HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) A correction has been published Respond to This Article Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Search for Related Content

2006 National Teaching Institute Research Abstracts

	RESEARCH ORAL PRESENTATIONS

Top RESEARCH ORAL PRESENTATIONS RESEARCH ABSTRACTS

To Feed or Not To Feed? That Is the Question: Aspiration Prevention in Acute Stroke Patients

Harris D, Gadishaw S, Larghi, P. Memorial Hospital West, Fla.

Purpose: To determine whether nurses (RNs) could quickly and accurately assess the swallowing of acute stroke patients before administering medications or food. Background/Significance: Dysphagia is a frequent manifestation following acute stroke. The term dysphagia is used to describe disorders that occur in the oral or pharyngeal phases of swallowing. Dysphagia develops in 27% to 50% of all stroke patients. Furthermore, 43% to 54% of stroke patients with dysphagia will experience aspiration, and among those patients, pneumonia will develop in 37%. A swallowing tool (FAST-Feed After Stroke Tool) based on the Dysphagia Nursing Screen and the SStuff tool was developed.The RNs’ results were compared with those of speech therapy evaluations of the same patients. Methods: The FAST tool development and study was done at a Joint Commission on the Accreditation of Healthcare Organizations (JCAHO) primary certified community hospital that is participating in a 1-year JCAHO stroke study. Measure-7/Screen for Dysphagia was the catalyst for our hospitals to improve dysphagia assessment. Fifty patients were assessed with the FAST tool. The RN quickly (mean 9 min) and accurately assessed patients’ swallowing and consulted speech therapy for further evaluations. Results: The FAST results that led RNs to have patients take nothing by mouth correlated 100% with speech therapy results. Speech therapy further recommended alternative dysphagia diets such as thickened liquids and pureed foods. None of the patients were readmitted with aspiration pneumonia within 30 days. Conclusions: The success of the FAST tool will continue to be monitored. The tool is slated for implementation for use with acute stroke patients.

Research Oral Abstract Award Winner

Prevalence of Electrocardiographic Abnormalities in Patients With Subarachnoid Hemorrhage

Hravnak M, Crago E, Kong Y, Horowitz M. University of Pittsburgh, Pa.

Purpose: To examine the prevalence of electrocardiographic (ECG) abnormalities in patients after aneurysmal subarachnoid hemorrhage (SAH-A), and their relationship to sympathetic stimulation and myocardial injury. Background/ Significance: Although ECG abnormalities and their possible causes have been described to some degree for patients with embolic stroke, they have not been well explored in patients with hemorrhagic stroke, specifically SAH-A. Methods: This prospective longitudinal study recruited 74 SAH-A patients ages 38–75 (mean 55.8, SD 9.6) with a Fisher grade 2 and/or a Hunt/Hess grade 3 admitted to the neurovascular intensive care unit. ECG abnormalities were determined from initial 12-lead ECG and continuous 24-hour Holter monitoring for 5 days after SAH-A. Concentrations of total serum catecholamines were measured every 12 hours. Serum markers for myocardial ischemia and infarct (TCPK, CPK/MB, CPK/MBI and cardiac troponin I [cTnI]) were measured daily. Descriptive, ², and repeated measures analyses were performed by using SAS v8.2. Results: ECG abnormalities were present by initial 12-lead ECG in 82% of patients, and in 93% by Holter monitor. A significant relationship existed between 12-lead ECG and total serum catecholamines (P = .05), and a nonsignificant trend was noted between 12-lead ECG and cTnI (P = .06), but not between 12-lead ECG and CPK/MB (P = .1). There was no relationship between ECG abnormalities by Holter monitor and total serum catecholamines, cTnI, or CPK/MB. Conclusions: ECG abnormalities are common after SAH-A. There is some evidence to support relationships between abnormalities on the initial 12-lead ECG, sympathetic stimulation, and some markers of cardiac injury, but these relationships are not present with regard to 24-hour Holter monitoring data. Vigilance in monitoring for ECG abnormalities in this complex patient population is important for optimal care, and the reasons for causation require further study. Sponsored by: National Institutes of Health, National Heart, Blood, and Lung Institute.

Research Oral Abstract Award Winner

Trajectory of Quality of Life in Heart Failure Patients With Preserved and Nonpreserved Systolic Function

De Jong M, Riegel B, Armola R, Moser D. Wilford Hall Medical Center, Tex.

Purpose: To investigate whether the trajectory of health-related quality of life (HRQL) in heart failure (HF) patients differs by sex or systolic function during the first 6 months after hospitalization. Background/Significance: Heart failure patients have worse HRQL than do patients with other cardiac conditions, and HRQL predicts mortality. Clinicians often assume that patients with preserved systolic function (ejection fraction [EF] > 0.40) have better HRQL than those with nonpreserved systolic function (EF = 0.40). Differences between the sexes in HRQL have been proposed, yet little is known about how HRQL changes over time based on either sex or EF. Methods: This was a longitudinal study of HF patients enrolled in the Multicenter Quality of Life Registry. The Minnesota Living with Heart Failure Questionnaire (LHFQ) was used to measure HRQL at baseline (during hospitalization), 3 and 6 months after discharge. Repeated-measures analysis of variance determined whether trajectory of HRQL differed by sex or preserved versus nonpreserved systolic function. Results: A total of 419 patients (age 70 ± 12 years; 51% female; EF 42 ± 18%; New York Heart Association class III/IV 71%) completed the LHFQ at each interval. During hospitalization, HRQL was poor for all patients. Neither sex nor EF group interacted to produce an effect on the trajectory of HRQL. There was a main effect of time. All patients reported substantially worse HRQL during hospitalization than at 3 months and 6 months after discharge (P<.001). Regardless of sex or EF, HRQL improved from baseline at 3 months (P < .001) and 6 months (P < .001). Moreover, HRQL was better at 6 months than at 3 months (P<.001). Conclusions: HRQL was most impaired during hospitalization, but significantly improved after discharge for both men and women, irrespective of EF. Clinicians may advise male and female patients that HRQL will most likely improve after discharge. However, clinicians should be aware that HRQL is appreciably impaired even for HF patients with a preserved EF. Sponsored by: Philips Medical-AACN Outcomes Grant.

Research Oral Abstract Award Winner

Development and Application of the Workload Management System for Critical Care Nurses (WMSCN) Using the Workload Management System for Nurses (WMSN)

Cho Y, Shin H, Cho J, Jung M, Lee B. Samsung Medical Center, International.

Purpose: To develop a nursing workload classification system in Korea specific to intensive care units (ICUs), establish the validity and reliability of this system, and identify the conversion index of nursing hour per 1 point of nursing workload score. Background/Significance: In 1992, the Korean Association of Clinical Nursing had revised the factor-typed patient classification system (PCS) with WMSN, developed at Walter Reed Army Hospital. But validity and reliability testing and appropriate teaching of correct guidelines were not done. Many differences were apparent between general wards and ICUs and also between Korea and the United States. Therefore, the Korean Association of Critical Care Nurses decided to develop a valid and reliable WMSCN for ICUs in Korean hospitals. Methods: The items were revised at the 256 ICUs in 45 hospitals from WMSN by checking the frequencies of items. Nursing care hours (NCH) were checked by trained observer and WMSCN scores were checked by registered nurses (RNs) caring for patients after training on the guidelines. Pretraining and posttraining scores were compared. Validity was tested through the correlation between direct NCH and WMSCN score. Interrater reliability between head nurses and RNs was tested at 18 of 45 hospitals. Finally, the conversion index of NCH by point of score was measured at 5 of the 18 hospitals. Results: The WMSCN consisted of 8 categories and 82 nursing indicators. The average WMSCN score decreased from 124.0 to 82.2 after training on the guidelines. Scores of RNs were highly correlated with scores of head nurses (r = .94). WMSCN scores were highly correlated with the direct NCH (P = .007). Finally, the conversion index was 7.99 minutes. Conclusions: WMSCN is valid and reliable for classifying ICU patients according to their nursing workload. Training on the guidelines and repeated evaluations of validity and reliability are required to use WMSCN effectively. The conversion index should be adjusted to estimate the appropriate staffing in Korea. Sponsored by: Korean Association of Critical Care Nurses.

Practices of Nurses and Nursing Assistants in Preventing Incontinence Dermatitis in Acutely/Critically Ill Patients

Peterson K, Bliss D, Nelson C, Savik K. Methodist Hospital and University of Minnesota, Minn.

Purpose: To describe current practices of nursing staff in preventing and managing incontinence-associated dermatitis (IAD) in acutely and critically ill patients. Background/Significance: Acutely and critically ill patients are at risk for IAD. There is no standard protocol for preventing or managing IAD, and numerous skin care products are available. Little is known about the practices of nursing staff to address IAD in these patients. Methods: 8 nurses (8 female, 1.5–30 years hospital employment) and 8 nursing assistants (6 female, 4–6 years employment) in 3 community hospital units (ICU, oncology, and a medical unit) participated in 1-hour focus groups. A 3-day prospective surveillance showed that the prevalence of IAD in incontinent patients in those units was 95%, 35%, and 32%. Focus group responses were audiotaped, transcribed, and analyzed by 2 investigators. Results: Staff recognized the importance of preventing IAD. Timely cleansing of soiling was identified as essential, but incontinence frequency, lack of time, and patients’ acuity were barriers. Staff used a variety of cleansers including soap and water, baby shampoo, shaving cream, and commercial cleansers whose names they did not know. Moisture barriers or body lotion were used inconsistently, and staff did not usually know product names or ingredients. To keep perineal skin dry, staff used baby powder, placed cloths in skin folds of obese patients, and turned patients. Absorbent briefs were avoided in 2 units, cotton underpads were preferred, rectal tubes were uncommon and reserved for liquid stool, and rectal pouching was rare. Product fragrance and availability at the bedside were primary factors in selection and use of skin care products/devices. Conclusions: Staff opinions about appropriate IAD care differed, and practice varied considerably among individuals and units. Findings support plans to develop a protocol and computerized education program about perineal skin care, IAD prevention, and use of related products/devices. Sponsored by: Park Nicollet Institute and an AACN Evidence-Based Clnical Practice Grant.

