in this admission outcome; approximately 18 % of patients in the usual care group were admitted for chest pain, while only approximately 8 % of patients in the intervention group were admitted. Additional analyses explored the effect of health literacy between groups and for the total sample, but no statistically significant differences were revealed for any of the outcomes of interest when health literacy was included as a predictor variable in regression models, or as an interaction variable. 3.3. Limitations This study has notable limitations. While the sample size was adequate to detect significant differences in primary outcome variable between the two comparison groups, a larger sample may have contributed to more robust findings related to differences in health literacy among the sample participants. Additionally, once these best practices and the patient-centered tool were introduced, we proposed that standard practice would be affected in an irreversible way, and it would not be feasible to expect providers to not use the enhanced communication elements of the intervention with patients who would be randomized to a control condition. Because this study was implemented, at least in part, as a demonstration of best practice in the Emergency Department, it was important to observe standard practice before intervening with new best practices. This quality improvement design relied on collecting baseline data for a time period before the intervention was launched for all providers in the ED. While a randomized design with blinding would have eliminated potential differences in outcomes related to time, and perhaps Hawhorne effect, it was not feasible for this quality improvement study.
Table 2 Effect of Health-Literate Decision Aid on outcomes of interest. Usual Care N = 78 Intervention N = 91 P value Knowledge Mean Decisional Conflict Mean Satisfaction Mean Admit for Chest Pain (yes) *
0.54 10.42 3.33 n = 14 (18%)
1.7 10.10 3.26 n = 7 (8%)
< 0.0001* 0.4460 0.7976 0.0439*
p < 0.05.
Please cite this article in press as: K.B. Hadden, et al., Implementation of a health-literate patient decision aid for chest pain in the emergency department, Patient Educ Couns (2019), https://doi.org/10.1016/j.pec.2019.11.009
G Model PEC 6447 No. of Pages 6
K.B. Hadden et al. / Patient Education and Counseling xxx (2019) xxx–xxx
4. Discussion and conclusion 4.1. Discussion Similar to prior investigations [29], patients in the usual care group were more likely to be admitted for chest pain and have lower knowledge scores compared to the intervention group in both unadjusted and adjusted analytical models when shared decision-making was used to inform patients of their disposition options. The use of shared decision-making to discuss patient disposition options could result in meaningful reductions of unnecessary testing, radiation exposure, and adverse events related to hospitalization. Additionally, although not measured in this trial, reductions in admissions for a common complaint like chest pain could improve ED throughput, mitigate ED boarding, and reduce time on pre-hospital diversion. A recent prospective randomized trial concluded that shared decision-making in patients with cardiovascular conditions improved decision-making quality and safely matched resource utilization to patient needs and preferences [29]. There were no statistically significant differences between the two groups for decisional conflict or satisfaction. In larger trials, decisional conflict was reduced in the populations where a decision aid was used [29]. In those trials, the populations studied had a substantially higher level of health literacy as compared to ours. Perhaps those with lower health literacy are equally satisfied with independent informed decision-making as they are with physician led determinations. It is also possible that our small sample size limited our ability to detect a true difference if one actually exists. Interestingly, patient satisfaction has not been shown to be positively affected by use of a decision aid in our trial and others [29]. While it would seem intuitive that better informed engagement in personal health care decisions would improve satisfaction, the data does not support this assumption. This is likely due to the complex combination of factors that influence patient satisfaction. As noted in a recent meta-analysis, it is not entirely clear what the major determinants of patient satisfaction are [30]. It should also be noted that using a decision-aid has not been shown to negatively impact satisfaction. The differences between the groups and for the total sample for the significant variables (admission and knowledge) were consistent across health literacy level; there was no interaction effect between group and NVS score. While a main intervention effect was demonstrated, we expected there to be differences either between the groups or within the total sample. Based on previous research [31] we hypothesized that patients with higher health literacy in the total sample would have higher knowledge scores and would be less likely to be admitted. This lack of significant