Printer Friendly

Learning advanced cardiac life support: a comparison study of the effects of low- and high-fidelity simulation.

ABSTRACT To increase cardiopulmonary arrest survival, the American Heart Association developed basic and advanced cardiac life support (ACLS) courses that expose participants to realistic learning situations. This experimental study compared results of two ACLS classes on measures of knowledge (content exam) and resuscitation skills (performance exam). Both the control and experimental groups consisted of physicians, nurses, emergency medical technicians, respiratory therapists, and advanced health care providers. The control group used low-fidelity simulation (LFS); the experimental group was exposed to enhanced realism via high-fidelity simulation (HFS). The findings showed a positive correlation between enhanced practice and learning but no significant correlation between posttest and skills test scores for the LFS and HFS groups. The HFS group did score higher on both cognitive and behavioral tests, but the difference was not statistically significant. Participants from both groups indicated satisfaction with their forms of simulation experience and course design. In addition, participants' self-confidence to care for a victim of cardiopulmonary arrest was increased after completing their course; profession and work experience had no effect on responses. The largest difference noted was in verbal responses to course satisfaction. The experimental group stated that learning using HFS was enjoyable and adamantly recommended that ACLS should only be taught using HFS. Further study is required to assess if practicing beyond the course enhances short- and long-term retention of ACLS techniques.

Key Words Advanced Cardiac Life Support--Experiential Learning--High-Fidelity Simulation--Resuscitation Education--Knowledge Retention

**********

ADVANCED CARDIAC LIFE SUPPORT (ACLS) IS A COURSE DESIGNED BY THE AMERICAN HEART ASSOCIATION (AHA) TO ENHANCE TREATMENT SKILLS OF HEALTH CARE PROVIDERS WHO DELIVER DIRECT PATIENT CARE OR INCLUDE RESUSCITATIONS AS A JOB RESPONSIBILITY. Through active learning, participants study and practice resuscitation techniques, completing critical thinking activities involved in situations leading to and during actual cardiopulmonary arrests. Because resuscitation skills are diverse and difficult to learn, course participants practice the integration of resuscitation information and skills until they are able to initiate resuscitation without hesitation (AHA, 2007). * Current AHA courses allow participants to view taped demonstrations, practice with peers until a level of skill confidence is attained, and test while being monitored by qualified AHA instructors. Just as advanced life support treatments require an elevated skill level, so does the ability to comprehend the gravity of resuscitation situations. The goal of ACLS is to provide safe, progressive care to arresting individuals for stabilization to the next level of treatment. This requires critical thinking and the ability to transform or transfer knowledge from the classroom to the clinical area (Braslow, Brennan, Batcheller, & Goodman, 2005). * A variety of teaching methods have been attempted to achieve the level of thinking required for successful resuscitation, but there seems to be a missing link. ACLS students learn content and pass both the behavioral and cognitive tests, yet victims of sudden cardiac arrest continue to die (Breitkreutz, Walcher, & Seeger, 2007). Studies on validity, reliability, and practicality of skill assessment sheets have shown that despite researchers' beliefs that skill performance was appropriate at the time of testing, retention of usable knowledge dropped off significantly after brief time periods (Brennan & Braslow, 1995; Brennan, Braslow, Batcheller, & Kaye, 1996). * Standardized skills checklists plus feedback devices, such as debriefing and reflection exercises, could greatly address the limitations of human evaluators. And the ability of high-fidelity simulation (HFS) to portray realistic resuscitation events and accurately record student actions for debriefing could enhance students' retention skills. THE PURPOSE OF THIS STUDY WAS TO DETERMINE IF ACLS COURSE PARTICIPANTS USING HFS WERE MORE SATISFIED WITH SIMULATION DESIGN AND LEARNING AND TESTED HIGHER, COGNITIVELY AND BEHAVIORALLY, AT THE END OF THE COURSE THAN THOSE WHO PARTICIPATED IN AN ACLS COURSE WHERE LOW-FIDELITY SIMULATION (LFS) WAS USED.

