Improving Patient-Provider Communication for Patients Having Surgery: Patient Perceptions of a Revised Health LiteracyYBased Consent Process

Transcription

1 ORIGINAL ARTICLE Improving Patient-Provider Communication for Patients Having Surgery: Patient Perceptions of a Revised Health LiteracyYBased Consent Process Michael J. Miller, RPh, DrPH,* Mary Ann Abrams, MD, MPH,Þ Barb Earles, RN, MHA, CPHRM,Þ Kirk Phillips, PhD,Þ and Erin M. McCleeary, PharmDþ Objectives: This research sought to describe and compare perceptions of consent-related health communication between surgical patients undergoing procedures at facilities that did and did not adopt a new health literacyybased consent form and process. Methods: A self-administered, mail survey was used to collect information about demographic characteristics, health locus of control, and perceptions of surgical consent-related health communication from patients aged 18 years or older, approximately 2 to 4 months after undergoing laparascopic cholecystectomy, total hip replacement, or total knee replacement surgery within a 10-hospital integrated health system in Iowa. A static group comparison design with multivariable logistic regression analyses was used to compare perceptions about 12 aspects of surgical consent-related health communication between the adopting and nonadopting facilities while controlling for observed differences in respondent background characteristics using a threshold of P G 0.05 for model inclusion. Results: Respondents from facilities implementing the new consent form and process had significantly higher odds of strongly agreeing that the nurses asked them to restate the type of surgery being performed in their own words (adjusted odds ratio, 1.92; 95% confidence interval, 1.30Y2.82) and they were comfortable asking questions about their surgery (adjusted odds ratio, 1.53; 95% confidence interval, 1.04Y2.26). Conclusions: The consent process can be refined to stimulate communication and comfort with asking questions, and promote use of health literacyybased techniques (i.e., teach-back) in the perioperative care setting. Adopting a health literacyybased informed consent process promotes patient safety and supports health providers obligations to communicate in simple, clear, and plain language. Key Words: health literacy, informed consent, quality improvement, patient safety, risk management, teach-back (J Patient Saf 2011;7: 30Y38) The informed consent process serves as a safeguard for surgical patients to make informed, autonomous choices about a proposed procedure while understanding its associated benefits, risks, and treatment alternatives. 1 Therefore, it is critical to deploy the consent process in a way that optimizes patient From the *College of Pharmacy, The University of Oklahoma, Tulsa, Oklahoma; Iowa Health System, Des Moines, Iowa; and Avera McKennan Hospital & University Health Center, Sioux Falls, South Dakota. Correspondence: Michael J. Miller, RPh, DrPH, College of PharmacyYTulsa, The University of Oklahoma, 4502 East 41st St, Tulsa, OK ( michael-miller@ouhsc.edu). Grant Support: This study was supported, in part, by the Community Responsive Grant Program through the Wellmark Foundation, Des Moines, Iowa. Conflict of Interest: None. Copyright * 2011 by Lippincott Williams & Wilkins understanding across the continuum from preoperative to postoperative care. Unfortunately, a systematic review revealed that a majority of published research studies on surgical informed consent show inadequate-to-moderate understanding of information given to patients, including the risks and benefits of the operation. 2 Implementing the informed consent process poses a challenge for health care systems and providers as they seek to balance liability concerns with provision of the adequate amount of information necessary for autonomous decision-making. Often, a significant focus is placed on obtaining a signature on a consent form. However, dependence on the consent form alone may contribute to suboptimal patient understanding as research has demonstrated that 4 basic elements of informed consent (i.e., nature of procedure, risks, benefits, and alternatives) were included on consent forms only 26.4% of the time. 3 Moreover, many factors demonstrating shared patient-physician decision making were included on only a minority of consent forms with the exception of a signature line for a patient to acknowledge understanding. 3 Although signing a consent form is essential for documenting that an informed consent discussion took place, it is a discrete event that may overshadow the need for ongoing dialogue and shared decision making between patient and health care provider. 4 A 2010 systematic review of 44 studies concluded that patient comprehension of informed consent for medical and surgical procedures may be improved using additional written information, audiovisual/multimedia interventions, extended discussions, and test/feedback techniques. 5 Seven of the 44 studies reviewed specifically included knee or hip arthroplasty, anterior cruciate ligament reconstruction, or cholecystectomy procedures relevant to the current study. In 6 of those studies, information leaflets/booklets (n = 5) or an educational video (n = 1) with or without verbal counseling were commonly used, whereas 1 study used an iterative teach-back process. Subsequent to each of these interventions, comprehension of risks, benefits, and/or general knowledge was improved. Until recently, there has been limited rigorous evaluation of use of the teach-back method in surgical informed consent reported in the literature. Findings from a randomized controlled trial comparing the standard consent process plus the repeatback (i.e., teach-back) method to the standard consent process alone in 4 common surgical procedures (i.e., carotid endarterectomy, laparascopic cholecystectomy, radical prostatectomy, and total hip arthroplasty) resulted in modestly increased patient comprehension while not affecting patient anxiety and satisfaction with care. 6 In 2003, the National Quality Forum (NQF) endorsed Ask(ing) each patient or legal surrogate to recount what he or she has been told during the informed consent discussion as one of 30 practices that should be used universally to reduce the risk of harm resulting from processes, systems, or environments of health care. 7 The NQF also recommended improving the quality 30 J Patient Saf & Volume 7, Number 1, March 2011