Managing Diarrhea and Fecal Incontinence: Results of a Prospective Clinical Study in the Intensive Care Unit

Gallagher J, Wishin J. University of Florida-Shands Hospital, Fla.

Purpose: This prospective study evaluated safety and performance of F* in 42 subjects with diarrhea and incontinence in 7 US hospitals. Background/Significance: Managing fecal incontinence and diarrhea challenges intensive care unit (ICU) and hospital staff. An innovative fecal management system F uses an inflated balloon to retain a tube within the rectum. An external pouch collects fecal material for patients with uncontrolled diarrhea. Methods: Endoscopic rectal vault proctoscopies assessed anorectal mucosal condition before insertion and after removal on 8 initial subjects. Investigators reported on ease of F insertion and removal, device retention and leakage, patients’ comfort, perineal skin condition, and odor during use. F performed well with duration of use ranging from 1 to 14 days. Results: The device was generally well-tolerated. Five deaths occurred from non–product-related illnesses and one subject with multisystem organ disease and history of gastrointestinal bleeding had an episode of lower gastrointestinal bleeding while using the device. Clinicians rated F easy-to-insert (97% of subjects rated), remove (97%), and dispose of (100%) with easy-to-follow instructions (100%). On 200 daily assessments, F was rated as improving fecal incontinence control in 83%, as time efficient in 89%, and as efficacious in 86%. Four subjects were unable to retain the device. Clinicians reported no odor on 85% of daily assessments and no or limited leakage on 82%. Perineal and buttock skin condition were maintained/ improved in 92% of patients with diarrhea and incontinence. Conclusions: The device was determined to have an overall favorable safety profile in this study, helping reduce risk of perineal and buttock skin breakdown. Product notation: *Flexi-Seal Fecal Management System, ConvaTec, Princeton, NJ, USA. Flexi-Seal is a registered trademark of E. R. Squibb & Sons, L. L. C. Sponsored by: ConvaTec, A Bristol-Myers Squibb Co.

Assessment of Intensive Care Unit (ICU) Delirium and Depression

Idemoto B. University Hospitals of Cleveland, Ohio.

Purpose: To examine the accuracy of critical care nurses’ clinical assessment of delirium and depression compared with standardized validated tools. Accuracy of nurses’ assessment of these syndromes can be improved. Background/Significance: Studies have shown high incidence, poor outcomes, and long-term sequelae for ICU patients who experience delirium or depression. Methods: A total of 126 adult postoperative patients with >24-hour stay in the surgical ICU of a large urban medical center were assessed and findings were compared with the assessments of nurses caring for the patients on the day of interview. The Confusion Assessment Method–Intensive Care Unit was used (CAM-ICU) for delirium, and the Hospital Anxiety and Depression Scale (HADS-D, depression subscale) to assess depression, and the nurse caring for the patient was asked for an informal clinical assessment. Results: Following study completion, August 2004, data were examined for concordance using percent agreement and the kappa statistic. Logistic regression was used to assess the influence of patients’ characteristics (age, sex, race, hospital day on day of interview, and primary procedure or diagnosis) and selected characteristics of nurses (sex, educational preparation, years of nursing experience) to assess the accuracy of the nurses’ judgment. The regression model for depression was significant with these predictors of concordance: hospital day (odds ratio [OR] 1.119, P = .02) and nursing years of experience (OR .725, P = .03); the regression model for delirium was not significant. Clinical significance or insights for clinical practice may be found with further analysis. Conclusions: This surgical ICU study is the first to assess nurses’ accuracy of detection of delirium and depression in the postoperative ICU patients. This study demonstrates the feasibility of investigating these complex phenomena in the ICU environment and can be pivotal in the building the foundation for future studies with the ultimate goal of improved outcomes for critically ill patients. Sponsored by: Sigma Theta Tau Alpha Mu Chapter Research Grant and Frances Payne Bolton School of Nursing Alumni Research Award.

Interrater Agreement of the Checklist of Nonverbal Pain Indicators in Intubated and Sedated Patients in Surgical Intensive Care Units (ICUs)

Tyberg K, Chlan L. University of Minnesota, Minn.

Purpose: To determine the interrater agreement of the Checklist for Nonverbal Pain Indicators (CNPI) in intubated and sedated ICU patients, and to have nurses evaluate the feasibility of using the CNPI in their practice. Background/Significance: Pain management in the ICU is a challenge with intubated and sedated ICU patients because of their inability to use the traditional pain scales to describe their pain levels. The CNPI was designed for use in persons with Alzheimer disease who are unable to use traditional pain scales because of cognitive impairment. The CNPI consists of 6 items where the observers rate yes/no if a certain behavior is present at rest and with activity. The CNPI might be useful to evaluate pain for intubated and sedated patients and warrants testing in the ICU. Methods: Eligible patients (n = 8) were enrolled from 1 surgical ICU in the urban Midwest after consent was obtained from family. A pair of nurses observed patients for the presence of CNPI behaviors (bracing, restlessness, grimacing, rubbing) at rest and with repositioning; nurses independently recorded their observations. A total of 22 paired nurse observations were collected. Nurses evaluated the feasibility of the CNPI on a 0–5 Likert scale for ease of use, whether it reflected the patient’s pain, and likelihood of using it in the future. Results: Overall interrater agreement of nurses’ observations of patients’ pain behaviors using the CNPI were high (mean = 97.72%, range 90.91%–100% overall). Nurses thought the CNPI was easy to use (mean = 4.4, SD = 0.62) and that it reflected the patient’s pain (mean = 3.7, SD = 1.1). Only 73% of nurses indicated that they would be likely to use the tool in the future. Conclusions: Nurse ratings of patients’ pain behaviors on the CNPI were in close agreement and the nurses found the tool easy to use. The CNPI has the potential for use in the ICU but requires further testing with a larger sample and investigating the nurses’ hesitancy to use the CNPI in their practice. Sponsored by: AACN/STT Clinical Practice Grant.

Judgments of Nurses and Physicians Regarding Futility and Withdrawal of Treatment in Medical and Surgical Intensive Care Units (ICUs)

Miller C, Funk M, Wiegand D. Yale University School of Nursing, Conn.

Purpose: To compare judgments of nurses vs physicians and clinicians in medical (MICU) vs surgical intensive care units (SICU) regarding futility and withdrawal of treatment; examine the accuracy of judgments in predicting hospital mortality; and determine factors influencing judgments about withdrawal of treatment. Background/Significance: Many patients face death in an ICU. Patients, families, and clinicians often deal with decisions regarding withdrawing and withholding therapy in situations that are deemed medically futile. Methods: Data were collected on 101 adult patients who had been in the MICU or SICU for >48 hours. A brief questionnaire eliciting information on judgments about futility and withdrawal of treatment was completed daily by the patient’s nurse and physician for the duration of the patient’s ICU stay. A total of 1263 questionnaires were completed by 36 nurses and 39 physicians. Both t test and chi-square analyses with kappa coefficients were performed to determine the association of type of clinician and type of unit with judgments, accuracy of judgments, and factors affecting judgments. Results: Nurses were more likely to judge treatment as futile (35.8%) and believe that treatment should be withdrawn (16.9%), compared with physicians (26.6% and 11.4%; P = .001 and P = .02). Clinicians in the MICU were more likely to consider treatment futile (26.2%), and believe that treatment should be withdrawn (24.1%), compared with clinicians in the SICU (8.1% and 3.0%; P < .001 for both). Judgments about futility indicate that nurses and physicians were moderately accurate in predicting hospital mortality (kappa=0.43). Type of unit (P <.001), physician specialty (P <.001), and nurse’s education (P = .004) influenced judgments surrounding withdrawal of treatment. Conclusions: Clinicians should initiate end-of-life discussions earlier in the patient’s ICU course, which can prevent later misunderstandings, and perhaps avoid unnecessary invasive and painful procedures.

Implementing Research-Based Out-of-Bed Guidelines After Percutaneous Transluminal Coronary Angiography (PTCA)

Breton E, Merdinger P, Buonocore D. Bridgeport Hospital, Conn.

Purpose: To safely reduce time in bed after PTCA/stenting by implementing a research-based protocol. Background/Significance: Patients after PTCA in our institution typically spend anywhere from 6 to 18 hours in bed depending on the attending physician, type of anticoagulant used during the procedure, size of the introducer catheter used, and what method was used for femoral artery hemostasis. Increased length of time in bed after femoral arterial sheath removal has been associated with increased discomfort of patients. We had a research-based guideline that was several years old and did not reflect the use of several new anticoagulants and artery closure devices. Methods: In an effort to standardize time out of bed safely, a thorough review and evaluation of the current literature was undertaken. Based on this literature search, the guideline was updated. The new updated guideline allows for ambulation 4 hours after femoral artery hemostasis in PTCA/stent patients with a 6F sheath or smaller and closure by manual compression, CClamp, FemoSTop, Angioseal, or Perclose and using either heparin or bivalrudin for anticoagulation with or without the use of a IIa/IIIb inhibitor. Ambulation is 6 hours for those patients with an 8F or larger sheath. Implementation of the updated guideline was facilitated by staff education, physician education, and an information bulletin board. Results: Data on postprocedural ambulation regarding hematoma formation as well as bleeding complications were tracked through our quality improvement program. Conclusions: Since the implementation of the program, no significant hematoma or bleeding complication related to the earlier time out of bed has been observed.