Experiential Learning and Reflection The theoretical framework for this study is based on John Dewey's experiential learning philosophy (1947). Dewey described experience as a reciprocal interaction between an individual and his environment; this interface is dynamic in nature and encourages a change in the person's attitude toward future relations (Luekner & Nadler, 1997). Knowledge is based on experience. When an experiential conflict occurs, thought is the product; repeat experiences aid in validating the thought. Dewey believed that pure memorization was not learning, but that combining traditional methods with progressive, intuitive educational tactics was the way to teach (Dewey, 1997).

In ACLS courses, the incorporation of experiential learning and reflection/debriefing enhances knowledge by developing one or both of the following behaviors: retention and transference. By promoting experiential learning in courses, instructors help students retain the information (AHA, 2006). Integrating reflection into an educational opportunity advances experiential learning, thereby promoting retention and transference into the clinical area. When HFS is used to present ACLS in an experiential learning format that provides a safe environment for resuscitating arrest victims, students are able to practice for extended periods while doing no harm.

Studies have shown that the ACLS course offered by the AHA is a viable and effective way to teach students how to manage emergency cardiopulmonary arrests, as evidenced by improved test scores at course end. However, Kidd and Kendall (2006) disputed past studies and questioned whether course participants truly learned or merely memorized facts. While death rates due to cardiac arrest have declined due to the provision of more and improved ACLS courses (Moretti et al., 2007), there are few studies about the effects of ACLS courses on the ability to provide appropriate resuscitation interventions over time. Due to the contradictory evidence for improved test scores translating into better resuscitation efforts and the uncertainty of retention and practice applicability, this researcher chose to isolate one teaching method (the use of HFS) and study its effects on resuscitation knowledge and skill acquisition.

HFS has proven efficacious in cognitive and behavioral health care provider education. Students report higher self-esteem and self-confidence when performing procedures, increased internalization of information, and greater satisfaction in the learning process when HFS is used (Jeffries, 2007). The limited study of HFS in ACLS courses reveals that it is preferred to other forms of teaching (Rodgers, 2007).

By conducting this study using HFS to provide cognitive and psychomotor learning experiences in an ACLS program, this researcher hoped to support the premise that evidence-based therapy would be understood and transferred to the clinical area. If the information is deployed using the experiential learning method, a long-term goal of the AHA, that is, increased survival rates, may be realized (AHA, 2006). Dewey's experiential learning theory holds promise when explaining how and why HFS promotes the learning and retention of ACLS knowledge and skills.

Methodology An experimental, two-group design was used in this study. The goal was to determine if the independent variable, level of simulation (high versus low), could explain variability in the development of the dependent variable (increase in knowledge base and skills of resuscitation). Outcomes were measured by pretest and posttest scores and skills performance scores. Both groups received the same curriculum for ACLS with the exception of the type of manikins used (low-fidelity vs. high-fidelity simulators).

The population for this study included health care providers requiring ACLS certification. The groups were comprised of physicians, nurses, respiratory therapists, advanced practice health care providers, and medics at the basic, intermediate, and paramedic levels. Demographic information collected at the beginning of the course allowed the researcher to minimize differences in group variances. Each group consisted of members from each demographic category to mimic true resuscitation teams.

The two settings for this study were an AHA training center based in a 250-bed hospital and a college of nursing simulation lab, affiliated with a level 1 trauma medical center. The control group, a traditional, instructor-led course, used LFS at the AHA training center. The intervention group used HFS at the college of nursing skills lab. Permission to conduct the study was obtained from two Institutional Review Boards--one from the city where the study was conducted and the other from the researcher's university.

Four hypotheses were developed:

1. ACLS course participants will have significantly higher scores on the ACLS posttest when they experience computerized, high-fidelity simulation rather than instructor-led, low-fidelity simulation for resuscitation practice.

2. ACLS course participants will have significantly higher scores on the ACLS Mega Code Performance Score Sheet when they experience computerized, high-fidelity simulation rather than instructor-led, low-fidelity simulation for resuscitation practice.

3. ACLS course participants will have significantly higher scores on the Simulation Design Scale when they experience computerized, high-fidelity simulation rather than instructor-led, low-fidelity simulation for resuscitation practice.

4. ACLS course participants will have significantly higher scores on the Student Satisfaction and Self-Confidence with Learning Scale when they experience computerized, high-fidelity simulation rather than instructor-led, low-fidelity simulation for resuscitation practice.