difference could be attributed to the lack of sensitivity in our outcome measures; the knowledge measure was very specific to chest pain and not sensitive to general health knowledge that has been shown to be a contributor to health literacy in conceptual models [32]. Additionally, the small sample size and low percentage of admissions could have contributed to the lack of significant difference in the admission outcome. We also hypothesized that we would observe a difference in the outcomes of interest in the usual care group related to health literacy, but not in the intervention group, though our results did not reveal this scenario. Because we made deliberate efforts to address known health literacy barriers to understanding and using the chest pain decision tool in its adaptation, we anticipated that the health literacy-adapted tool would mitigate outcome disparities in the intervention group related to health literacy. However, when the analyses were repeated within each arm, and included health literacy level, no interaction was observed in either arm; the differences observed were consistent across all health literacy
5
levels. Even though health literacy was not a significant predictor of knowledge or other outcomes in either arm, the success of the intervention and the added value of ensuring patient understanding through health literacy best practices and field testing with patients are all important contributions of this study. Our project demonstrated that using health literacy best practices and decision support can address quality issues related to provider communication, patient knowledge, and admission rates for low risk patients in the ED. 4.2. Conclusion Using the adapted chest pain decision tool in emergency medicine may improve knowledge and reduce admissions, while addressing known barriers to understanding related to patient health literacy. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Declaration of Competing Interest None. References [1] U.S.Department of Health and HumanServices, Healthy People 2010 Available at:, (2000) . https://www.healthypeople.gov/sites/default/files/ HP2020_brochure_with_LHI_508_FNL.pdf. [2] M. Kutner, E. Greenberg, Y. Jin, C. Paulsen, S. White, Health Literacy of America’s Adults: Results From the 2003 National Assessment of Adult Literacy Available at:, (2006) . www.nces.gov. [3] A.L. Golbeck, C.R. Ahlers-Schmidt, A.M. Paschal, S.E. Dismuke, A definition and operational framework for health numeracy, Am. J. Prev. Med. 29 (4) (2005) 375–376. [4] Institute of Medicine, Health Literacy and Numeracy: Workshop Summary, (2014) . [5] R. Garcia-Retamero, E.T. Cokely, Communicating health risks with visual aids, Curr. Dir. Psychol. Sci. 22 (5) (2013) 392–399. [6] E. Garrison, C. Graves, R. Gregory, et al., Numeracy and literacy in pregant women with pregestational diabetes, Am. J. Obstet. Gynecol. 206 (1) (2012) S123. [7] H. Huang, Y. Chan, D. Feng, Health numeracy confidence among racial/ethnic minorities in hints 2007: sociodemographic, attitudinal, and knowledge correlates, Lit. Numer. Stud. 20 (2) (2012) 3–16. [8] M. Goodman, R. Finnegan, L. Mohadjer, T. Krenzke, J. Hogan, Literacy, Numeracy, and Problem Solving in Technology-rich Environments Among U.S. Adults: Results From the Program for the International Assessment of Adult Competencies 2012: First Look (NCES 2014-008), (2013) . [9] G. Elwyn, D. Frosch, R. Thomson, et al., Shared decision making: a model for clinical practice, J. Gen. Intern. Med. 27 (10) (2012) 1361–1367. [10] D.M. Muscat, H.L. Shepherd, S. Morony, et al., Can adults with low literacy understand shared decision making questions? A qualitative investigation, Patient Educ. Couns. (2016). [11] K.J. McCaffery, M. Holmes-Rovner, S.K. Smith, et al., Addressing health literacy in patient decision aids, BMC Med. Inform. Decis. Mak. 13 (Suppl 2) (2013) S10. [12] A. Delanoe, J. Lepine, M.E. Leiva Portocarrero, et al., Health literacy in pregnant women facing prenatal screening may explain their intention to use a patient decision aid: a short report, BMC Res. Notes 9 (2016) 339. [13] L.J. Malloy-Weir, C. Charles, A. Gafni, V.A. Entwistle, Empirical relationships between health literacy and treatment decision making: a scoping review of the literature, Patient Educ. Couns. 98 (3) (2015) 296–309. [14] Agency for Healthcare Research and Quality, Hospital guide to reducing medicaid readmissions, (Prepared by Collaborative Healthcare Strategies, Inc., and John Snow, Inc., Under Contract No. HHSA290201000034I). AHRQ Publication No. 14-0050-EF, Agency for Healthcare Research and Quality, Rockville, MD, 2014. [15] D.A. DeWalt, L.F. Callahan, V.H. Hawk, et al., Health Literacy Universal Precautions Toolkit, 1st ed., Agency for Healthcare Research and Quality, Rockville, MF, 2010. [16] H.K. Kanzaria, R.H. Brook, M.A. Probst, D. Harris, S.H. Berry, J.R. Hoffman, Emergency physician perceptions of shared decision-making, Acad. Emerg. Med. 22 (4) (2015) 399–405. [17] E.P. Hess, M.A. Knoedler, N.D. Shah, et al., The chest pain choice decision aid: a randomized trial, Circ. Cardiovasc. Qual. Outcomes 5 (3) (2012) 251–259.