2 J Patient Saf & Volume 7, Number 1, March 2011 Patient Perceptions of a Revised Health Literacy-Based Consent Process of informed consent documents to ensure they are Ireaderfriendly, simple, and useful to patients, particularly those with limited health literacy, while also educating providers about the central role of verbal discussion in the informed consent process. 8 Use of the teach-back method in the consent process remains one of the 34 best practices as of Although the teach-back method is endorsed as an effective strategy to enhance patient comprehension during the surgical consent process, there is a dearth of research documenting its adoption and effective implementation in routine practice under nonresearch conditions. As part of a cross-cutting, system-wide quality initiative to address health literacy that began in 2003, the Iowa Health System (IHS) targeted improvement of the informed consent process and documents as important priorities. The goal was to create a reader-friendly document that would stimulate dialogue and enhance patients comfort asking questions about their surgical procedure while integrating the use of teach-back into the informed consent process and not adding additional complexity to the perioperative care environment. Ultimately, the document would confirm that the informed consent discussion had occurred, the elements of informed consent were addressed, and the patient had a chance to ask questions and describe the procedure in their own words. The strategy undertaken by IHS to develop a reader-friendly document and enhanced informed consent process for surgery has been previously described 10,11 and recognized. 12 In brief, an iterative process was used to develop the reader-friendly document in collaboration with IHS health literacy teams, adult learners, risk managers, health care providers, and the IHS Law Department. The document incorporated the key elements of informed consent 1 and triggers the use of the health literacyy based teach-back technique 13 by providing space to record patients description of their procedure in their own words. Beginning in 2004, the new document was introduced on a voluntary basis at IHS affiliate hospitals, with pilot-testing, staff training, and physician involvement that led to minor revisions. (See Appendix 1.) This voluntary implementation provided the opportunity to describe and compare surgical patient perceptions of consent-related, perioperative communication at facilities where the new document and process were adopted to those where they were not yet implemented. METHODS Study Design A self-administered, anonymous mail questionnaire was used to collect information related to patient background, health locus of control (HLOC), and perceptions of health communication during the perioperative period among communitydwelling adults 18 years or older. To improve response rate, a follow-up questionnaire was sent to persons not responding within 1 month of the initial mailing. A static group comparison design 14 was used to evaluate differences in perceptions of health communication between patients at facilities that had adopted the new consent form and process and those that had not. This design was used to capture the experience of patients in a setting that reflects deployment of an intervention in a large health system with semi-autonomous affiliate hospitals that includes early and late program adopters. 15 The study was approved by institutional review boards within IHS and its affiliates, as well as at Drake University. Study Setting All 10 IHS senior affiliate hospitals, ranging from 38 to 584 beds in 7 population centers, were used for recruitment to represent a broad range of patient and facility characteristics across rural and urban settings. At the time of this study, 6 hospitals had adopted the new consent process, 1 was transitioning to the new consent process, and 3 had not adopted the new process. Participant Recruitment and Selection Eligible patients were those patients who had undergone one of 3 of the most common non-pregnancy-related surgical procedures (i.e., laparascopic cholecystectomy, total hip replacement, or total knee replacement) at any of the 10 IHS affiliate hospitals and were discharged during September and October Patients were identified by any of the following International Classification of Diseases, Ninth Revision, Clinical Modification, procedure codes: 51.23, 81.51, or through electronic billing records. Patients included were admitted through physician referral or the emergency department and discharged to home, a skilled nursing facility, home health care, or a Medicare swing bed or rehabilitation unit within the facility where they had surgery. Initial survey distribution was accomplished approximately 2 to 4 months after discharge to ensure a complete sampling frame was available in the database, maximize the likelihood that patients were physically at their address of record, and avoid contacting