SAMPLE The study sample consisted of 53 health care providers who asked to attend an ACLS course conducted in a medical center in the Midwest. Random assignment was completed after potential participants provided consent and registered for a course; every other registrant was designated as either control group or experimental group until the registration was completed. Individuals declining study participation were allowed to participate, but their scores were not included as study data. Due to physical space limitations, student-to-instructor ratios, and student-to-manikin ratio mandates, each course was limited to 30 participants. Waiting lists were created to compensate for pre-course attrition, as the goal of the study was to provide group equality.

Students received preparation packets upon registration and payment. The first evening of the course, all students completed an ACLS pretest and a demographics survey and participated in interactive learning sessions with ACLS instructors. For students participating in the LFS course, the AHA instructors provided all the information required to resuscitate the manikin: vital signs, heart tones, breath sounds, and a description of the patient's overall condition. Students participating in the HFS course gathered all the same assessment data, but this information came from the manikin and its monitoring devices. Instructors in the HFS course facilitated the reality of the resuscitation by acting as helpers for the student resuscitation directors.

The second evening, students actively participated in skills practice stations and tested in adult CPR skills. The third evening, students completed the rated mega-code skills testing and written ACLS content posttest.

If the student did not pass the posttest, remediation and further assessment with a second posttest was completed until the course director deemed the individual met the course objectives or needed to participate in a future class. Results of the second posttest were not included in study data.

INSTRUMENTS Demographic surveys, two AHA testing tools, and two instruments from the National League for Nursing (NLN) were used for this study. The demographic instrument enabled the researcher to determine study participants' expertise, abilities, ACLS experiences, and other potential group variances.

ACLS written examinations, which evaluate cognitive knowledge and content mastery before and after the program, measure learned constructs essential to the health care provider's abilities to resuscitate an arrested patient and provide basic-level health care practice in high-acuity and emergency unit environments (AHA, 2006). The ACLS Mega Code Performance Score Sheet numerically evaluates resuscitation skills at the conclusion of the course. Content validity for this tool was established by three content experts in resuscitation and skills testing. The tool was also pilot tested with expert instructors for ease of use and interrater reliability prior to the study course (AHA, 2007).

The NLN Simulation Design Scale (SDS) was designed to elicit student preferences regarding simulation. This 20-item instrument evaluates "objectives/information, support, problem solving, feedback/guided reflection, and fidelity (realism)" (Jeffries, 2007, p. 152). The SDS is divided into two sections: the specific functions of simulation and the importance of these features to the participant. Content validity was established by 10 simulation content experts. Cronbach's alphas were reported as 0.92 for features and 0.96 for students' perceptions of the importance of the attributes.

The NLN Student Satisfaction and Self-Confidence in Learning Scale measures five items regarding students' satisfaction with simulated learning activities and eight items to measure students' confidence in their ability to care for patients. Content validity for both concepts was established by nine clinical experts. Cronbach's alphas were 0.94 and 0.87 for the Satisfaction and Self-Confidence subscales, respectively (Jeffries, 2007). In this study, students rated how confident they were in resuscitating a victim of cardiopulmonary arrest.

Results DEMOGRAPHIC FINDINGS All 53 participants held advanced practice or specialty certification. In the control group, 15 participants (62.5 percent) were male and nine (37.5 percent) were female; in the experimental group, nine (31.03 percent) were male and 20 (68.97 percent) were female. The mean age was 35.5 years for control group participants (SD = 12.19) and 34.10 years (SD = 10.39) for participants in the experimental course. Years of experience for the control group were greater than for the experimental group (control, M = 10.26 years, SD = 9.97; experimental, M = 8.38 years, SD = 9.01).

Thirteen control group participants (54.17 percent) reported working in a hospital setting; seven (29.17 percent) worked for an ambulance service; two (8.33 percent) worked in a private practice; and one each (4.17 percent) worked in a physician's office or different health care delivery system. Fourteen participants in the experimental group (48.28 percent) worked in a hospital; seven (24.14 percent) worked for an ambulance service; five (17.24 percent) worked in a physician's office; and three (10.34 percent) worked in an area not previously listed.