Please cite this article in press as: K.B. Hadden, et al., Implementation of a health-literate patient decision aid for chest pain in the emergency department, Patient Educ Couns (2019), https://doi.org/10.1016/j.pec.2019.11.009
G Model PEC 6447 No. of Pages 6
6
K.B. Hadden et al. / Patient Education and Counseling xxx (2019) xxx–xxx
[18] D. Stacey, F. Legare, N.F. Col, et al., Decision aids for people facing health treatment or screening decisions, Cochrane Database Syst. Rev. 1 (2014)CD001431. [19] M.A. Pierce, E.P. Hess, J.A. Kline, et al., The chest pain choice trial: a pilot randomized trial of a decision aid for patients with chest pain in the emergency department, Trials 11 (2010) 57. [20] K. Hadden, The stoplight method: a qualitative approach for health literacy research, HLRP Health Lit. Res. Pract. 1 (2) (2017) e18–e22. [21] S.A. Mahler, R.F. Riley, B.C. Hiestand, et al., The HEART Pathway randomized trial: identifying emergency department patients with acute chest pain for early discharge, Circ. Cardiovasc. Qual. Outcomes 8 (2) (2015) 195–203. [22] J. Laureano-Phillips, R.D. Robinson, S. Aryal, et al., HEART score risk stratification of low-risk chest pain patients in the emergency department: a systematic review and meta-analysis, Ann. Emerg. Med. (2019). [23] B.E. Backus, A.J. Six, J.C. Kelder, et al., A prospective validation of the HEART score for chest pain patients at the emergency department, Int. J. Cardiol. 168 (3) (2013) 2153–2158. [24] Pfizer, The Newest Vital Sign Available at:, (2011) . http://www.pfizer.com/ files/health/2016_nvs_flipbook_english_final.pdf. [25] J.R. Nurss, R.M. Parker, M. Williams, D.W. Baker, Test of Functional Health Literacy in Adults (TOFHLA), (2001) .
[26] A.M. O'Connor, Decisional Conflict Scale - User Manual, The Ottawa Hospital Research Institute, 2010 Available at: https://decisionaid.ohri.ca/docs/ develop/User_Manuals/UM_Decisional_Conflict.pdf. Accessed April 5, 2017. [27] A.M. O’Connor, Validation of a decisional conflict scale, Med. Decis. Making 15 (1) (1995) 25–30. [28] Agency for Healthcare Research and Quality, CAHPS Health Literacy Item Sets, U S Department of Health and Human Services, 2012 Available at: https:// www.ahrq.gov/cahps/surveys-guidance/item-sets/literacy/index.html. Accessed January 4, 2017. [29] E.P. Hess, J.E. Hollander, J.T. Schaffer, et al., Shared decision making in patients with low risk chest pain: prospective randomized pragmatic trial, BMJ 355 (2016) i6165. [30] E. Batbaatar, J. Dorjdagva, A. Luvsannyam, M.M. Savino, P. Amenta, Determinants of patient satisfaction: a systematic review, Perspect. Public Health 137 (2) (2017) 89–101. [31] N.D. Berkman, S.L. Sheridan, K.E. Donahue, et al., Health literacy interventions and outcomes: an updated systematic review, Evid. Rep. Technol. Assess (Full Rep) 199 (2011) 1–941. [32] M.K. Paasche-Orlow, M.S. Wolf, The causal pathways linking health literacy to health outcomes, Am. J. Health Behav. 31 (Suppl 1) (2007) S19–S26.
Please cite this article in press as: K.B. Hadden, et al., Implementation of a health-literate patient decision aid for chest pain in the emergency department, Patient Educ Couns (2019), https://doi.org/10.1016/j.pec.2019.11.009