individuals who may have died during or shortly after their surgical hospitalization. Assuming that 50% of patients at facilities using the original consent form and process would report strong agreement with each of the statements about consent-related, perioperative health communication during their recent surgical stay, and at least 70% of patients would report strong agreement at facilities adopting the new consent form and process, approximately 206 subjects (103 per study arm) were needed to detect a statistically significant difference with an > = 0.05 and power = A minimum sample of 400 patients (200 each from intervention and control hospitals) was sought over a 2-month recruitment period to ensure a sample with adequate power assuming a 50% response rate. All eligible patients were recruited for a complete 2-month period to ensure a sufficient sample. Data Collection and Measurement The first section of each questionnaire included questions eliciting demographic information to describe the respondent sample and included age, sex, race, education level, and selfreported health status. The second section collected information about respondents multidimensional HLOC. 16,17 The HLOC is the generalized belief about one s ability to control events, which may act as a barrier to adequate patient-provider communication. The HLOC includes 3 domains (internal, external or powerful others, and chance), each comprised of 6 belief statements. A 6-level response set is used to allow respondents to strongly disagree, moderately disagree, slightly disagree, slightly agree, moderately agree, or strongly agree with each statement. Minor changes were made to the standard instructions for completing HLOC questions to ensure clarity and plain language. The third section of each questionnaire included statements related to respondent perceptions of consent-related perioperative communication. In the absence of a previously validated data questionnaire related to perceptions of the consent-related perioperative communication process, 13 statements were developed by a multidisciplinary group of health care providers and health services researchers using an iterative process. Statements represented a wide range of content, including issues related to patient-perceived understanding of provider (i.e., physician and nurse) communication, provider listening skills, opportunities to ask questions and restate (teach back) provider-communicated * 2011 Lippincott Williams & Wilkins 31

3 Miller et al J Patient Saf & Volume 7, Number 1, March 2011 information, understanding of surgical procedure risks and benefits, comfort with asking questions, consent form readability, and general feeling of being informed about the surgical procedure. A 6-level response set similar to that for the HLOC statements was used to allow respondents to strongly disagree, moderately disagree, slightly disagree, slightly agree, moderately agree, or strongly agree with each statement. Consistent with top box analysis used in previous surgery-relevant quality-of-life research 18 and patient satisfaction surveys reported by Press Ganey Associates, Inc. ( the response set was dichotomized to compare the percentage of participants responding strongly agree to the percentage with any other response. The first and last statements relating to respondents sense of being generally informed about the surgical procedure were similar with the exception that the last statement was negatively-worded to assess response acquiescence. After determining there was no response acquiescence, the negatively-worded question was dropped to simplify interpretation and avoid redundancy with the first statement, leaving a set of 12 questions for bivariate and multivariable analyses. Data Management and Analysis All data were collected and entered into a secure database. Data were exported to Stata 10.1 statistical software for analysis. 19 Univariate descriptive statistics were computed and reported for all measures collected. The proportion of respondents reporting overall as well as strong agreement with each of the 12 statements related to consent-related perioperative health communication between facilities that adopted the revised consent form and process and those that did not were compared in bivariate analyses. Chi-square analyses were used to assess the bivariate relationships. Background variables significantly associated (P G 0.05) with participation in the new consent process were included in multivariable analyses as control variables. Multivariable logistic regression was used to evaluate the relationships between respondent strong agreement with various aspects of the consent process and participation in the new consent process while accounting for significant observed differences in respondent background demographic characteristics. An a priori > level of 0.05 was used as the criterion for statistical significance for all analyses. RESULTS Participants/Response Rate Questionnaires were mailed to 1100 patients. Fourteen patients were not eligible because they either reported no surgery or an unrelated surgery (n = 13) or had died (n = 1) and were removed from the denominator for the eligible sample. A total of 568 of the 595 questionnaires returned were usable, yielding a final response rate of 52.3% (568/1086). Response rate varied slightly by affiliate hospitals that adopted the new consent process (47.2% [233/494]), retained the original consent process (55.9% [312/558]), or were in transition to the new process (67.6% [23/34]). Questionnaire responses from the one affiliate hospital that was transitioning were dropped from all analyses because they could not be attributed to either the original or new consent