DATA ANALYSIS To respond to Hypothesis 1, that experimental group participants will have significantly higher scores on the ACLS posttest, the researcher assessed participants' between- and within-group pretest and posttest mean scores. Parametric t-tests were run to assess for statistical significances in testing scores. Results of pretest scores showed no statistically significant difference in pre-course knowledge between the groups (control, M = 79.06, SD = 19.88; experimental, M = 81.52, SD = 16.48). These results demonstrated that neither group possessed more ACLS knowledge as measured by the pretest.

Results of the posttest scores also showed no statistically significant difference in ACLS knowledge between the groups (control, M = 87.67, SD = 9.28; experimental, M = 90.34, SD= 7.75); t(51) = -1.15, p = .26). Even though neither group excelled posttest, learning did occur.

The change in mean pretest to posttest scores for the control and experimental groups was found to have a statistically significant difference, as was the overall group mean. Analysis of pretest and posttest scores for the control group resulted in a pretest mean of 79.06 (SD = 19.88) and a posttest mean of 87.67 (SD = 9.28), respectively; t(23) = -2.73, p = .01. For the experimental group, the pretest mean was 81.52 (SD = 16.48) and the posttest mean was 90.34 (SD = 7.75); t(28) = -3.91, p = .001. Change in the total study population between the pretest and posttest was also statistically significant (pretest, M = 80.41, SD = 17.96; posttest, M = 89.13, SD = 8.50; t(52) = -4.67, p = .00). Hypothesis 1 was not supported.

The second hypothesis was tested using the ACLS Mega Code Performance Score Sheet. Twelve expert raters evaluated students' resuscitation skills during mega code testing in four testing stations for both courses. The same expert instructors established interrater reliability (IRR) prior to each of the course sessions. The IRR coefficient for January testing sessions was .94; for February testing, the IRR coefficient was 1.00, indicating high rater agreement.

The mean skills test score for the control group was 24.70 (SD = 3.68), in contrast to 26.12 (SD = 2.51) for the experimental group. There was no statistically significant difference between the mean scores for the two groups, t(51) = -1.61, p = .12. Hypothesis 2 was not supported.

For Hypothesis 3, it was found that the entire study population indicated satisfaction with the features of their respective simulated teaching methods. Mean scores for the five subscales of the Simulation Design Scale are found in Table 1.

Course participants were also asked to share their perceptions of the importance of simulation elements. See Table 2 for findings for the five subscales. Based on the statistical analysis presented, Hypothesis 3 was not supported.

To address Hypothesis 4, that participants in the experimental group would have significantly higher Student Satisfaction and Self-Confidence scores compared with the control group, t-tests were run to assess for differences in means scores between the groups. No statistically significant difference was found. The Satisfaction mean scores were 22.54 (SD = 2.69) for the control group and 22.52 (SD = 2.43) for the experimental group. Mean scores for the Self-Confidence subscale, indicating confidence in resuscitation skills, were 35.08 (SD = 4.34) for the control group and 35.03 (SD = 3.28) for the experimental group. Hypothesis 4 was not supported.

Limitations One limitation of the study involves the debriefing sessions. Debriefings were conducted in the same manner for both study groups following mega code resuscitations in order to isolate whether the HFS teaching method was superior to the LFS teaching method. But one has to question whether the debriefing, rather than the fidelity of the manikins, facilitated learning. Rodgers (2007) found that nursing students perceived that debriefing enhanced their learning. Following a gold standard, debriefing is an expectation of the researcher and her training center's instructors. Future studies could exclude debriefing sessions for the control group, comparing outcomes with that of the experimental group.

Another limitation was due to the instructor- and manikin-to-participant ratios mandated by the AHA. To maintain the six students per one instructor and three students per manikin ratio, courses must remain small. Although the power was established to be appropriate for this study, a larger sample might be able to demonstrate differences not apparent in this study. However, increasing the group size may not be feasible to maintain control and organization. When course sizes increase, possibilities arise in which students perform at basic levels (rote memorization) instead of at the highest potential (synthesis, analysis, and evaluation) and lack total understanding of the resuscitation process. This missed competency level could be the assessment or skills required to resuscitate a victim of cardiopulmonary arrest.

Due to the instructor and manikin ratios mandated by the AHA, if the courses were larger, providing greater generalizability, the ability to detect potential differences in the results could have occurred (Rodgers, 2007). A prospective solution for this problem would be to study the participants in ACLS courses conducted over a one-year period. The natural remedy would be to increase course size, but space, equipment necessities, and instructor availability for multiple, 15- to 16-hour courses would be prohibitive.