A Comparison of Sex and Age Differences in Symptoms After Myocardial Infarction and After Coronary Artery Bypass Graft

Sethares K, Carroll D, Buselli E. University of Massachusetts Dartmouth, Mass.

Purpose: To determine if differences between the sexes are present in the symptoms of pain, shortness of breath, and fatigue in persons with a diagnosis of either myocardial infarction (MI) or coronary artery bypass graft surgery (CABG) over time. Background/Significance: Pain, shortness of breath, and fatigue are among the most frequently reported symptoms in persons after MI or CABG. These symptoms can impair quality of life and functional status for up to 1 year after acute hospitalization. Yet, few studies have longitudinally explored symptom change over time. Further, research suggests that differences between the sexes may exist in the experience of pain, shortness of breath, and fatigue. Methods: Data on pain, shortness of breath, and fatigue were collected from 248 persons by telephone at 6 weeks (t1), 3 months (t2), and 6 months (t3) after either an acute MI or CABG. All symptoms were measured on a 0-to-10 scale (0 = absent and 10 = worst possible symptom experience). All data were analyzed by using repeated measures analysis of variance. Results: Of the 248 subjects enrolled in the study, 94 had an MI and 154 underwent CABG. All of the subjects were unpartnered cardiac elders with a mean age of 76 (85 men, 163 women). There were significant reductions in pain, shortness of breath, and fatigue noted over time, but the symptom experience did not differ by sex or cardiac diagnosis. The mean levels of fatigue dropped from 4.54 (t1) to 4.28 (t2) to 3.25 (t3), with the greatest reduction between 3 and 6 months. The mean levels of pain dropped from 3.37 (t1) to 0.82 (t2) to 0.78 (t3), with the greatest reductions between 6 weeks and 3 months. Finally, mean levels of shortness of breath dropped from 3.36 (t1) to 2.18 (t2) to 1.69 (t3), with the greatest reductions between 6 weeks and 6 months. Conclusions: The presence of symptoms decreased over time with a variable pattern of change. It is surprising to note that there were no significant differences in the symptom experience between men and women and between persons with MI or CABG.

Demonstrating the Impact of a Web-Based Education Program for Family Members of Intensive Care Patients

Kleinpell R, Silva N, Tully M, Hancock B. Rush University Medical Center, Ill.

Purpose: To assess the impact of a unique web-based education program for families of intensive care unit (ICU) patients on several outcomes: family education, family satisfaction, content of educational information taught by the nursing staff, and nurses’ satisfaction with use of the program. Background/ Significance: Research has demonstrated that the need for information is a top priority for family members of ICU patients. Yet, retaining information that is verbally communicated is difficult, and printed pamphlets are frequently misplaced. Methods: This descriptive study was conducted with family members and nurses in 3 ICUs in a midwestern university-affiliated medical center. Preimplementation surveys assessing satisfaction with communication and education were conducted with ICU nurses (N = 50) and ICU family members (N = 51), which revealed that a significant amount of time was being spent re-educating family members on ICU equipment. A dedicated computer was provided in the family waiting room to enable families to access educational information using the program ICU-USA. Results: During the first 3 months of use, family members visited more than 300 pages of the educational program. Frequently visited sites included web pages describing the ICU, ICU equipment, diagnoses, procedures, and hospital services including dining and general accommodations. Family satisfaction surveys completed on the computer provided immediate feedback and revealed a high degree of satisfaction with daily communication and information. Postimplementation surveys with the ICU nursing staff are currently being conducted to assess further the impact of the web-based educational program on educational needs of family members and communication content. Conclusions: The use of a web-based educational initiative can improve ICU family member satisfaction with information and enhance communication in the ICU. Sponsored by: Golden Lamp Society.

High School Students’ Perceptions of Nursing Versatility—Stories Told!

Porter G, Boggs P, Bonifield S, Bride W, Granger B. Duke University Health System, NC.

Purpose: To explore high school (HS) students’ perceptions of nursing as a career choice, and to gain insight into their impressions of the nurse as a key healthcare team member. Background/Significance: Consistent with national trends, our institution is experiencing a nursing shortage. To stem this trend, a study was designed to intervene at the HS level. Methods: The Academy of Health Sciences (AHS) was created in our area to assist HS students in selecting health-related careers. We offered students an opportunity to fulfill part (30–40 hours) of their AHS observation requirement by shadowing nurses in our Heart Center Observational Program (HCOP). Three information sessions were held to elicit student interest. Students completed study and shadowing consent forms, a preprogram information form, postobservation questionnaires, a postprogram questionnaire, and a guided interview with the program coordinator. Narrative methods were used to explore 16 taped interviews (45 minutes average). Constant comparison and thematic coding identified salient themes of field notes and recorded narratives. Results: Of 24 students enrolled, 16 completed the program. Misconception of nursing was the dominant theme. Subcategories included misperceptions of tools/ technology used, helping relationships, role responsibility, interdisciplinary respect, patient fragility, and knowledge synthesis. Experiential knowledge of nursing was identified as a core need for students interested in nursing and health careers. Conclusions: These data suggest experiential knowledge of nursing at the HS level may alter perceptions of nursing and influence perceptions on a broader scale than is currently achieved by contemporary media. One student’s impression summarized this experience well, "My idea of nursing has definitely changed!" Sponsored by: Duke Heart Center Fellowship.

Acute Effects of Music on Stress in Patients Receiving Mechanical Ventilatory Support

Chlan L, Engeland W, Anthony A, Guttormson J. University of Minnesota School of Nursing, Minn.

Purpose: To evaluate the effect of relaxing music on the stress response (SR) in mechanically ventilated patients (MVPs) in the intensive care unit (ICU). Background/Significance: Non-pharmacological interventions such as music have been suggested as effective means for interrupting the SR in MVPs. Previous investigations have focused solely on indirect markers of SR activity in the sympathetic nervous system (SNS) only. There is an absence of research measuring direct serum markers of activity in both the hypothalamic-pituitary-adrenal (HPA) axis and SNS. Methods: A 2-group experimental design with repeated measures pilot study was used to address the study purpose. Ten (6 female, 4 male) alert, hemodynamically stable MVPs (mean age 64.9 [SD 8.2] years) with intact renal function were recruited from a university-affiliated tertiary care center in the urban Midwest. Subjects were randomized to 60 minutes of listening to self-selected relaxing music through headphones or to resting quietly for 60 minutes. Heart rate and serum samples were obtained at baseline, +15 minutes, +30 minutes, +60 minutes; samples were assayed for epinephrine, norepinephrine, adrenocorticotropic hormone, and cortisol. Results: Data analysis via Kruskal-Wallis and Friedman test revealed no significant differences between groups on any measures due to the small sample size and wide variability among subjects. The experimental group demonstrated a downward trend for all measures, while the control group showed a variable pattern of increases and decreases for all measures over time. Conclusions: Data trends indicate that music may interrupt the SR in MVPs as demonstrated by a general pattern of reductions in serum markers of HPA axis and SNS activity. Additional research is warranted with an adequately powered sample size to build further the knowledge base of effective nonpharmacological nursing interventions for stress reduction in intensive care patients receiving mechanical ventilation that can improve patients’ outcomes. Sponsored by: Grant-in-Aid University of Minnesota Graduate School.

The Effects of Music Twice Daily on Various Outcomes in Intensive Care Patients Receiving Mechanical Ventilation: Improving Umbilical Venous Catheter Care by Implementing an Evidence-Based Practice Guideline in the Neonatal Intensive Care Unit

Sarin-Gulian A, Heliker B, Gawlinski A. UCLA Medical Center, Calif.

Purpose: To decrease the number of clotted umbilical venous catheters (UVCs) in the neonatal intensive care unit (NICU) by implementing an evidence-based practice (EBP) guideline. Background/Significance: Variability exists in UVC care practices, specifically regarding the frequency of and the solution and amount used for flushing. This variability may lead to lumen clotting, which necessitates placement of alternative catheters. Methods: The study used a pretest-posttest design. The intervention was an EBP UVC care guideline. Outcomes measured were nurses’ knowledge, catheter patency, and practices before and after implementing an EBP guideline. A convenience sample of 64 NICU nurses was surveyed. UVC care practice was observed in 13 cases before and 17 cases after intervention. Results: A significantly higher percentage of nurses achieved mastery of knowledge after the intervention (78%), compared with before the intervention (21%, P < .05). Significant increases also occurred in nurses’ correct stopcock use (23% before, 81% after, P < .05) and consistency in flushing (22% before, 64% after, P < .05). UVC patency increased to 92% (after), compared with 86% (before). Conclusions: The use of an EBP UVC care guideline successfully improved UVC care practice in the NICU. Statistically significant findings were (1) increase in nurses’ self-reported stopcock use and flushing of capped lumens, and (2) increase in the number of nurses who achieved mastery of knowledge and nurses’ total knowledge scores regarding frequency of selecting the appropriate lumen to flush and flushing lumens after procedures. Clinically significant findings were (1) increase in knowledge scores related to minimum amount of flush solution and (2) increase in the number of patent UVCs after the intervention.

Methicillin-Resistant Staphylococcus aureus and Vancomycin-Resistant Enterococcus Screening in the Intensive Care Unit

Schweon S, Zanders S, Roethke F. St Luke’s Hospital, Pa.

Purpose: Critically ill patients who acquire the antibiotic resistant pathogens methicillin-resistant Staphylococcus aureus (MRSA) or vancomycin-resistant enterococcus (VRE) during their hospitalization are at increased risk for infection, resulting in a greater length of stay, increased hospital resource utilization, and mortality. MRSA and VRE can be carried on the hands of healthcare workers from unidentified colonized patients. Background/Significance: During the past 4 years, an increasing number of community-acquired and healthcare-associated MRSA and VRE infections have occurred in the intensive care unit (ICU). Methods: The team decided to initiate surveillance cultures for MRSA and VRE for all ICU patients being admitted, regardless if they are being admitted from the emergency department, operating room, or another inpatient unit. The ICU staff and physicians were educated about the program. All patients upon admission are assessed for a history of MRSA or VRE. Patients that are colonized with either of these organisms are placed into contact isolation precautions. Patients who are not colonized then have samples from both the anterior nares cultured to assess for MRSA colonization and rectal samples cultured to assess for VRE colonization. Additional surveillance cultures are then obtained from the patients weekly. Culture samples are not obtained when the patient refuses or there is a physical limitation that prohibits the obtaining of a culture (eg, packing of nares after surgery). Results: During an 11-week period, 290 patients were screened for MRSA. Seven (2.4%) were newly discovered to be colonized. Three patients, initially testing negative upon admission, later tested positive within 3 weeks of their ICU stay. A total of 298 patients were screened for VRE. Fifteen (5%) were newly discovered to be colonized. Six patients, who initially tested negative upon admission, later tested positive within 3 weeks of their ICU stay. Conclusions: The multidisciplinary team caring for patients is notified when cross-transmission occurs and recommendations are given to improve practices.

Successfully Launching a Rapid Response Team by Using Quantitative Data to Prove Efficacy and Dispel Misconceptions

Linck J, Wilson J, Rock S, Henderson S. Park Nicollet Health Services, Minn.

Purpose: To evaluate the efficacy of the rapid response team (RRT) model, especially to dispel misconceptions and determine actual crisis intervention versus nuisance calls. Background/Significance: RRT is gaining widespread attention, but has proven challenging to implement as it requires significant cross-disciplinary staff support. At Methodist Hospital, a 426-bed facility, our RRT consists of a resident physician, respiratory therapist, critical care nurse, and the bedside nurse. Methods: An ongoing quantitative data analysis with 100% chart review of all RRT calls and collection of the following data: time, location and length of initial call; triggers for call; interventions within 3 to 5 hours of initial call; and short-term outcomes. Results: There were 70 calls in the program’s first 10 weeks. Sixty percent occurred on second shift (3 PM to 11 PM). Remaining calls were split evenly between first and third shifts. Most patients had multiple triggers for an RRT call, averaging 2.1. The top 3 triggers were respiratory changes (55.7%), neurological changes (40%), and blood pressure changes (32.9%). After the initial RRT call, a critical care nurse did follow-up visits at 2–4 hours and then again at 12 to 15 hours after the initial call. Interventions were tracked from the initial visit through the first follow-up. During this 3 to 5 hour period, patients received an average of 2.8 interventions. Eleven interventions emerged as significant, used in 10% or more cases. Also, 50% of RRT patients required transfer to higher level of care (ie, intensive care). Conclusions: Quantitative data dispelled the 2 most common staff misconceptions about the RRT program. First, the false perception that most RRT calls were nuisances or unnecessary; second, that most RRT calls occur on the day shift, interrupting the resident’s work when the primary doctor was only footsteps away. Finally, the data showed that our program was reaching its ultimate goal: to intervene sooner in a crisis and rescue patients earlier.

Progress Toward Meeting Backrest Elevation Standards in Patients Treated With Mechanical Ventilation

Grap M, Munro C, Sessler C, Russell S. Virginia Commonwealth University School of Nursing, Va.

Purpose: To identify progress toward meeting head-of-bed (HOB) elevation standards in patients receiving mechanical ventilation. Background/Significance: Recommendations from the Centers for Disease Control and Prevention (CDC) for prevention of nosocomial pneumonia in patients receiving mechanical ventilation include HOB elevations to 30° to 45°. Both the Joint Commission on the Accreditation of Healthcare Organizations (JCAHO) and the Institute for Healthcare Improvement include HOB elevation at greater than 30° as part of their pneumonia reduction recommendations. Studies published from 1994 to 2005 document HOB elevation in this population ranging from 19° to 23° without significant progress in meeting the above standards. Recently these recommendations have resulted in systematic documentation and evaluation of HOB elevation, which may increase compliance. Methods: HOB elevation was measured daily, using the bed’s electronic pad, over a 4-year period during a large clinical trial related to ventilator-associated pneumonia reduction. Data were obtained from subjects receiving mechanical ventilation during their first 7 days of intubation in 3 intensive care units (medical, surgical-trauma, neuroscience). Year by year comparisons of HOB elevation were made as well as comparisons between the years before and after implementation of systematic HOB documentation and evaluation (January 2004). Results: HOB elevation was measured in 323 patients over 1148 patient days. When analyzed by year, HOB elevation increased significantly over time, from 17.4° in 2002, 21.9° in 2003, 26.9° in 2004, to 28.2° in 2005. Whereas only 30% of HOB measurements met the standard in 2003, 56% met it in 2004 and 66% in 2005. By units, only the medical unit met the standard overall in 2005. Conclusions: Although recommendations are clear for minimal HOB elevation in patients treated with mechanical ventilation (30°), and HOB elevation is approaching this standard, the change in practice has been slow and is not yet at a complete level of compliance. Sponsored by: National Institutes of Health/National Institute of Nursing Research.

	RESEARCH ABSTRACTS

Top RESEARCH ORAL PRESENTATIONS RESEARCH ABSTRACTS

 
Job Satisfaction and Stressors Among Experienced Nurses in the Pediatric Intensive Care Unit

Shaw J, Huth M. Cincinnati Children’s Hospital Medical Center, Ohio.

Purpose: To explore job satisfaction and dissatisfaction of experienced staff nurses in a pediatric intensive care unit (PICU). Background/Significance: Job satisfaction surveys of all nurses in the PICU in 2003 indicated that nurses with greater than 3 years of experience had higher levels of dissatisfaction than less experienced nurses. Nurse’s role satisfaction may be a contributing factor in PICU retention. Growth in numbers of patients and the nursing shortage make retention of experienced PICU nurses a high priority. Methods: Herzberg’s (1959) 2-factor theory of employee satisfaction provided the conceptual foundation for the study. A convenience sample of 11 nurses with > 3 years of experience chose to participate. Focus groups, individual interviews, and demographic surveys were used to collect data. Content analysis was performed. Data were categorized and compared with the Herzberg model. Results: Participants were 26 to 60 years old. Fifty-five percent had 10 years or less of PICU experience. The major dissatisfier was relationships with physicians and other co-workers. Other dissatisfiers included low staffing, lack of supplies, and lack of financial recognition for longevity or extended roles. Motivating factors (satisfiers) included recognition of nursing excellence, the fulfilling nature of the work, and making a difference for children. There was no significant difference in work-related and non–work-related stress. Conclusions: It is important to address relationships between experienced nurses, physicians and co-workers in the PICU to decrease dissatisfaction. Further dialogue with nurses and physicians to seek understanding of relationship issues that are present and ideas for improved relationships will occur. Limitations of this study include a small number of participants, interviews that were not tape recorded, and interviews that were structured to elicit information about dissatisfaction more than satisfaction.

A Cumulative Analysis of Bispectral Index Monitoring in Critical Care

Olson D, Wong G, Bissett J, Dioguardi M, Kovitch L. Duke University Hospital, NC.

Purpose: The volume of literature discussing bispectral index (BIS, Aspect Medical Systems) exceeds 2000 peer-reviewed articles. The purpose of this project was to perform a systematic review of literature addressing validity and reliability of using physiological sedation assessment (BIS) in the intensive care unit (ICU). Background/Significance: Many ICUs now use BIS as an adjunct to routine observational forms of sedation assessment. Despite its multiple correlation studies with subjective assessment tools, authors have arrived at different conclusions on adopting BIS. It is erroneous to expect that physiological and observational methods of assessment would correlate higher than a pair of observational methods. Two different methods of assessing sedation are expected to correlate at lower levels than 2 similar methods. The significance of each correlation is a statistical query. Methods: A computer-aided literature search identified articles that were then reviewed for inclusion in this study. Articles included discussed psychometric evaluations of BIS and at least 1 sedation scale. Average weighted correlations were computed for each sedation score by converting r² values to standard correlation coefficients. Results: Of 2005 publications, 245 were ICU-related, 73 of these were peer-reviewed, and 22 met inclusion criteria for our study. There were 5 correlations of BIS to the Sedation Agitation Scale (SAS), 7 to Ramsay scale, 4 to COMFORT, and 2 to the Richmond Agitation Sedation Scale (RASS). Average weighted correlations were explored for analyses of BIS and SAS (r = .60), Ramsay (r = .58), COMFORT (r = .57) and RASS (r = .66). Conclusions: Despite a range of correlations, most authors found statistically significant correlations between BIS and one or more observational assessment tools. Future research into BIS use should focus on nursing decision making and clinical outcomes as a result of incorporating physiological data into current sedation assessment protocols.

Clinical Factors Associated With Agitation

Gardner K, Sessler C, Grap M. Virginia Commonwealth University Schools of Nursing and Medicine, Va.

Purpose: To examine the relationship of clinical, laboratory, and intervention characteristics of consecutive patients in the medical respiratory intensive care unit (MRICU) to the development of agitated behavior Background/Significance: Sixty percent of ICU patients experience agitated behavior ranging from apprehension or anxiety, to self-removal of indwelling tubes or frankly combative behavior that are potentially life-threatening. Identification of patients at particularly high risk for developing agitation would provide an opportunity to implement preventative strategies to protect patients from self-induced injury. However, there are few data to identify those at risk. Methods: Retrospective chart review of 83 subjects admitted to the MRICU in a 1-month period included medical/ medication history, admitting diagnosis, severity of illness, frequency of agitated behavior, and number of tubes and lines pulled. Data were collected for 274 patient days of ICU stay. Results: Subjects’ mean age was 50, and they were primarily male (55%), with an Acute Physiologic and Chronic Health Evaluation (APACHE) II score of 20.3, and an ICU length of stay of 5.9 days. Of the 83 subjects, 35 (42%) were agitated during at least 1 day of their ICU stay. Agitation was observed 32% of the time (n=86 days). APACHE II scores were significantly greater (23.8 vs 17.5; P = .002) in those subjects who experienced agitation during their ICU stay compared with those who did not. Daily multiorgan dysfunction scores (MODS) were also higher (8.2 vs 6.8; P = .002) on days when subjects were noted to exhibit agitated behavior. The pulmonary (P = .001), cardiovascular (P < .001) and neurological (P < .001) subscores of the MODS were higher in agitated subjects. No relationship was found between presence of agitation and age, sex, admitting diagnosis, history of psychiatric disorder or alcohol/drug abuse. Conclusions: Patients with greater levels of illness severity appear to have greater risk for agitation developing.

Determining Correlation Between Clinical and Electroencephalogram (EEG)-Based Sedation Assessment

Arbour R, Bucher L, Seckel M, Waterhouse J. Christiana Care Health System and the University of Delaware, Del.

Purpose: To evaluate correlation between clinical assessment of sedation using the Sedation-Agitation Scale (SAS) and an electroencephalogram (EEG)-based parameter, the bispectral index (BIS). Background/Significance: Oversedation masks neurological changes and increases mortality/morbidity. Undersedation risks prolonged stress mobilization and patient injury. In situations such as deep sedation/analgesia, an adjunct to clinical assessment of sedation using BIS may be useful in determining depth of sedation and navigating between sedation extremes. Determining correlation between clinical and EEG-based measures of sedation may validate BIS for use in practice. Methods: Intensive care patients meeting inclusion criteria were monitored using the SAS and BIS. Nurses assessing sedation level were blinded to BIS values. To generate paired measurements, data collectors initiated event markers with BIS at the time of SAS assessment. SAS values were extracted from medical records and corresponding BIS data from monitor downloads. Statistical analysis was performed on the paired observations to determine correlation. Results: Data were collected on 40 subjects, generating 209 data points. Moderate positive correlation between BIS and SAS values was shown with a Spearman rank coefficient (r-) value of .502 and an r² of .252 (P < .001). Strong positive correlation was noted between BIS and electromyography (EMG) with an r- value of .749 (P < .001). Conclusions: In situations where the clinical assessment is compromised or equivocal, BIS monitoring may have an adjunctive role in sedation assessment given the moderate positive correlation between BIS and SAS. BIS values should be interpreted with caution, however, because EMG activity remains a potentially significant confounding factor that may produce BIS elevations independent of level of sedation. More research is necessary to determine the optimal role of BIS monitoring in intensive care units. Sponsored by: AACN/Sigma Theta Tau Critical Care Research Grant.

Adopting Technology to Enhance Patient Care

Baker K, Bokovoy J, Soete R, Matchett S. Lehigh Valley Hospital & Health Network, Pa.

Purpose: To evaluate the impact of an intensive care information system (ICIS) on nurse workflow in a medical-surgical intensive care unit (ICU) and a trauma ICU. Background/Significance: Several studies have been done to evaluate the impact of an ICIS on nursing workflow. Using a combination of these earlier reported workflow sampling methods, our study evaluates ICU nurse workflow on all shifts, including weekdays and weekends at random times. Methods: We used workflow sampling methods to randomly evaluate frequency of nursing workflow activities on all shifts in 2 ICUs during a 2-week period before and after ICIS implementation. Workflow observations were done in January 2004 and in April 2005, and ICIS went live in June 2004. Activity types were defined using previous research. These activities were then grouped under the following 4 categories: (1) patient care, (2) documentation, (3) unit-related, and (4) personal. Results: At baseline, 1373 activities were observed and after the ICIS was implemented, 1249 activities were observed. Using ² analysis, the distribution of before and after values for all 4 categories were significantly different (P < .001). Unit-related, documentation and personal categories decreased 2.8%, 2.8% and 4.8%, respectively, whereas patient care increased by 10.4%. Extrapolating data to a theoretical 24-hour shift, percentages of observation were converted to minutes. Conclusions: The use of ICIS to support patient care in 2 ICUs improves nursing workflow. It reduces the time for documentation and increases time for direct patient care.

The Course of Delirium in Older Patients in Surgical Intensive Care Units

Balas M. University of Pennsylvania School of Nursing, Pa.

Purpose: To examine the frequency, course, and duration of delirium in older adults admitted to surgical intensive care units (SICUs). A secondary objective was to compare delirious and nondelirious subjects with respect to hospital and SICU length of stay (LOS), mortality rates, postdischarge institutionalization rates, and discharge functional and cognitive ability. Background/Significance: Delirium is one of the most frequent, dangerous, and costly complications associated with hospitalization in the older adult population. Prior studies on delirium, however, lacked data on the course of delirium in older adults admitted to SICUs. Methods: The sample included 114 subjects aged 65 and older admitted to the SICUs of a 672-bed university teaching hospital. Baseline characteristics were obtained through surrogate interviews and medical chart reviews. Subjects were screened for evidence of dementia and impairment in activities of daily living by using validated surrogate-rated instruments. Subjects were screened for delirium daily throughout their hospitalization. The subjects’ functional and cognitive ability and postdischarge placement were assessed within 24 hours of hospital discharge. Results: Eighteen percent of older adults were found to have evidence of dementia on admission to the SICU. This diagnosis was frequently unrecognized by both healthcare providers and the elders’ surrogates. While few elders (2.6%) were admitted to the SICU with evidence of pre-existing delirium, 28% had delirium develop during their SICU stay and 23% after their SICU stay. Delirious older adults had significantly longer hospital and SICU LOS, more frequent discharge to institutional care, and decreased functional ability at hospital discharge than did nondelirious elders. Conclusions: Delirium is a significant clinical and societal problem in need of further exploration and intervention. Future research is needed to elucidate the exact role delirium plays in the outcomes of older adults admitted to SICUs. Sponsored by: John A. Hartford Foundation Building Academic Geriatric Nursing Capacity Predoctoral Scholarship, AACN Educational Grant.

The Experience of Critically Ill Patients Receiving Neuromuscular Blocking Agents

Ballard N, Robley L, Barrett D, Fraser D, Mendoza I. Wellstar Health System, Ga.

Purpose: To understand the remembered experiences of persons who are given neuromuscular blocking agents and sedatives and/or analgesics to facilitate mechanical ventilation, improve hemodynamic stability, and improve oxygenation while in the critical care unit. Background/Significance: Neuromuscular blocking agents, used in the critically ill for therapeutic purposes, paralyze patients but leave them fully conscious. Sedatives and analgesics are necessary to reduce awareness, relieve fear, produce comfort, decrease anxiety, induce unconsciousness, and minimize latent complications such as posttraumatic stress syndrome. The extent to which patients experience awareness during therapeutic paralysis has been the subject of very little research. Methods: A phenomenological approach using in-depth interviews with 11 participants was employed. Data were analyzed by applying the constant comparative approach. Results: Four themes and 3 subthemes were identified from the data. The first theme was back and forth with a subtheme of having weird dreams. The second theme was loss of control, with subthemes of fighting/being tied down and being scared. The third theme was almost dying, and the fourth theme was feeling cared for. Conclusions: Patients can recall both negative and positive experiences during neuromuscular paralysis. Healthcare professionals are encouraged to search for improved assessment parameters, advocate for the development and use of sedation/analgesia guidelines and invest in quality improvement programs to assess for awareness during therapeutic paralysis and provide follow up and referral as necessary. Sponsored by: Sigma Theta Tau Mu Phi Chapter.

The Effects of Massage Therapy on Cardiac Transplant Recipients Following Coronary Angiography

Ballew C, Pocanic F, Potts W, Bergin J. University of Virginia Health System, Va.

Purpose: To understand how massage therapy influences pain, anxiety, and overall satisfaction in cardiac transplant recipients who undergo coronary angiography. Background/Significance: Cardiac transplant recipients commonly undergo routine coronary angiography to evaluate for graft vasculopathy. Studies suggest anxiety is commonly expressed in patients during coronary angiography. Because no published studies about transplant recipients quantify anxiety, pain, and overall satisfaction during the procedure, we evaluated the effects of massage therapy in this unique population. Methods: After receiving approval from the human investigation committee, we approached 40 cardiac transplant recipients who were scheduled for routine coronary angiography. After consent was obtained, we randomized subjects into a treatment group (massage) or control group (no massage). Within 40 minutes of returning to the recovery area, all subjects were asked to rate their pain and overall satisfaction with the procedure. Subjects were also asked to complete the State Trait Anxiety Inventory (STAI). Next, subjects randomized into the treatment group received a 20-minute massage from a licensed massage therapist. Subjects in the control group received no massage. Forty minutes after completing the baseline questionnaires, all subjects were asked to again rate their pain and satisfaction and to complete an STAI. A ² and Fisher exact test were used to analyze the data. Results: No demographic differences between the groups were found. Subjects in the treatment group had a significant reduction in anxiety scores (P .05) and a trend toward pain reduction (P = .07). Overall satisfaction with their angiography experience did not differ between the groups (P = .3). Conclusions: It appears that massage therapy in this population helps reduce patients’ anxiety. These results validate the need for more research in the area of understanding patients’ coronary angiography experience.

Factors Associated With Prolonged Prehospital Delay Time in African Americans With an Acute Myocardial Infarction

Banks A, Dracup K. University of San Francisco and University of California, San Francisco, Calif.

Purpose: To characterize the symptom experience of African Americans after an acute myocardial infarction (AMI). Background/Significance: Research to date has shown prolonged delays from onset of symptoms to arrival at the hospital in African Americans experiencing an AMI. Methods: African Americans (N = 61) diagnosed with an AMI were interviewed on average 2.6 (SD 1.1) days after hospital admission. Patients were aged 60 (SD 12) years on average and 52% (n = 32) were women. Patients were interviewed using a structured interview. Delay times were calculated from patient interview. Results: Median delay time was 4.3 hours and did not differ significantly between women and men (4.4 hours vs 3.5 hours). Most patients (69%) experienced their initial symptoms at home with the most common witness being a family member (36%). The ambulance was the most common means of transportation to the hospital (59%) with higher use among women (52.8%) than men (47.2%). Three factors were associated with increased delay times: insurance status, marital status, and diabetes. Median delay time for insured patients was longer than that for uninsured patients (4.5 vs 0.5 hours, P = .03). Single patients had longer median delay times than did married patients (5.3 vs 2.5 hours, P = .04), and patients with diabetes had longer median delay times than nondiabetics (7.3 hours vs 3.5 hours, P = .02). Conclusions: In this sample of African American AMI patients, median delay times were substantially longer than the recommended time of less than 1 hour, making it difficult for most patients to benefit fully from reperfusion therapies. Public education and counseling of patients and their families must be a major strategy in optimizing patients’ outcomes and decreasing the time to definitive treatment. Sponsored by: National Institute for General Medical Sciences.

Temperature Measurement in Critically Ill Adults

Lawson L, Bridges E, Ballou I, Eraker R, Greco S, Shively J, Sochulak V. University of Washington Medical Center, Wash.

Purpose: To describe the accuracy and precision of oral (OR), tympanic (TYMP), temporal artery (TMPL), and axillary (AX) temperature measurements compared with pulmonary artery (PA) temperature in critically ill adults. Background/Significance: Research is equivocal on the accuracy/precision of noninvasive temperature methods in critically ill adults, particularly for TMPL. Methods: Repeated measures design. Inclusion: Patients requiring a PA catheter. Exclusion: oral inflammation/ trauma, tympanic membrane not intact/not visualized by otoscopy. Measurements were taken by trained ICU nurses using calibrated thermometers and manufacturer recommendations: OR/AX (SureTemp Plus, Welch Allyn), TYMP (Genius 3000A, Sherwood), TMPL (TemporalScanner, Exergen), and PA (VIP Swan-Ganz Catheter, Edwards). Temperatures were taken in sequence within 1 minute on the same side of the body, and were repeated 3 times at 20-minute intervals. Accuracy, precision and confidence limits (CL) were analyzed. Results: 60 critically ill patients (20 females, 40 males, 57 [SD 15] years) with cardiopulmonary disease (medical n = 33; surgical n = 27) were studied. PA temperature range: 35.3–39.4°C. Mean (SD) offset and CL differences were: PA-ORAL (0.09 [0.43]°C; CL = –0.75, 0.93), PA-TYMP (–0.36 [0.56]°C; CL = –1.46, 0.74), PA-TMPL (–0.02 [0.47]°C; CL –0.92, 0.88) and PA-AX (0.23 [0.44]°C; CL = –0.64, 1.12). Percentage of pairs with differences greater than 0.5°C: PA-ORAL 19%, PA-TYMP 49%, PA-TMPL 23%, PA-AX 27%. There was 1 outlier for TMPL (PA > TMPL 2.6–2.9°C). The subject was febrile and diaphoretic with a fan blowing. Intubation significantly increased oral temperatures (OR > PA 0.3 ± 0.3°C). Conclusions: The OR and TMPL were most accurate. The AX underestimated PA temperature. The TYMP was least accurate and had the greatest variability. Intubation affected the accuracy of OR measurements, and diaphoresis/airflow across the face may affect TMPL. Further study is needed for the TMPL in patients with fever, sepsis, or trauma and with oxygen delivery via face mask/tent.

Development of Perineal Dermatitis in Critically Ill Adults With Fecal Incontinence

Ehman S, Thorson M, Lebak K, Bliss D, Savik K, Beilman G. University of Minnesota School of Nursing and School of Medicine, Minn.

Purpose: To describe the onset and severity of incontinence associated dermatitis (IAD) in critically ill (ICU) adults with fecal incontinence. Background/Significance: Incontinent ICU patients are at high risk for IAD. IAD is manifested by redness when less severe and skin loss and secondary infections when more severe. Little is reported about the incidence and characteristics of IAD in ICU patients. Methods: Patients in surgical/trauma critical care units at 3 urban hospitals who were incontinent of feces and free of perineal skin damage were recruited for prospective observation of their perineal skin. The hospitals did not have a protocol for IAD prevention/management. Daily visual inspection and DermaSpectrometry of perineal skin was done by the investigators. Nurses reported characteristics of feces and incontinence each shift on a bedside record. Surveillance continued until participants transferred to another unit, received a rectal pouch/tube, or died. Results: 15 patients from each hospital were enrolled (n = 45, 34 male); 35 white, 8 African American, and 2 Asian. Patients’ mean age was 49 (SD 19) years. Diagnoses were multitrauma=28, traumatic brain injury = 6, cerebrovascular accident = 6, abdominal aortic aneurysm = 2, and sepsis = 1. Mean score at admission on the Acute Physiology and Chronic Health Evaluation (APACHE) II was 15 (SD 5). Patients were receiving mechanical ventilation a mean of 66% of observed time. Length of surveillance was 7 (SD 4) days. Of 514 stools, 98% were incontinent and 68% of these were loose/liquid. IAD developed in 35.5% of patients. Time to onset of IAD was 4 (range 1–6) days. Severity of perineal skin redness was mild, moderate, or severe for a mean of 10%, 7%, or 3% of the observed time. Partial tissue loss occurred 7% of the observed time and fungal infections 3.5% of the time. DermaSpectrometer measures and investigator ratings of skin redness were significantly correlated (Spearman’s = 0.262, P < .001). Conclusions: IAD is a problem in ICU patients. Its onset is rapid, and its severity is usually mild to moderate. An IAD prevention protocol is recommended. Sponsored by: Greater Twin Cities Area Chapter of the American Association of Critical-Care Nurses.

Stroke Knowledge of Patients With Atrial Fibrillation

Fowler S, Ruh D. Somerset Medical Center, NJ.

Purpose: To answer the following question: What is the stroke knowledge of patients with atrial fibrillation and is there a sex or age difference in this knowledge? Background/Significance: Atrial fibrillation (AF) is the most common sustained arrhythmia encountered in clinical practice. In the United States, about 2.2 million people have AF. Its incidence increases with age and the presence of structural heart disease. It is a major cause of stroke, especially in the elderly. On average, about 700 000 Americans have a stroke each year, and 15% of strokes are related to AF. Assessing patients’ knowledge can help healthcare professionals in planning measures directed at prevention. Methods: Patients (N = 50)are being interviewed prospectively during hospitalization if they meet the following criteria: (1) 18 years of age and older, (2) atrial fibrillation in admitting diagnosis, (3) alert and oriented at time of interview, (4) medically stable at the time of the interview, (5) English speaking, and (6) able to understand informed consent and provide consent. Patients complete a demographic form and risk factor and stroke knowledge questionnaires (Test Your Stroke IQ). Results: Frequencies and percentages for demographic and questionnaire data and mean/range, when appropriate, will be calculated. Differences in questionnaire scores based on demographic variables (eg, sex, age) will be detected with a t test. Conclusions: Prevention of stroke and management of complications associated with atrial fibrillation targets one of AACN’s research priority areas: prevention and management of complications. If critical and acute care nurses are to make their optimal contribution to a healthcare system driven by the needs of patients, we must first understand those needs by asking our atrial fibrillation patients about their stroke knowledge.

The Relationship of Light, Sound, and Rest Periods on Sedation, Physiological Stability, and Movement in Patients Receiving Mechanical Ventilation

Mellott K, Grap M, Hamilton A, Best A, Wetzel P, Munro S, Sessler C. Virginia Commonwealth University Schools of Nursing, Medicine and Engineering, Va.

Purpose: To describe the level of light and sound for patients treated with mechanical ventilation during and outside of unit rest periods and the effect of rest periods on sedation, physiological stability, and movement. Background/Significance: The U.S. Environmental Protection Agency recommends hospital sound levels of less than 45 decibels (dB, daytime) and less than 35 dB (nighttime). Sound levels in critical care units average 50 to 60 dB and frequently peak above 80 dB. Recommendations for lighting in hospitals range from 10 to 20 foot-candles (fc). Excessive stimuli may impede recovery and present a barrier to effective sleep. Sedation may be used to blunt environmental stimulation, potentially resulting in oversedation. Rest periods with diminished lighting and sounds are used theoretically to improve patients’ rest. Methods: Fourteen patients receiving mechanical ventilation (73% female; mean age 55, mean score of 27 on the Acute Physiology and Chronic Health Evaluation II) were studied. All measurements were obtained every 15 seconds. Sound and light were measured by calibrated meters positioned at the head of the bed. Patient State Index (PSI), a processed EEG, was used to determine sedation level, physiological stability consisted of heart rate and respiratory rate, and patients’ movement was measured with arm/leg actigraphy. Results: A total of 38 165 samples were obtained, resulting in 159 hours of data. The mean light was 5.3fc (SD 4.25). Mean sound was 70.6 dB (range 61–82). During unit rest periods, sound was significantly lower (P < .001), and subjects’ heart and respiratory rates were lower than during nonrest periods (P < .001). Sedation level was lighter during the rest period, and there was no difference in light or patients’ movement. Conclusions: Although sound levels are above those recommended, they were reduced by use of a rest period and some positive effects on patients were noted. Although significant differences were found, clinical differences in PSI, heart rate, and respiratory rate were small. Further study of the efficacy of rest periods is warranted. Sponsored by: Physiometrix, Inc.

Physiological Stability and Patients’ Comfort During Varying Levels of Sedation in Adults Receiving Mechanical Ventilation

Grap M, Sessler C, Best A, Wezel P, Hamilton A, Mellott K, Munro C. Virginia Commonwealth University Schools of Nursing, Medicine and Engineering, Va.

Purpose: To examine the effect of level of sedation on sedation outcomes of physiological stability and comfort. Background/Significance: 85% of ICU patients receive sedation to help attenuate the anxiety, pain, and agitation associated with mechanical ventilation. Inappropriately high levels of sedation prolong duration of mechanical ventilation and increase ventilator-associated pneumonia, whereas low levels of sedation place patients at risk for unplanned (self) extubation, hemodynamic instability, and physical injury. Sedation scales are used to assess sedation level; however, the extent to which various levels of sedation actually achieve the goals of physiological stability and comfort is unknown. Methods: Fourteen subjects in the medical respiratory ICU were continuously monitored and data were recorded every 15 seconds for a total of 159 patient-hours. Sedation level was measured by using the Patient State Index (PSI; processed electroencephalogram), physiological stability was documented by using heart rate (HR) and respiratory rate (RR), and comfort was evaluated by using arm and leg actigraphy, a method to detect patients’ movement. All data were downloaded for analysis and the percent of time outside normal range for HR, RR, and actigraphy were evaluated. Sedation level was categorized as deep (PSI < 60), mild/moderate (PSI 60–80) or awake/alert (PSI > 80). Results: Subjects were predominantly female (73%), with a mean age of 55 years, and admitted for acute respiratory failure. Subjects were physiologically unstable (either HR or RR outside of normal limits) 63% of the time during deep sedation, 56% of the time during mild/moderate sedation, and 56% of the time when alert. The percentage of time patients were moving, which may indicate discomfort, was 2% during deep sedation, 7% during mild/ moderate sedation, and 13% while alert. Conclusions: Although patients’ movement increases as expected with less sedation, physiological stability is not being achieved even with deep levels of sedation. The present methods of sedation evaluation may not be adequate for assessing all domains of sedation efficacy. Sponsored by: Physiometrix Inc.

Joint Professional Analysis of a Novel Proton Pump Inhibitor (PPI) Dosage Form

Haak S, Nickman N, Rusho W, Beck S, Lafleur J, Brixner D. University of Utah Colleges of Nursing & Pharmacy, Utah.

Purpose: A collaborative evaluation of a novel proton pump inhibitor (PPI) dosage form (Prevacid SoluTab, lansoprazole orally disintegrating tablet, or LODT) conducted by nursing and pharmacy faculty used 3 approaches to analyze pharmacy/nursing tasks and labor/supply costs. Background/Significance: Patients requiring acid-suppressing therapy need special dosage forms to accommodate nasogastric (NG) tubes and difficulty swallowing. An alternative to current intravenous (IV) and oral forms could improve patients’ care and require less preparation and administration time. Methods: A survey of nurses and pharmacists in 5 practice settings assessed usage experience and tasks required to prepare and administer PPIs. Results from 48 surveys were used to develop a simulated Time-and-Motion (TM) analysis of tasks and costs related to 5 dosage scenarios. Results: A TM analysis of videotaped pharmacy staff preparing/dispensing doses and nurses preparing/ administering doses to Lifeform adult manikins indicated IVs (Protonix IV, pantoprazole, Wyeth) were the most time intensive for pharmacy (P < .001) but the least time intensive for nursing (P < .001). For nursing, NG administration of both lansoprazole capsule granules and LODT required more time than oral capsules in applesauce or oral LODT, but LODT administration was significantly faster regardless of route. Cost minimization analysis indicated the LODT dosage form offered potentially significant cost savings. Savings for LODT over IV administration were estimated at $71.57/dose, LODT over oral capsule at $2.20/dose, and LODT NG over capsule NG at $2.32/dose. The LODT dosage form also appeared to be easier to administer both orally and by the NG route. Conclusions: Cost savings and ease of use associated with this novel PPI dosage form may provide a compelling argument for its use in patients with special PPI needs. Sponsored by: TAP Pharmaceutical Products, Inc.

Averting High-Risk Dosing Errors Associated With Infusion Pump Programming

Jacobs B. George Washington University Hospital, DC.

Purpose: To determine the incidence of averted IV medication errors resulting from utilization of dosing limits during intravenous infusion pump delivery. Background/Significance: Of all the medication errors that result in significant harm, 35% are the result of infusion pump errors, with the most common error being incorrect programming of the infusion parameter into the pump. Methods: A total of 550 infusion pumps with dose-limiting technology were put in use throughout all departments of George Washington University Hospital. After a study period of 5 months, data logs were collected from a random sample of 150 pumps. The data logs were analyzed for incidence of dose alerts and the medications involved. Results: Among 42,837 programmed doses, 122 dose alerts were identified. Twenty-seven doses (22%) were programmed below the allowable dose limit, and 95 doses (78%) were programmed above the allowable dose limit. Of the 122 dose alerts, we felt that 7 represented potentially high-risk dosing errors that could have led to a significant patient event. Medications associated with these 7 "near-miss" events were vasopressin, milrinone, nesiritide, fentanyl, heparin, and oxytocin. Conclusions: Dose-limiting technology enabled our hospital to improve patients’ safety by averting potential high-risk medication errors. Dose alerts served as a double-check system for our clinicians when dosing IV medications. In addition, our study helped us identify the need for staff education on the dosing guidelines for particular drugs. Further data collection over a longer period of time is warranted to determine if there is a relationship between incidence of dose alerts and time of day, time of year, and the number of critical IV medications infusing simultaneously.

Natural History of Dental Plaque Accumulation in Mechanically Ventilated Adults

Jones D, Munro C, Grap M. Virginia Commonwealth University School of Nursing, Va.

Purpose: To describe the pattern of dental plaque accumulation in adults receiving mechanical ventilation. Background/Significance: Dental plaque may serve as a reservoir in critically ill patients for potentially pathogenic microorganisms. Accumulation of dental plaque and bacterial colonization of the oropharynx are associated with a number of systemic diseases including ventilator-associated pneumonia (VAP). Understanding the accumulation pattern of plaque in intubated patients is essential to the development of an effective standardized oral care program. Methods: Data were collected from 66 critically ill adults who were control subjects in a large clinical trial related to oral care intervention. Subjects were enrolled within 24 hours of intubation; dental plaque was assessed on study days 1, 3, 5 and 7. Dental plaque was measured by using the University of Mississippi Oral Hygiene Index; each tooth was divided into 10 standardized sections, and every section of every tooth was scored (plaque present or absent). A fluorescein plaque-disclosing rinse, visible only in ultraviolet light, was used to enhance discrimination of plaque. Results: The sample demonstrated diversity in race and sex. The average length of intubation was 7 days. The mean number of decayed, missing, or filled teeth was 7.7. Plaque: All surfaces had greater than 50% plaque coverage from day 1 to day 7; lingual surfaces contained the greatest plaque average on all days (>70%); molars and premolars contained the greatest plaque (mean > 70%). Systemic antibiotic use on day 1 had no significant effect on plaque accumulation on day 3 (P=.73). Conclusions: Patients arrive in critical care units with preexisting oral hygiene issues that vary over their stay and theoretically increase their vulnerability to systemic disease. Knowing accumulation trends of plaque will guide the development of effective oral care protocols. Such protocols may decrease length of stay, length of intubation, and hospital costs in critically ill patients. Sponsored by: National Institutes of Health.

End-of-Life Care: Can We Predict to Prepare?

Johnson R, Granger B, Bride W. Duke University Health System, NC.

Purpose: To assist nurses in identifying demographics that correlate with comfort care at end of life. Background/Significance: We questioned if there was a method to predict when to change the focus of patient care from curative to comfort, thereby enabling staff to better anticipate and tailor end-of-life care. Methods: Demographics were kept on 996 patients who died during the past 8 years in a 16-bed intensive care unit (ICU). The patients were divided into 3 categories: full code (FC), do not resuscitate (DNR), and active withdrawal of life support (AW). Results: The mean age of death was 67 years (median 68), with 63% of deaths occurring in patients more than 65 years of age. Regardless of the level of care delivered, FC, DNR, or AW, more men than women died (53% vs 47%, P < .001). The only outlier was the category of brain death with a total of 6 men and 8 women. Most patients died with an endotracheal tube (ETT) in place (53%). The majority with an ETT were FC (81% died with an ETT). By contrast, only 22% of AW patients had an ETT at the time of death (indicating that the healthcare team was successful in adjusting the focus of care for this subgroup.) The majority of patients did not have a pulmonary artery catheter (PAC) in place. For FC 89%, for DNR 82% and for AW 95% did not have a PAC. The time of day that deaths occurred was distributed relatively evenly: morning 31.4%, evening 36.3%, and night 32.4%. Conclusions: This preliminary information has prompted us to take the next step and consider additional analysis to more clearly identify criteria that would enhance our ability to determine the most appropriate level of care for individual patients. Length of ICU stay and cause of death will be examined to ascertain if patients’ outcomes can be more accurately predicted. With this additional information, we hope to be able to foresee patients’ outcomes earlier, allowing us to care for patients as they should be cared for at the end of life—with comfort, dignity, and compassion.

How Low Should You Go? Nursing Practice of Aggressive Blood Glucose Management Following Cardiovascular Surgery

Kinney T, Frank A, Steinberg K. Mayo Foundation, Minn.

Purpose: To describe the role of intensive care unit (ICU) nurses in the implementation of an aggressive blood glucose management protocol following adult cardiovascular surgery. Background/ Significance: Intense review of the background of hyperglycemia, along with the conventional postoperative hyperglycemic nursing protocol of the adult cardiac surgical population was implemented. The support of evidence-based literature summarizing strict glycemic control displays its significance. Methods: A retrospective observational study was completed on 409 consecutive adult patients undergoing cardiac surgery at Mayo Clinic Rochester between June 2002 and August 2002, looking at how much perioperative hyperglycemia affected patients’ outcomes. This study lead to a prospective insulin study comparing outcomes, length of stay (LOS), and glucose control with 2 strategies: intensive therapy of continuous intravenous infusion of insulin to maintain blood glucose at 4.4 to 5.6 mmol/L (80–100 mg/dL), and conventional glucose management practiced by the ICU nurse. Results: Results summarized the insulin study and its effects on primary and secondary outcomes up to 30 days postoperatively. Outcomes compared included prolonged intubation, stroke, heart block requiring pacemaker, new-onset atrial fibrillation, acute renal failure, cardiac arrest, hospital LOS, ICU LOS, and infection. Conclusions: The intensive perioperative insulin protocol had no significant change on outcomes when compared with the conventional postoperative protocol. Although no significant change was demonstrated, the importance of aggressive nursing management of postoperative hyperglycemia continues to be a relevant nursing practice.

Replace the Glasgow Coma Scale? Validation of a New Coma Scale: The FOUR Score

Miers A, Pfrimmer D, Gusa D. Mayo Clinic-Saint Marys Hospital, Minn.

Purpose: To validate a new coma scale, the Full Outline of Un-Responsiveness (FOUR) score. Background/Significance: The Glasgow Coma Scale (GCS) is widely used even though the scale cannot be used to assess the verbal score in an intubated patient or to test brainstem reflexes. A new coma scale, the FOUR score, was recently devised. It consists of evaluation of 4 components (eye, motor, brainstem, and respiration), and each component has a maximum score of 4. Methods: An interdisciplinary, prospective study was undertaken to determine tool validity. The FOUR score was tested on 120 intensive care unit patients and compared with the GCS score by interrater pairs of neuroscience nurses, neurology residents, and neurointensivists. Results: The overall interater reliability was .82, which was the same as the reliability of the GCS (.82). The FOUR score provided greater neurological detail than the GCS, allowed recognition of "locked-in syndrome," and was superior to the GCS because it allowed scoring of the brainstem reflexes, breathing patterns, and allowed different states of herniation to be recognized. Evidence also indicated that the probablility of in-hospital mortality was higher for the lowest total FOUR scores when compared with the lowest total GCS scores. Conclusions: The FOUR score is a valid and reliable assessment tool to evaluate coma status of patients in the intensive care unit. It shows superiority in several areas over the GSC. Further testing is planned to test the tool in the emergency unit and pediatrics. Details of the validation process and subsequent testing will be presented.

Description of Enteral Feedings in Adults Receiving Mechanical Ventilation

Munro C, Grap M, Sessler C, Russell S. Virginia Commonwealth University, Va.

Purpose: To describe enteral feedings in critically ill adults during the first 7 days of mechanical ventilation. Background/ Significance: Nutritional support through enteral feeding is an important aspect of the care of critically ill patients. Because enteral feeding improves clinical outcomes, reduces length of stay, and reduces cost, current recommendations suggest that nutritional support should begin within 24 to 48 hours of admission to the intensive care unit (ICU). However, enteral feeding has been identified as a risk factor for ventilator-associated pneumonia (VAP), particularly when associated with large residual volumes or alkaline pH of gastric secretions. Many questions related to optimal feeding strategies remain. Methods: The sample consisted of 296 critically ill adults who were subjects in a large clinical trial related to reduction of VAP. Enteral feeding data were obtained from subjects receiving mechanical ventilation during their first 7 days of intubation in 3 ICUs (medical, surgical-trauma, neuroscience), resulting in 1147 observation days. Results: Percentages of subjects receiving enteral feedings increased over time (38.0% day 1, 55.4% day 2, 66.3% day 3, 72.1% day 4, 77.5% day 5, 78.5 day 6, 78.7% day 7). Most feedings were delivered via nasogastric or orogastric tube (86%) and by gastric route (89.4%) rather than small bowel (9.6%). Continuous delivery was more common (84.4%) than bolus feedings (10.4%). Residual volumes greater than 0 were present in 63.3% of observations. Of observations with residual volume, mean residual volume was 107.5 mL (SD 132.5), and 21.0% were greater than 150 mL. Recording of pH was not routine (recorded on only 25.4% of observation days); pH was greater than 7 in 9.6% of documented observations. Conclusions: Enteral nutrition support was initiated in most patients by the second day of intubation . The relationships of enteral feeding variables and outcomes in patients receiving mechanical ventilation require further research. Sponsored by: National Institutes of Health.

Bereavement After-Care Project

Parker L, Clark S, Patterson C, Ulrich L, Caldwell M. Charles F. Kettering Memorial Hospital, Ohio.

Purpose: The current practice in critical care units (CCUs) is that "end-of-life care" ends when the patient dies. The nursing staff is focused on saving the patient’s life rather than supporting the families’ bereavement process. With the acuity and morbidity of the patients in the CCU increasing, nurses are challenged to assess the steps in the bereavement process. Background/ Significance: A caring program was developed to address the needs of the bereaved family. This program will help guide them through the process without abruptly ending the ties with the nursing staff at the time of death. Methods: After a literature review, a tool was identified to obtain data from the patient’s families regarding their needs and formulate their experiences to improve the care given. The survey was created to establish a follow-up program. During the stay in the CCU, a relationship is formed with staff. Information regarding the support groups and community resources are provided. Verbal permission is obtained to continue contact. Communication via notes and telephone calls to the bereaved person(s) is scheduled. Within the first year, a questionnarie and survey regarding the end-of-life care along with the grieving process will be collected and the results will be shared with the CCU staff. Results: When we give support to the community in regards to education and resources, the grieving process has less of an impact. Grieving persons reestablish ties within their communities, families, and jobs at a faster pace. Children are quickly acclimated into their social and academic environment. CCU nurses have a better rapport, knowing that our family-centered support system is intact and ongoing after the grieving person leaves the CCU. Conclusions: The survey results from the families will provide information that will be used to assess the bereavement program. Adjustments will be made to the program as needed. Continued improvement during the second stage of bereavement is expected.

Development of a Scale to Measure the Risk of Skin Breakdown in Critically Ill Patients

Rose P, Cohen R, Amsel R. McGill University Health Centre and McGill University, International.

Purpose: To develop a scale that assesses the risk of skin breakdown in critically ill patients and to test its feasibility, reliability, and validity. Background/Significance: Critically ill patients are at risk of developing skin breakdown. Scales currently used to measure this risk do not include factors specific to critical illness. Scales that have been explicitly developed for use with this population demonstrate serious limitations. Methods: Generation and preliminary reduction of scale items was done through literature review, clinical observations, and 5 focus groups of staff. The 36-item scale was tested on a consecutive sample of 111 patients admitted to the intensive care unit (ICU) of a university teaching center. Data collected included baseline chart data, assessment of risk of skin breakdown using both the Braden scale and the 36-item scale, and assessment of skin integrity. Risk and skin assessments were independently rated so that data collectors remained blind to the other measures. Assessments were performed within 48 hours of admission and every 2 days until day 8. Data collection ceased if a pressure ulcer developed or the patient died, went home, or was transferred to another institution. Results: Inter-rater reliability, estimated by calculating percentage agreement on 10% of the sample, was 0.93. Content and face validity of the scale were supported by reliance on scientific literature, expert opinions, and clinical observations to develop items. Comparison of sensitivity and specificity of the 36-item scale and the Braden Scale provided an estimate of concurrent validity. Multiple regressions determined which group of items best predicts pressure ulcers: skin quality, restricted movement, and temperature. This 3-item scale demonstrates better sensitivity and specificity than the Braden Scale. Conclusions: This research suggests a practical 3-item scale that provides clinicians with a reliable and valid tool to identify those critically ill patients at risk for skin breakdown. Sponsored by: Newton Foundation, Canadian Nurses’ Foundation, and the Foundations of the Montreal General Hospital, Royal Victoria Hospital and Montreal Childrens’ Hospital.

Symptom Clusters in Acute Myocardial Infarction

Ryan C, Devon H, Horne R, King K, Milner K, Moser D, Quinn J, Rosenfeld A, Zerwic J. University of Illinois at Chicago, Ill.

Purpose: To identify common clusters of symptoms in a population of people with acute myocardial infarction (AMI) and to compare symptom clusters across demographic groups. Background/Significance: Previous studies identifying AMI symptoms have indicated that people do not experience single symptoms but rather a constellation of symptoms. However, clusters of symptoms have yet to be identified. Methods: The sample for this study consisted of AMI symptom data obtained from 9 different researchers (1073 subjects) in the US and the UK. The data were originally collected for purposes other than clustering during face-to-face interviews with persons who had recently experienced AMI. Results: Twelve common symptoms and 5 clusters were identified. Age (P < .001), sex (P</