process. Sample Characteristics The respondent sample is described in Table 1. In general, respondents were predominantly white (97%), female (66%), and with good to excellent health (92%). More than half of the respondents were 65 years or older (57%) and had more than a high school education (54%). Respondents were most likely to have knee replacement surgery (48%), followed by cholecystectomy (34%) and hip replacement (19%). Respondents reported a higher internal HLOC score than external or chance HLOC scores (26.8, 20.9, and 16.8, respectively on a scale of 6Y36). Calculated alpha reliabilities for each of the dimensions of the HLOC scales ranged from 0.68 to Tables 2 and 3 describe respondent agreement with each of the 12 statements related to consent-related perioperative health communication. In summary, 90% or more of the respondents slightly, moderately, or strongly agreed with each of the statements (Table 2). Top box analyses (Table 3) demonstrated the majority of the respondents strongly agreed that they were well informed about their surgery (68.8%) and that their physician explained the surgical procedure in a way they could understand (71.2%) and listened carefully to patient questions (70.5%). In contrast, slightly more than half of the respondents strongly agreed that the nurses explained the surgery in a way they could understand (51.5%) and listened carefully to patient questions (53.8%). A slight majority of the respondents strongly agreed that the surgery consent form was easy to read (57.5%). Importantly, a majority of the respondents strongly agreed that they understood the risks (71.5%) and benefits (76.7%) of their surgery. A smaller majority of the respondents strongly agreed that they had the chance to ask questions about their surgery (64.9%), had the opportunity to make decisions about their surgery (65.6%), felt comfortable asking questions about their surgery (67.5%), and were asked by the nurse to restate the type of surgery being performed in their own words (64.2%). Calculated > reliability for the 12 questions related to respondent perceptions of consent-related perioperative communication was 0.93 when coded for top box analysis. Bivariate Analyses Respondent perceptions for affiliates adopting the new surgical consent process and those not adopting the new process were compared in bivariate analyses (Table 3). A difference was found between the proportion of respondents strongly agreeing that the nurse asked them to restate the type of surgery being performed in their own words for the new and original consent process (71.2% versus 58.9%, P = 0.004). No other differences were noted between the original and new consent process. Multivariable Analyses Multivariable logistic regression analysis was performed to evaluate the relationship between the perceptions of consentrelated, perioperative health communication and the type of consent process used, while controlling for observed background differences in education and type of surgery (Table 4). Respondents who experienced the new consent process had significantly higher odds of strongly agreeing that the nurses asked them to restate the type of surgery being performed in their own words (adjusted odds ratio [AOR], 1.92; 95% confidence interval [CI], 1.30Y2.82) and that they were comfortable asking questions about their surgery (AOR, 1.53; 95% CI, 1.04Y2.26) compared with those experiencing the original consent process. Respondents with at least some college education had significantly higher odds of strongly agreeing that they had the chance to ask questions about the surgery (AOR, 1.55; 95% CI, 1.07Y2.24), felt comfortable asking questions about the surgery (AOR, 1.59; 95% CI, 1.09Y2.32), and understood nurses explanations of the surgery (AOR, 1.59; 95% CI, 1.92Y2.25), when compared with those with a high school education or below * 2011 Lippincott Williams & Wilkins

4 J Patient Saf & Volume 7, Number 1, March 2011 Patient Perceptions of a Revised Health Literacy-Based Consent Process TABLE 1. Background Characteristics Total n (%) Original Process n (%) New Process n (%) P Sex Total 545 (100) 312 (100) 233 (100) Male 183 (33.6) 105 (33.7) 78 (33.5) Female 362 (66.4) 207 (66.4) 155 (66.5) Race Total 540 (100) 308 (100) 232 (100) White 523 (96.9) 301 (97.7) 222 (95.7) Nonwhite 17 (3.2) 7 (2.3) 10 (4.3) Education Total 538 (100) 307 (100) 231 (100) High school or below 248 (46.1) 130 (42.4) 118 (51.1) Some college or higher 290 (53.9) 177 (57.7) 113 (48.9) Health status Total 539 (100) 308 (100) 231 (100) Poor 1 (0.19) 0 (0) 1 (0.4) Fair 43 (8.0) 25 (8.1) 18 (7.8) Good 269 (49.9) 152 (49.4) 117 (50.7) Very good 189 (35.1) 107 (34.7) 82 (35.5) Excellent 37 (6.9) 24 (7.8) 13 (5.6) Type of surgery Total 545 (100) 312 (100) 233 (100) Gall bladder 184 (33.8) 85 (27.2) 99 (42.5) Knee 259 (47.5) 164 (52.6) 95 (40.8) Hip 102 (18.7) 63 (20.2) 39 (16.7) Age, y Total 545 (100) 312 (100) 233 (100) G65 y 235 (43.1) 124 (39.7) 111 (47.6) Q65 y 310 (56.9) 188 (60.3) 122 (52.4) Multidimensional HLOC (mean [SD], n) Internal 26.8 (4.4), (4.5), (4.3), Chance 16.8 (5.7), (5.7), (5.7), External (powerful others) 20.9 (5.7), (5.6), (5.9), Respondents undergoing orthopedic surgery had significantly and consistently higher odds of strongly agreeing that they understood the risks (knee [AOR, 1.76; 95% CI, 1.15Y2.71] and hip [AOR, 2.09; 95% CI, 1.18Y3.69]) and benefits (knee [AOR, 1.75; 95% CI, 1.11Y2.74] and hip [AOR, 2.93; 95% CI, 1.52Y5.64]) of their surgery and were comfortable asking questions about their surgery (knee [AOR, 1.90; 95% CI, 1.24Y2.91] and hip [AOR, 1.81; 95% CI, 1.06Y3.11]) compared with those undergoing cholecystectomy. DISCUSSION This evaluation of a health literacyybased informed consent process demonstrated important patient safetyyrelated findings about patients perceptions of the perioperative care environment. Patients in hospitals using the new process strongly agreed that they felt more comfortable asking questions and that the nurses were more likely to ask them to restate the type of surgery they were having using their own words, suggesting that the new surgical consent process is associated with an environment conducive to more patient-provider dialogue. This new consent process has the potential to serve as both a patient safety and risk management tool. Informed consent is an ongoing process that begins with the informed consent discussion between the physician and patient in the preoperative period and continues through the perioperative and postoperative phases. 