Importance for the Nursing Profession As nurses become accountable for resuscitating arresting patients, their requirement to practice critical thinking exercises multiplies, Resuscitation algorithms must be learned and understood (Field, 2004). Nurse team members must achieve competency by practicing ACLS critical thinking exercises under the auspices of expert instructors, learning to apply, analyze, and synthesize information necessary for successful resuscitations.

It is understood that motivation is key to increasing knowledge (Dewey, 1947). When motivated and having fun, one learns faster and is inspired to continue learning. The AHA course focuses on core knowledge and experiential learning. For nurses to identify and challenge established concepts and explore the evidence on which practice is based, they must build a knowledge reservoir based on theory and practice (Field, 2004). The expert ACLS instructor aids in filling this reservoir, providing mentoring and developing lifelong learning practices that are essential to the future of all nurses and health care providers (Kim, Kim, Min, Yang, & Nam, 2002).

ACLS courses also aid in developing leadership skills. Nurses who act as team leaders in resuscitation must develop self-reflection, flexibility, advocacy, and communication skills. But for leadership to be learned, the educational program must be creative and redefine nurses' roles.

HFS simulation provides a safe environment for nurses in all positions to practice leadership skills to the level of proficiency. With the use of HFS, nurses can improve skill versatility and increase self-confidence. Potential results will be decreased training time for new nurses to achieve competency, decreased use of training dollars based on the premise that staff becomes proficient quicker, and improved education methods to strengthen learning and improve patient care.

Recommendations for Further Research Very few articles exist regarding ACLS and even fewer for using HFS to teach ACLS. The AHA advanced life support classes have been effective tools for saving victims of cardiopulmonary arrest for over 20 years. Whereas short-term knowledge is clearly improved, how long does this last? Noted by several researchers, further studies are needed to establish participants' abilities to transfer knowledge learned in class to the bedside, long-term retention of this knowledge, and ultimately the increased survival rates of victims of cardiopulmonary arrests.

Studies should also be considered to analyze costs of resuscitating arrest victims who are treated immediately by ACLS-proficient practitioners versus bystanders or medics with less emergency assessment skills and ability to treat. In this era of cost containment, advanced technological teaching methods such as HFS could prove efficacious in reducing resuscitation costs, especially if sentinel events occur due to lack of training.

Conclusions The purpose of this study was to test the degree of simulation fidelity for its potential influence on learning outcomes. None of the study's hypotheses were supported. While significant gains in knowledge were demonstrated on the posttest by both groups, the HFS group was not significantly higher than the control group, although the results trended in that direction. HFS teaching methods did not produce higher skills scores, satisfaction with simulation design features and their importance, or satisfaction with the learning experience, which promoted self-confidence in performing resuscitation techniques. But HFS should not be discarded. Further research must be conducted with more participants, collecting qualitative data, and analyzing costs to assess if HFS simulation is a passing technological phase or a glimpse of improved health care provision.

References

American Heart Association. (2006). Advanced cardiovascular life support instructor manual. Dallas, TX: Author.

American Heart Association. (2007). History of the American Heart Association. Retrieved July 10, 2007, from www.americanheart.org/ presenter.jhtml?identifier=10860

Braslow, A., Brennan, R., Batcheller, A., & Goodman, I. (2005). Continuation of research activities for ECC BLS health care provider CPR evaluation instrument development. American Heart Association Emergency Cardiovascular Care, 8, 1-45.

Breitkreutz, R., Walcher, F., & Seeger, F. (2007). Focused echocardiographic evaluation in resuscitation management: Concept of an advanced life support-conformed algorithm. Critical Care Medicine, 35(5), S150-S161.

Brennan, R., & Braslow, A. (1995). Skill mastery in cardiopulmonary resuscitation training classes. American Journal of Emergency Medicine, 13(5), 505-508.

Brennan, R., Braslow, A., Batcheller, A., & Kaye, W. (1996). A reliable and valid method for evaluating cardiopulmonary resuscitation training outcomes. Resuscitation, 32(2), 85-93.

Dewey, J. (1947). Experience and education. New York: Macmillan.

Dewey, J. (1997). How we think. Mineola, NY: Dover Press.