4 Each member of the health care team plays a role in ensuring patient understanding. The physician initiates the discussion, and, most often, a nurse subsequently obtains a patient s signature on a consent document reflecting that the informed consent discussion took place, the content was understood, and the patient had the opportunity to ask and receive answers to any questions. In this study, more than 70% of the respondents reported strong agreement that physicians explained their surgery and listened to their questions about the surgery. Only a slight majority (52%Y54%) of the respondents reported strong agreement related to explanations and listening by nurses. This patient perception reinforces that physicianpatient communication is critical to patients feelings of being well-informed about their surgery and reflects that the primary responsibility for the informed consent discussion resides with the physician. Although a majority of the respondents reported strong agreement that they were well informed about and understood the risks and benefits of their surgery, respondents in hospitals using the new consent process had increased odds of reporting * 2011 Lippincott Williams & Wilkins 33

5 Miller et al J Patient Saf & Volume 7, Number 1, March 2011 TABLE 2. Agreement With Perceptions of Consent-Related Perioperative CommunicationVBivariate Analysis Total n (%) Original Process n (%) New Process n (%) P In general, I was well informed about my surgery. Total 538 (100) 308 (100) 230 (100) 0.83 Slightly, moderately, or strongly agree 523 (97.2) 299 (97.1) 224 (97.4) Slightly, moderately, or strongly disagree 15 (2.79) 9 (2.9) 6 (2.6) My doctor explained my surgery in a way I could understand. Total 538 (100) 308 (100) 230 (100) 0.24 Slightly, moderately, or strongly agree 518 (96.3) 294 (95.5) 224 (97.4) Slightly, moderately, or strongly disagree 20 (3.7) 14 (4.6) 6 (2.6) My doctor listened carefully to my questions about surgery. Total 538 (100) 308 (100) 230 (100) 0.59 Slightly, moderately, or strongly agree 524 (97.4) 299 (97.1) 225 (97.8) Slightly, moderately, or strongly disagree 14 (2.6) 93 (2.92) 5 (2.2) My nurses explained my surgery in a way I could understand. Total 532 (100) 304 (100) 228 (100) 0.37 Slightly, moderately, or strongly agree 496 (93.2) 286 (94.1) 210 (92.1) Slightly, moderately, or strongly disagree 36 (6.8) 18 (5.9) 18 (7.9) My nurses listened carefully to my questions about my surgery. Total 535 (100) 306 (100) 229 (100) 0.52 Slightly, moderately, or strongly agree 504 (94.2) 290 (94.8) 214 (93.5) Slightly, moderately, or strongly disagree 31 (5.8) 16 (5.23) 15 (6.6) I had the chance to ask questions about my surgery. Total 536 (100) 305 (100) 231 (100) 0.65 Slightly, moderately, or strongly agree 520 (97.0) 295 (96.7) 225 (97.4) Slightly, moderately, or strongly disagree 16 (3.0) 10 (3.3) 6 (2.6) I had the chance to make decisions about my surgery. Total 537 (100) 306 (100) 231 (100) 0.12 Slightly, moderately, or strongly agree 513 (95.5) 296 (96.7) 217 (93.9) Slightly, moderately, or strongly disagree 24 (4.5) 10 (3.3) 14 (6.1) I understood the risks of my surgery. Total 537 (100) 306 (100) 231 (100) 0.55 Slightly, moderately, or strongly agree 528 (98.3) 300 (98.0) 228 (98.7) Slightly, moderately, or strongly disagree 9 (1.7) 6 (2.0) 3 (1.3) I understood the benefits of my surgery. Total 536 (100) 306 (100) 230 (100) 0.74 Slightly, moderately, or strongly agree 533 (99.4) 304 (99.4) 229 (99.6) Slightly, moderately, or strongly disagree 3 (0.6) 2 (0.7) 1 (0.4) The consent form for my surgery was easy to read. Total 532 (100) 302 (100) 230 (100) 0.26 Slightly, moderately, or strongly agree 518 (97.4) 292 (96.7) 226 (98.3) Slightly, moderately, or strongly disagree 14 (2.6) 10 (3.3) 4 (1.7) I was asked to tell the nurse what surgery I was having done, using my own words. Total 523 (100) 297 (100) 226 (100) 0.23 Slightly, moderately, or strongly agree 472 (90.3) 264 (88.9) 208 (92.0) Slightly, moderately, or strongly disagree 51 (9.8) 33 (11.1) 18 (8.0) I felt comfortable asking questions about my surgery. Total 535 (100) 306 (100) 229 (100) 0.95 Slightly, moderately, or strongly agree 516 (96.5) 295 (96.4) 221 (96.5) Slightly, moderately, or strongly disagree 19 (3.6) 11 (3.6) 8 (3.5) strong agreement that they were comfortable asking questions about their surgery. The new consent process also was associated with increased odds of respondents reporting strong agreement that nurses asked them to report what surgery they were having done, in their own words. This use of teach-back is a key component of the new health literacyybased consent process, supporting the NQF best practices 9 and may serve as an additional safeguard to protect patients from harm as it has previously been shown to improve patient comprehension of the surgical consent process. 5,6 Use of the reader-friendly consent document, with its inherent prompt for teach-back, allows providers to carry out a final check for patient understanding and if they identify lack of clarity about their procedure or potential misunderstanding to stop the line before the procedure starts * 2011 Lippincott Williams & Wilkins

6 J Patient Saf & Volume 7, Number 1, March 2011 Patient Perceptions of a Revised Health Literacy-Based Consent Process TABLE 3. Strong Agreement With Perceptions of Consent-Related Perioperative CommunicationVBivariate Analysis Total n (%) Original Process n (%) New Process n (%) P In general, I was well informed about my surgery. Total 538 (100) 308 (100) 230 (100) Strongly agree 370 (68.8) 210 (68.2) 160 (69.6) All other responses 168 (31.2) 98 (31.8) 70 (30.4) My doctor explained my surgery in a way I could understand. Total 538 (100) 308 (100) 230 (100) Strongly agree 383 (71.2) 213 (69.2) 170 (73.9) All other responses 155 (28.8) 95 (30.8) 60 (26.1) My doctor listened carefully to my questions about surgery. Total 538 (100) 308 (100) 230 (100) Strongly agree 379 (70.5) 215 (69.8) 164 (71.3) All other responses 159 (29.6) 93 (30.2) 66 (28.7) My nurses explained my surgery in a way I could understand. Total 532 (100) 304 (100) 228 (100) Strongly agree 274 (51.5) 155 (51.0) 119 (52.2) All other responses 258 (48.5) 149 (49.0) 109 (47.8) My nurses listened carefully to my questions about my surgery. Total 535 (100) 306 (100) 229 (100) Strongly agree 288 (53.8) 165 (53.9) 123 (53.7) All other responses 247 (46.2) 141 (46.1) 106 (46.3) I had the chance to ask questions about my surgery. Total 536 (100) 305 (100) 231 (100) Strongly agree 348 (64.9) 199 (65.3) 149 (64.5) All other responses 188 (35.1) 106 (34.8) 82 (35.5) I had the chance to make decisions about my surgery. Total 537 (100) 306 (100) 231 (100) Strongly agree 352 (65.6) 205 (67.0) 147 (63.6) All other responses 185 (34.5) 101 (33.0) 84 (36.4) I understood the risks of my surgery. Total 537 (100) 306 (100) 231 (100) Strongly agree 384 (71.5) 221 (72.2) 163 (70.6) All other responses 153 (28.5) 85 (27.8) 68 (29.4) I understood the benefits of my surgery. Total 536 (100) 306 (100) 230 (100) Strongly agree 411 (76.7) 236 (77.1) 175 (76.1) All other responses 125 (23.3) 70 (22.9) 55 (23.9) The consent form for my surgery was easy to read. Total 532 (100) 302 (100) 230 (100) Strongly agree 306 (57.5) 173 (57.3) 133 (57.8) All other responses 226 (42.5) 129 (42.7) 97 (42.2) I was asked to tell the nurse what surgery I was having done, using my own words. Total 523 (100) 297 (100) 226 (100) Strongly agree 336 (64.2) 175 (58.9) 161 (71.2) All other responses 187 (35.8) 122 (41.1) 65 (28.8) I felt comfortable asking questions about my surgery. Total 535 (100) 306 (100) 229 (100) Strongly agree 361 (67.5) 198 (64.7) 163 (71.2) All other responses 174 (32.5) 108 (35.3) 66 (28.8) Adoption and use of a health literacyybased consent document and process could have positive implications for all health care settings with respect to patient understanding and satisfaction. Incorporating a similar document and process for all procedures requiring formal consent could promote patient understanding through the use of teach-back. It could be surmised that incorporation of teach-back used in the consent process could extend to everyday communication processes with patients. Depending on the situation, passive diffusion and ineffective dissemination of new interventions can be inefficient, costly, * 2011 Lippincott Williams & Wilkins 35

7 Miller et al J Patient Saf & Volume 7, Number 1, March 2011 TABLE 4. Strong Agreement With Perceptions of Consent-Related Perioperative CommunicationVMultivariable Analysis Type of Surgery New Process Gall Bladder Education No Knee High School or Less Yes Hip College Q1 Year In general, I was well informed about my surgery. AOR (95% CI) 1.17 (0.80Y1.71) 1.76 (1.15Y2.69) 1.20 (0.82Y1.75) AOR (95% CI) NA 1.57 (0.92Y2.68) NA My doctor explained my surgery in a way I could understand. AOR (95% CI) 1.32 (0.89Y1.96) 1.42 (0.91Y2.20) 1.21 (0.82Y1.78) AOR (95% CI) NA 0.96 (0.56Y1.63) NA My doctor listened carefully to my questions about my surgery. My nurses explained my surgery in a way I could understand. My nurses listened carefully to my questions about my surgery. I had the chance to ask questions about my surgery. I had the chance to make decisions about my surgery. AOR (95% CI) 1.11 (0.75Y1.64) 1.19 (0.78Y1.82) 1.19 (0.81Y1.74) AOR (95% CI) NA 1.66 (0.94Y2.94) NA AOR (95% CI) 1.06 (0.75Y1.52) 1.08 (0.72Y1.60) 1.59 (1.12Y2.25) AOR (95% CI) NA 0.89 (0.54Y1.46) NA AOR (95% CI) 0.97 (0.68Y1.39) 0.98 (0.66Y1.46) 1.23 (0.86Y1.74) AOR (95% CI) NA 0.95 (0.58Y1.55) NA AOR (95% CI) 1.02 (0.70Y1.47) 1.26 (0.83Y1.90) 1.55 (1.07Y2.24) AOR (95% CI) NA 1.21 (0.72Y2.04) NA AOR (95% CI) 0.91 (0.63Y1.31) 1.56 (1.03Y2.36) 1.27 (0.88Y1.83) AOR (95% CI) NA 1.07 (0.64Y1.78) NA I understood the risks of my surgery. AOR (95% CI) 1.02 (0.69Y1.51) 1.76 (1.15Y2.71) 1.20 (0.82Y1.78) AOR (95% CI) NA 2.09 (1.18Y3.69) NA I understood the benefits of my surgery. AOR (95% CI) 1.06 (0.70Y1.61) 1.75 (1.11Y2.74) 1.34 (0.89Y2.04) AOR (95% CI) NA 2.93 (1.52Y5.64) NA The consent form for my surgery was easy to read. I was asked to tell the nurse what surgery I was having done, using my own words. I felt comfortable asking questions about my surgery. AOR (95% CI) 1.06 (0.74Y1.52) 1.47 (0.99Y2.20) 0.90 (0.63Y1.29) AOR (95% CI) NA 1.09 (0.66Y1.80) NA AOR (95% CI) 1.92 (1.30Y2.82) 2.03 (1.33Y3.11) 1.02 (0.70Y1.49) AOR (95% CI) NA 1.31 (0.78Y2.21) NA AOR (95% CI) 1.53 (1.04Y2.26) 1.90 (1.24Y2.91) 1.59 (1.09Y2.32) AOR (95% CI) NA 1.81 (1.06Y3.11) NA and even harmful. 15 To our knowledge, this is the first study demonstrating the positive impact of a systematic organizational approach to dissemination and implementation of a health literacyybased surgical consent process in the real-world setting. Identifying effective interventions that are acceptable, practical, and usable by all members of the health care team in varied health care settings is fundamental to improving health literacy. 20 This successfully-implemented revised health literacyybased surgical consent process enhanced patients comfort with asking questions and their recollection of being asked to explain their pending procedure in their own words, and may contribute to patients empowerment by encouraging more active participation in their health care. Early research into health literacy focused on evaluating and improving the readability of written health information. 21 However, this study also demonstrates that improving the consent process involves more than just simplifying written documents. Despite significant document revision, simplification, and improved readability, there were no reported differences in respondent perceptions about consent form readability between those experiencing the new or original consent process. Although only a slight majority (58%) of the respondents strongly agreed that the consent form was easy to read, more than twothirds (69%) of the respondents strongly agreed that they were well informed about their surgery. In this project, patients actual reading of the consent form was not assessed, although during pilot tests, nurses reported anecdotally that more patients read the new consent form compared with the original consent form. Variations among the type of surgical procedure and patient perceptions about aspects of the perioperative communication process were observed. Compared with those who had a cholecystectomy, respondents who had undergone orthopedic surgery 36 * 2011 Lippincott Williams & Wilkins

8 J Patient Saf & Volume 7, Number 1, March 2011 Patient Perceptions of a Revised Health Literacy-Based Consent Process had higher odds of reporting strong agreement that they understood the risks and benefits of surgery and comfort with asking questions about their surgery. This may reflect differences in the urgency of the procedure. For example, orthopedic surgical procedures, such as knee or hip replacements, may be elective or planned in advance. Hence, there may be more opportunity to ask questions, learn about, and understand the risks and benefits of surgery. Recent research has underscored the importance of feasibility and provider acceptance of interventions to improve the consent process. 5 The findings from this study demonstrate deployment of an effective health literacyybased intervention (i.e., teach-back) recognized to improve comprehension of the surgical consent process 6 in a large integrated health system without observed negative consequences. The respondents did not express concern with the perioperative communication processes. Although not measured as part of this study, no evidence of dissatisfaction related to increased work or time pressure, delays, excess calls to physicians, or procedure cancellations has been observed during implementation of the new consent process, mitigating concerns about potential adverse impact on physicians and nurses involved in the informed consent continuum. Limitations Although the results of this study are important, several factors must be considered when applying them to other health care environments and patient samples. The study was carried out within a single health care system in a defined geographic area with hospitals ranging in size from 38 to 584 licensed beds. The findings are limited by lack of diversity with respect to race/ethnicity, as 97% of the responding study sample was white. Moreover, the sample was highly educated and reported a strong internal HLOC. Finally, only patients who underwent one of 3 types of surgery were included in this study, and the patient care experience may vary with the acuity, urgency, and complexity of the procedure. The characteristics of the responding study sample were compared with representative claims data from the IHS during the same time frame. The survey sample respondents were similar, with respect to age and sex distribution, to patients undergoing these procedures in the IHS. However, a larger percentage of survey respondents had undergone a cholecystectomy than would be expected. This imbalance likely biased findings in a conservative direction, given that patients undergoing cholecystectomy were less likely to report strong agreement that they understood the risks and benefits of their surgery and comfort asking questions about their surgery. Other common demographic characteristics (e.g., race, education) collected in the survey are not available in the claims data set used for this study and could not be assessed for comparability. Additional evaluation should be conducted among other types and sizes of health delivery systems, more diverse patient samples, and patients undergoing surgical procedures other than those studied in this research. This study did not include a measure of patients health literacy. Future research should incorporate brief patient health literacy assessments and explore ways to target and enhance response rates among minority and disadvantaged population subgroups to understand the impact of this health literacyybased consent process on the underserved and those with low health literacy. Another potential limitation in this study relates to differences in observed and unobserved respondent characteristics between the comparison groups as is common in practice-based, real-world research. Although observed differences between comparison groups were accounted for in multivariable models, it is possible that respondents differed on some unobserved characteristics. There also was a difference in response rates between the comparison groups, with study participants from affiliates using the original consent process being more likely to respond than those at facilities adopting the new consent process. However, respondent HLOC did not differ between comparison groups, suggesting no significant difference in respondent beliefs about who was in control of their health. Responses also may have been influenced by the lack of specificity of the data collection instrument and ability to recall or remember specific aspects of the consent document and process that occurred approximately 2 to 4 months before data collection. This study was conducted concurrently with ongoing efforts to address health literacy in IHS. Initiatives to foster questions (e.g., Ask Me 3 22 ) and use of teach-back were encouraged among staff at all facilities. The degree to which individual health care providers were aware of, had experience with, and actually used health literacyyrelated techniques, such as plain language communication principles and teach-back, was not assessed. Hence, differential uptake of these efforts by staff at IHS affiliate hospitals may have influenced the findings of this study. CONCLUSIONS Health literacy is fundamental to safe, high-quality health care. 23 The surgical consent process is a multidimensional process across a continuum that requires ongoing patient-provider interaction and the need to document effective patient-provider communication. This research demonstrates that the consent process can be refined to stimulate patient-centered communication and comfort with asking questions, promoting the use of health literacyyrelated techniques, such as teach-back, in the perioperative care setting. Adopting a health literacyybased informed consent process promotes patient safety and supports health providers obligations to communicate in simple, clear, and plain language. ACKNOWLEDGMENTS The authors thank the Iowa Health System Health Literacy Teams and other staff who were and continue to be involved in this initiative. REFERENCES 1. American Medical Association. Patient Physician Relationship Topics. Informed Consent [American Medical Association website]. Available at: legal-topics/patient-physician-relationship-topics/informed-consent.shtml. Accessed April 9, Falagas ME, Korbila IP, Giannopoulou KP, et al. Informed consent: how much and what do patients understand? Am J Surg. 2009;198:420Y Bottrell MM, Alpert H, Fischbach RL, et al. Hospital informed consent for procedure forms. Facilitating quality patient-physician interaction. Arch Surg. 2000;135:26Y Bernat JL, Peterson LM. Patient-centered informed consent in surgical practice. Arch Surg. 2006;141:86Y Schenker Y, Fernandez A, Sudore R, et al. Interventions to improve patient comprehension in informed consent for medical and surgical procedures: a systematic review. Med Decis Making. 2010; March 31 epub. DOI: / X Fink AS, Prochazka AV, Henderson WG, et al. Enhancement of surgical informed consent by addition of repeat back. * 2011 Lippincott Williams & Wilkins 37

9 Miller et al J Patient Saf & Volume 7, Number 1, March 2011 A multicenter randomized controlled trial. Ann Surg. 2001;252:27Y National Quality Forum. Safe Practices for Better Healthcare: A Consensus Report. Washington, DC: National Quality Forum; National Quality Forum. Improving Patient Safety Through Informed Consent for Patients With Limited Health Literacy: An Implementation Report. Washington, DC: National Quality Forum National Quality Forum. Safe Practices for Better HealthcareV2010 Update: A Consensus Report. Washington, DC: National Quality Forum; Abrams MA, Earles B. Developing an informed consent process with patient understanding in mind. NC Med J. 2007;68:352Y Lorenzen B, Melby CE, Earles B. Using principles of health literacy to enhance the informed consent process. AORN J. 2008;88:23Y Matiasek J, Wynia MK. Reconceptualizing the informed consent process at eight innovative hospitals. Jt Comm J Qual Patient Saf. 2008;34:127Y Schillinger D, Piette J, Grumbach K, et al. Closing the loop. Physician communication with diabetic patients who have low health literacy. Arch Intern Med. 2003;163:83Y Gay LR, Airasian P. Education Research. Competencies for Analysis and Application. 6th ed. Upper Saddle River, NJ: Prentice-Hall Inc.; Berwick DM. Disseminating innovations in health care. JAMA. 2003;289:1969Y Wallston KA, Wallston BS, DeVellis R. Development of the multidimensional health locus of control (MHLC) scales. Health Educ Monogr. 1978;6:160Y Wallston KA. Multidimensional health locus of control (MHLC) scales. June 15, Available at: Accessed April 7, Velanovich V. Behavior and analysis of 36-item Short-Form Health Survey data for surgical quality-of-life research. Arch Surg. 2007;142:473Y StataCorp. Stata statistical software [computer program]. Version College Station, TX: StataCorp LP, Department of Health and Human Services, Office of Disease Prevention and Health Promotion. National Action Plan to Improve Health Literacy. Washington (DC): US. Available at: HLActionPlan/pdf/Health_Literacy_Action_Plan.pdf. Accessed November 7, Rudd RE, Moeykens BA, Colton TC. Health and literacy: a review of medical and public health literature. In: Comings J, Garners B, Smith, eds. Annual Review of Adult Learning and Literacy. New York, NY: Jossey-Bass; Partnership for Clear Health Communication. Ask Me 3. Available at: Accessed April 17, Nielson-Bohlman L, Panzer AM, Kindig DA, eds. Health Literacy. A Prescription to End Confusion. Washington, DC: The National Academies Press; APPENDIX * 2011 Lippincott Williams & Wilkins