Field, D. (2004). Moving from novice to expert. The value of learning in clinical practice: A literature review. Nurse Education Today, 24, 560-565.

Jeffries, P. (Ed.). (2007). Simulation in nursing education: From conceptualization to evaluation. New York: National League for Nursing.

Kidd, T., & Kendall, S. (2006). Review of effective advanced cardiac life support using experiential learning. Journal of Clinical Nursing, 16, 58-66.

Kim, J., Kim, W., Min, K., Yang, J., & Nam, Y. (2002). Learning by computer simulation does not lead to better test performance than textbook study in the diagnosis and treatment of dysrhythmias. Journal of Clinical Anesthesia, 14(5), 395-400.

Luckner, J., & Nadler, R. (1997). Processing the experience: Strategies to enhance and generalize learning. Dubuque, IA: Kendall/Hunt.

Moretti, M., Cesar, L., Nusbacher, A., Kern, K., Timerman, S., & Ramires, J. (2007). Advanced cardiac life support training improves long-term survival from in-hospital cardiac arrest. Resuscitation, 72,458-465.

Rodgers, D. L. (2007). The effect of high-fidelity manikin-based human patient simulation on educational outcomes in advanced cardiovascular life support classes. Unpublished manuscript.

Theresa A. Hoadley, PhD, RN, TNS, is an assistant professor at OSF Saint Frances Medical Center College of Nursing, Peoria, Illinois. The author is grateful to the A CLS instructors from the Proctor Hospital Training Center, Tom and Danelle Geraci, Lisa Roth, Matt Ross, Gerri Hellhake Hall, Jill Weberski, Julie Smith, Mona Aberle, Barb Ekstrum, Janet Callahan, Cyndi Colgan, Patti O'Connor, Maggi Ballard, and Christa Fuller, who facilitated the courses and worked to establish interrater reliability for the study's tools. She also acknowledges the members of her dissertation committee, Dr. Deborah Leners, Dr. Carol Roehrs, Dr. Pamela Jeffries, and Dr. David DanieL For more information, write to Dr. Hoadley at Theresa.hoadley@osfbealthcare.org.
Table 1. Satisfaction with Features of Simulated Teaching
Methods, Simulation Design Scale

SUBSCALE CONTROL EXPERIMENTAL

 MEAN SD MEAN SD

Objectives/
Information 23.17 1.93 23.83 1.69

Support 18.92 1.82 18.66 2.01

Problem Solving 22.96 2.31 22.41 3.15

Feedback/
Guided Reflection 19.13 1.45 19.31 1.26

Fidelity (Realism) 9.13 1.39 9.38 0.9

Table 2. Perceptions of the Importance of Simulation Elements,
Simulation Design Scale

SUBSCALE CONTROL EXPERIMENTAL

 MEAN SD MEAN SD

Objectives/
Information 23.00 2.23 22.83 2.80

Support 18.25 2.05 18.86 1.71

Problem Solving 22.96 2.49 21.38 3.16

Feedback/
Guided Reflection 18.50 1.91 18.66 1.70

Fidelity (Realism) 9.08 1.32 9.28 1.16
COPYRIGHT 2009 National League for Nursing, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:NURSING EDUCATION RESEARCH
Author:Hoadley, Theresa A.
Publication:Nursing Education Perspectives
Date:Mar 1, 2009
Words:4216
Previous Article:Simulation as a means to foster collaborative interdisciplinary education.
Next Article:Student satisfaction with high-fidelity simulation: does it correlate with learning styles?
Topics:


Related Articles
Changing attitudes about persons with disabilities: effects of a simulation.
Janet L. Grady, DrPH. RN. began her career as a high school foreign language teacher, seeking innovative ways to increase student involvement in the...
Georgia nurses and end of life care: results of a statewide educational needs survey.
Tennessee Nursing Partnership promotes skill-advancement in simulation technology for nurse educators in Tennessee.
Congratulations to recipients of 2007-2008 AACN grants.
Dreams for the future for clinical simulation: research and development in this hot topic area will lead to more sophisticated manikins and robots...
A collaborative project to apply and evaluate the clinical judgment model through simulation.
Modeling and simulation: an innovative teaching-learning strategy.
The oxygen supply and demand framework: a tool to support integrative learning.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters