A Nursing Clinical Decision Support System and Potential Predictors of Head-of-Bed Position for Patients Receiving Mechanical Ventilation

Frank Lyerla, Cynthia LeRouge, Dorothy A. Cooke, Debra Turpin and Lisa Wilson
Am J Crit Care 2010;19:39-47 doi: 10.4037/ajcc2010836
2010 American Association of Critical-Care Nurses Published online http://www.ajcconline.org Personal use only. For copyright permission information: http://ajcc.aacnjournals.org/cgi/external_ref?link_type=PERMISSIONDIRECT

Subscription Information http://ajcc.aacnjournals.org/subscriptions/ Information for authors http://ajcc.aacnjournals.org/misc/ifora.shtml Submit a manuscript http://www.editorialmanager.com/ajcc Email alerts http://ajcc.aacnjournals.org/subscriptions/etoc.shtml

AJCC, the American Journal of Critical Care, is the official peer-reviewed research journal of the American Association of Critical-Care Nurses (AACN), published bimonthly by The InnoVision Group, 101 Columbia, Aliso Viejo, CA 92656. Telephone: (800) 899-1712, (949) 362-2050, ext. 532. Fax: (949) 362-2049. Copyright 2010 by AACN. All rights reserved.

Downloaded from ajcc.aacnjournals.org by guest on February 8, 2011

Pulmonary Critical Care

CLINICAL DECISION SUPPORT SYSTEM AND POTENTIAL PREDICTORS OF HEAD-OF-BED POSITION FOR PATIENTS RECEIVING MECHANICAL VENTILATION
URSING
By Frank Lyerla, RN, PhD, Cynthia LeRouge, PhD, CPA, Dorothy A. Cooke, RN, PhD, Debra Turpin, RN, MSN, ONC, and Lisa Wilson, RN, BSN
Background Patients receiving mechanical ventilation are at high risk for pneumonia due to aspiration. Published guidelines recommend elevating the head of the bed 30 to 45, if not contraindicated, to reduce risk, but this intervention is underused. Objectives To facilitate incorporating evidence-based practice by improving positioning of patients receiving mechanical ventilation and to identify patient and nurse characteristics that predict use of the guideline. Methods A modified interrupted time-series design was used. Data were collected on 43 patients and 33 nurses 3 separate times in a 12-bed intensive care unit at a medium-sized hospital. A total of 105 observations were recorded for analysis each time. Results Mean elevations of the head of the bed increased significantly from phase 1 (27.7) to phase 2 (31.7) and from phase 1 to phase 3 (31.1). Elevations were higher for tube-fed patients than for patients not given enteral tube feedings. Elevations were higher for patients with a pulmonary-related diagnosis and lower for patients with a gastrointestinal diagnosis than for patients with other diagnoses. Elevations were lower for patients with a body mass index between 25.0 and 29.9 (overweight) than for patients with other body mass index values. Nurse characteristics were not significant predictors of elevation. Conclusion A nursing clinical decision support system integrated into a patients electronic flow sheet can increase nurses adherence to guidelines. Pulmonary and gastrointestinal diagnoses, body mass index, and tube feeding are predictors of elevation of the head of the bed. (American Journal of Critical Care. 2010;19:39-47)

AN

2010 American Association of Critical-Care Nurses doi: 10.4037/ajcc2010836

www.ajcconline.org

AJCC AMERICAN JOURNAL OF CRITICAL CARE, January 2010, Volume 19, No. 1 Downloaded from ajcc.aacnjournals.org by guest on February 8, 2011

39

entilator-associated pneumonia (VAP) is defined as a lung infection that occurs after at least 48 hours of intubation and mechanical ventilation.1 Among patients receiving mechanical ventilation, the rate of VAP is 8% to 28%.2 VAP accounts for more than $40 000 in mean hospital charges per patient.3 Tracheobronchial aspiration of oropharyngeal secretions and gastric contents has been cited as a cause of VAP.4,5 Additionally, lower head-of-bed (HOB) elevations have been associated with higher rates of aspiration.6-11 Despite the evidence that HOB elevation (30-45) helps prevent aspiration in patients receiving mechanical ventilation, the intervention is underused.11-13

We developed and implemented a nursing clinical decision support system (CDSS) in the form of an electronic reminder. Our primary purpose was to determine the effect of a nursing CDSS on the use of evidence-based practice in improving the positioning of patients receiving mechanical ventilation. Our secondary purpose was to determine the effect of potential predictors of HOB elevation. Potential predictors included time of day, patients body mass index (BMI; calculated as weight in kilograms divided by height in meters squared), whether a patient was tube-fed, patients diagnoses, and nurses length of clinical experience.

Review of Relevant Literature

Expert panels have published guidelines on the positioning of patients receiving mechanical ventilation. The guidelines are based on evidence that HOB elevation reduces occurrence of aspiration and subsequent pneumonia. The Centers for Disease Control and Prevention5 recommend elevating the HOB to 30 to 45 to prevent aspiration in patients receiving mechanical ventilation. The same recommendation is made by the Society for Parenteral and Enteral Nutrition Consensus Conference on Aspiration in Critically Ill Patients,14 the Institute for

Clinical decision support systems can be used effectively in the clinical setting.

Healthcare Improvement,15 and the American Association of Critical-Care Nurses (AACN).16 The Joint Commission17 and the Canadian Critical Care Trials Group18 recommend a minimum HOB elevation of 30 for patients receiving mechanical ventilation. A CDSS is a computerized program that provides clinical knowledge and patient-related information, intelligently filtered or presented at appropriate times, to enhance patient care (modified from Osheroff et al).19 Design and implementation of a CDSS are complex and involve many variables, including function, user, setting, and desired outcome. Evidence20-23 indicates that clinical use of a CDSS can be effective. Functions of a CDSS may include alerting, reminding, critiquing, interpreting, predicting, diagnosing, assisting, and suggesting.24 The CDSS we used was based on the knowledge development process model.25 The model depicts the process of meeting the clinical decision-making needs of nurses via a decision support system. The process begins with identifying the decision-making need and the associated knowledge domain. The next step is searching for and selecting the best available evidence. Once the quality of evidence has been evaluated, data rules are developed, categorized, mapped, and linked. Data rules are if-then statements that link data or trigger messages. The final step is an evaluation of the support system by experienced clinicians.

Methods
About the Authors
Frank Lyerla is an assistant professor at Southern Illinois University Edwardsville, Edwardsville, Illinois. Cynthia LeRouge and Dorothy A. Cooke are associate professors at Saint Louis University, St Louis, Missouri. Debra Turpin is the chief nurse executive and Lisa Wilson is in risk management and safety at Alton Memorial Hospital, Alton, Illinois. Corresponding author: Frank Lyerla, AH 3334, Southern Illinois University Edwardsville, Edwardsville, IL 620261066 (e-mail: flyerla@siue.edu).

Study Design and Setting The study was approved by the appropriate institutional review board. A modified interrupted time-series design was used.26 The study took place in a 12-bed intensive care unit at Alton Memorial Hospital, Alton, Illinois, a medium-sized hospital. Initial interest in the study occurred when the hospitals quality improvement department reported that the mean HOB elevation for patients receiving

40

AJCC AMERICAN JOURNAL OF CRITICAL CARE, January 2010, Volume 19, No. 1 Downloaded from ajcc.aacnjournals.org by guest on February 8, 2011

www.ajcconline.org

mechanical ventilation was only 27.6 and that the recommended minimum of 30 was used for less than half (6 of 14) of the patients. At the time of the study, the hospital was not using any VAP prevention protocols or bundles. The study consisted of 3 phases. A total of 105 observations (including HOB elevation) were recorded in each phase. Phase 1 occurred 2 months before implementation of the nursing CDSS; phase 2, the first and second months after implementation; and phase 3, the fourth and fifth months after implementation. Thus, completion of the entire study required 7 months. The purpose of the third phase was to assess the continued effect of the CDSS. Although 330 measurements of HOB elevation were collected during the 3 phases, 15 were excluded because elevating the HOB to a minimum of 30 was contraindicated for the patients involved. Intervention The CDSS used in the study involved a pop-up alert window that appeared on the computer screen when a nurse electronically documented the HOB angle for a patient receiving mechanical ventilation. The section prompting the nurse to document the angle was added to the patients flow sheet and served as the prompt for the CDSS reminder. Additionally, nurses were required by hospital policy to document the HOB angle on the flow sheet every 4 hours. If the angle was 45 or greater, the CDSS was not triggered. If the angle was less than 45, the nurse was reminded of the recommended HOB angle for patients receiving mechanical ventilation: a minimum of 30 and preferably, if not contraindicated, 45. Because of system limitations, only a single message could be programmed as a reminder. Therefore, HOB angles less than 45 but greater than or equal to 30 triggered the same reminder. The CDSS then directed the nurse to select any contraindications, as specified by AACN, to the recommended HOB elevation. AACN lists 6 contraindications27: medical (physician) orders, hypotension, unstable physiological status, low cardiac index, procedures that preclude HOB positioning, and prone positioning. A seventh option was provided to identify (nurse perceived) contraindications not indicated on the AACN list. Some AACN contraindications (such as unstable physiological status) are open to subjective interpretation. If a nurse thought that 1 of the 6 AACN contraindications did not apply to a patient, the nurse might be inclined to elevate the patients HOB when a valid contraindication existed. The seventh option, otherspecify, was included to help lower the risk for inappropriate HOB eleva-

tion due to interpretation error by a nurse. Additional information on the development and implementation of the nursing CDSS used in this study is reported elsewhere.28 Data Collection Consent was obtained from and data were collected on 42 patients and 33 registered nurses. Written data were entered into and stored in an electronic database using Microsoft Access 2003 (Microsoft Corp, Redmond, Washington). Calculations indicated that 105 HOB measurements would be necessary for each phase of the study for a large effect size of 45% and a power of 90%. All HOB measurements collected were linked (via coding) to the nurse responsible for positioning the patient at the time the measurement was recorded. Additional data collected on nurses included number of years worked as a registered nurse and number of years in critical care. After discussions with administrative hospital personnel, a decision was made to limit data collected on nurses to experience alone. HOB measurements for patients receiving mechanical ventilation were collected 1 to 3 times per day by the principal investigator (F.L.) during each of the 3 phases. He entered the intensive care unit via a back door and recorded HOB measurements immediately to minimize any cueing or influence on the nursing staff and their patients HOB angle. Additionally, HOB measurements were collected at times that produced a fairly equal number of day- and night-shift observations. Of the 315 measurements, 175 (56%) were collected between 7 AM and 6:59 PM and 140 (44%) between 7 PM and 6:59 AM. The HOB measurements included those obtained with a leveled protractor or bed scales and the angles electronically documented by nurses. The following data were collected for each patient via review of medical records: medical diagnoses, use of tube feeding, height, weight, BMI, age, sex, and any contraindications to HOB elevation. Measurements obtained with the bed scales did not differ significantly from those obtained with the protractor (t = 1.19; P = .23). However, the protractor measurements were used for statistical analyses because some beds did not have working scales. The nurse-documented HOB angles differed significantly from the protractor measurements (t = -5.525; P < .001) and the bed scale measurements (t = -5.532; P < .001). The nurse-documented HOB

Nurses believed some patients experienced contraindications to head-of-bed elevations 45.

www.ajcconline.org

AJCC AMERICAN JOURNAL OF CRITICAL CARE, January 2010, Volume 19, No. 1 Downloaded from ajcc.aacnjournals.org by guest on February 8, 2011

41

Table 1 Characteristics of the sample

Characteristic Patients receiving mechanical ventilation (n = 42) Age, mean (SD), range, y Body mass Mean (SD), range Underweight , <18.5 Normal weight, 18.5 - 24.9 Overweight, 25.0 - 29.9 Obese, 30 Sex Men Women Primary diagnosis Pulmonary Cardiac Gastrointestinal Metabolic Other Nurses (n = 33) Years as a registered nurse Mean (SD), range <5 5 Years in critical care Mean (SD), range <5 5
a

Valuea

66.9 (14.6), 39 - 96 28.59 (7.88), 15.8 - 51.3 1 (2) 15 (36) 14 (33) 12 (29) 22 20 16 6 6 4 10 (52) (48) (38) (14) (14) (10) (24)

indexb

12.5 (8.9), 0 - 30 8 (24) 25 (76) 6.6 (7.2), 0 - 25 18 (55) 15 (45)

for the demographic data. Appropriate calculations included frequencies, ranges, percentages, means, and standard deviations. Several categories were collapsed to provide more meaningful data for cross-tabulation analysis, including BMI, medical diagnosis, and HOB angle. BMI was categorized in accordance with the US Department of Health and Human Services and the National Institutes of Health.29 Medical diagnoses were categorized as gastrointestinal, pulmonary, metabolic, cardiac, and other. HOB elevations were categorized as angles of 30 or greater but less than 45, and angles of 45 or greater. Gastrointestinal diagnoses included abdominal pain, gastrointestinal bleeding, diverticulitis, and abdominal obstruction. Pulmonary diagnoses included respiratory arrest, pneumonia, and hypoxia. Metabolic diagnoses included hypoglycemia, hyperkalemia, and drug overdose. Cardiac diagnoses were chest pain, acute myocardial infarction, and congestive heart failure. Other diagnoses included cancer, seizures, and cellulitis.

Results
Characteristics of the Sample Table 1 is a summary of the characteristics of both groups, patients and nurses. Patients were 39 to 96 years old, with a mean of 66.9 years (SD, 14.6). Just over half of patients were men (n = 22; 52%). The most common primary diagnosis was pulmonary related (n = 16; 38%). Only 1 patient (2%) was categorized as underweight, 12 patients (29%) were categorized as obese, 15 (36%) as normal weight, and 14 (33%) as overweight. The nurses enrolled in the study ranged from new graduates to those with 30 years of nursing experience and from 0 to 25 years of critical care nursing experience. Mean values were 12.5 years (SD, 8.9) for nursing experience and 6.6 years (SD, 7.2) for critical care experience. About one-quarter of the nurses (n = 8; 24%) had less than 5 years of nursing experience, and more than half (n = 18; 55%) had less than 5 years of critical care experience. HOB Angle Descriptive statistics were calculated for the number and percentage of HOB angles of 30 or greater but less than 45, and of HOB angles of 45 or greater. Less than half of the phase 1 HOB measurements (n = 46; 44%) were 30 or greater but less than 45 (Table 2). None of the angles were 45 or greater. In phase 2, both the number of measurements of angles 30 or greater but less than 45 (n = 70, 67%) and the number of measurements of angles of 45 or greater (n = 4; 4%) increased. Changes in

b Body

Values are number of patients or nurses (%) unless indicated otherwise. mass index calculated as weight in kilograms divided by height in meters squared.

Table 2 Descriptive statistics for head of bed angle for patients receiving mechanical ventilation, by study phase (105 observations per phase)
Phase Angle, No. 30 to <45 45 46 0 1 % (44) (0) No. 70 4 2 % (67) (4) No. 62 5 3 % (59) (5)

angle was about 4 higher than angles determined using the protractor and the bed scales. Data Analysis Data were coded and then transferred from the Access database to SPSS, version 15.0 (SPSS Inc, Chicago, Illinois), for statistical analyses. A 2-tailed level of .05 was used to determine significance for all calculations. Descriptive statistics were used

42

AJCC AMERICAN JOURNAL OF CRITICAL CARE, January 2010, Volume 19, No. 1 Downloaded from ajcc.aacnjournals.org by guest on February 8, 2011

www.ajcconline.org

Patient and Nurse Characteristics as Predictors Cross tabulations were completed between the dichotomized HOB angles and whether the patient was being tube fed at the time the measurement was recorded, and for each of the categorized medical diagnoses. Only 70 HOB angles (36%) were less than 30 if the patient was being tube fed; 122 (64%) were 30 or greater (Table 4). The difference between the 2 groups was significant (P = .002). Of the 5 diagnosis categories, only pulmonary (2 = 8.1; P = .004) and gastrointestinal (2 = 10.9; P = .001) were significant (Table 4). Most of the 313 HOB measurements were recorded for patients categorized as obese (n = 120; 38%). In total, 99 HOB measurements (36%) were recorded for patients categorized as overweight, and 94 (30%) were for patients categorized as normal weight. Cross tabulations and odds ratios were determined for each BMI category compared with the HOB angles of less than 30 or 30 or greater. A total of 53 of the 99 HOB measurements (54%) recorded for overweight patients were angles less than 30. Overweight patients had significantly lower HOB elevations (P = .01; Table 5) than did patients who were normal weight or obese. Nurses were grouped into those with 5 or more years of nursing and critical care experience (n = 8) and those with less than 5 years (n = 18). Nursing experience was not a significant predictor of HOB elevation. Contraindications Of the 330 HOB measurements obtained, 15 were excluded from data analyses because of contraindications against elevating the HOB 30 or more. During phase 2 of the study, it became apparent that nurses thought that some patients had contraindications to HOB elevations of 45 or greater, but not to elevations of 30 to 44. Therefore, contraindications to HOB elevations of both 30 and 45 were recorded during phases 2 and 3. An additional 61 HOB measurements were recorded for phases 2 and 3 for patients who had contradictions against HOB elevations of 30 or more.

Count

HOB elevation less than 30 and 30 or greater for all 3 phases of the study are depicted in the Figure. The mean HOB angle for phase 1 (27.7; SD, 6.2) was significantly lower than the mean for phase 2 (31.7; SD, 6.2; t = -4.69; P < .001) and the mean for phase 3 (31.1; SD, 6.9; t = -3.69, P < .001). However, the mean HOB angles for phase 2 and phase 3 did not differ significantly (t = 0.73; P = .46; Table 3).

60

40

20

0 Before intervention 1 month after intervention Study phase <30 30 5 months after intervention

Figure Change in head-of-bed angle less than 30 and 30 or more, as measured with a protractor, by study phase.

Table 3 Differences in head of bed angle for patients receiving mechanical ventilation, by study phase (105 measurements per phase)
Phase 1 2 1 3 2 3
a b

Mean 27.7 31.7 27.7 31.1 31.7 31.1

SD 6.2 6.2 6.2 6.9 6.2 6.9

ta

-4.69b -3.69b 0.73

Degrees of freedom = 208 for all ttests. P < .001.

The most common nurse-reported contraindication to an HOB elevation of 30 or greater or to 45 was hypotension. Additional contraindications reported by nurses included medical orders, unstable physiological status, undergoing a procedure, risk for decubitus ulcer, sliding down in bed, invasive femoral catheter, patients preference, abdominal distension, decrease in oxygenation, climbing

www.ajcconline.org

AJCC AMERICAN JOURNAL OF CRITICAL CARE, January 2010, Volume 19, No. 1 Downloaded from ajcc.aacnjournals.org by guest on February 8, 2011

43

Table 4 Comparison of predictors of head of bed angle in patients receiving mechanical ventilation (n = 315 observations)
Angle of head of bed, Potential predictor <30 No. Patient tube fed Primary diagnosis Gastrointestinal Pulmonary Metabolic Cardiac Other
a

shifts and HOB elevations of less than 45 and 45 or greater also indicated no significant difference between shifts (P = .50).
2 1 P

30 No. %

Discussion
Instruments We found no significant difference between the protractor measurements and the bed scales measurements, indicating that both methods are useful for recording HOB elevations. The significant differences between nurse-documented HOB angles and angles measured using a protractor or bed scales could be due to several reasons. Nurses were only required to document HOB angles every 4 hours. Predicting the exact time a nurse would measure the HOB angle and then document that measurement was impossible to correlate with the times the measurements obtained with the protractor and the bed scales were recorded. Therefore, at the time the protractor and bed-scales measurements were recorded, the last HOB angle electronically charted by the nurse was recorded. This time point could have been several hours before the HOB angle was measured via the protractor and the bed scales. Thus, a patients HOB angle measured using a protractor or the bed scales might differ from the last nursedocumented HOB angle because of bathing or other patient care activities. This possibility is a limitation of the study. In addition, the electronically documented HOB angle might have been higher because some nurses waited until the end of their shift to document care for their entire 12-hour shift. This documentation practice was witnessed on several occasions. Nurses could have forgotten the HOB angle measured several hours earlier and documented the angle incorrectly. Usefulness of the Nursing CDSS We found significant improvements in HOB elevations 1 and 5 months after implementation of the nursing CDSS. A mean increase of about 4 in HOB elevation was reported after the implementation of the nursing CDSS. Perhaps of more clinical importance was the increase in the percentage of patients who had HOB elevations greater than 30. Before implementation of the CDSS, only 44% of the 105 recorded HOB measurements were of angles of 30 or greater. After implementation, the percentage increased to 67% (phase 2) and 59% (phase 3). We think that the nursing CDSS contributed to this improvement and continued to be implemented after the conclusion of the study.

% (36) (61) (33) (39) (42) (49)

70 41 40 19 11 26

122 (64) 26 80 30 15 27 (39) (67) (61) (58) (51)

9.9 10.9 8.1 0.5 0.01 0.8

.002a .001a .004a .47 .90 .37

Signicant at P < .05.

Table 5 Impact of body mass indexa on head of bed angle for patients receiving mechanical ventilation (n = 313 measurements)a
Angle of head of bed, Body mass indexb <30 No. Normal weight, 18.5 - 24.9 Overweight, 25.0-29.9 Obese, 30
a Only

30 No. 56 46 75 % (60) (46) (62)

2 1

% (40) (54) (38)

38 53 45

0.5 6.0 2.8

.28 .01c .06

2 measurements were recorded on 1 patient with a body mass index <18.5; they were therefore excluded from this analysis. mass index calculated as weight in kilograms divided by height in meters squared. c Signicant at P < .05.
b Body

out of bed, obesity, safety, side-lying position, cardiac arrest, and not being tube fed. Time of Day Each of the 315 recorded HOB measurements was grouped into either a 12-hour day shift (7 AM6:59 PM) or night shift (7 PM-6:59 AM). Distribution of the number of HOB measurements was fairly even for those recorded on the day shift (n = 175; 56%) and those recorded on the night shift (n = 140; 44%). Of the day-shift HOB measurements, 79 (45%) were angles less than 30, and 96 (55%) were angles of 30 or greater. Among the nightshift measurements, 58 (41%) were angles less than 30 and 82 (59%) were angles of 30 or greater. A cross tabulation was completed for night and day shifts and HOB elevations of less than 30 and angles of 30 or greater; and they indicated no significant difference between shifts (P = .51). A cross tabulation completed for night and day

44

AJCC AMERICAN JOURNAL OF CRITICAL CARE, January 2010, Volume 19, No. 1 Downloaded from ajcc.aacnjournals.org by guest on February 8, 2011

www.ajcconline.org

Nurse and Patient Characteristics Patients with pulmonary diagnoses had significantly higher HOB elevations than did patients with other diagnoses. This finding could have occurred because the nurses knew that HOB elevation promotes chest expansion.30 The presence of a feeding tube was a significant predictor of higher HOB elevations. Nurses probably realized the associated aspiration risk for tube-fed patients and positioned the HOB higher. Patients with gastrointestinal diagnoses had significantly lower HOB elevations than did patients with other diagnoses. This finding appears to be related to the presence or absence of a feeding tube. Only 28 of 67 HOB measurements (42%) categorized as a gastrointestinal-related diagnosis were in tube-fed patients. However, 85 of 120 HOB measurements (71%) were for patients with pulmonary diagnoses who were tube fed. Overweight patients had significantly lower HOB elevations. Although HOB measurements recorded for obese patients were not significantly lower (P = .06) than those of patients in any other weight group, the values might have been lower if a larger sample had been used. Thus, BMI may be a useful predictor of HOB elevation. Research is needed on the effects of BMI on HOB elevation. The final demographic variable analyzed was years of nursing and critical care experience for each nurse. We found no statistical significance for years of nursing experience or critical care experience. Nursing experience was not a predictive variable for HOB elevation. Contraindications Additional research is needed on contraindications to HOB elevation for patients receiving mechanical ventilation. The AACN recommendations are somewhat general and are subject to misinterpretation by nurses. Several nurses in our study chose to electronically enter a contraindication rather than select 1 of the 6 contraindications provided in a drop-down menu. One of the more common contraindications entered by nurses was risk for decubitus ulcer. The Agency for Healthcare Research and Quality (AHRQ) published guidelines for the prevention of pressure ulcers.31 One guideline indicates that patients at risk for pressure ulcers should have their HOB maintained in the lowest degree possible consistent with medical conditions and other restrictions. Most likely many nurses included in our study were aware of this guideline and interpreted it as a contraindication. The guideline published by AHRQ was made known to the nurses involved in our study via a

suggested order printed automatically on the care plan of patients who had Braden scores of 15 or lower for risk of pressure ulcers. The suggested order stated that the patients HOB should be elevated no more than 30. The suggested order was not part of a patients chart but was communicated to nurses during report. The wound care team implemented the suggested order a couple of weeks before the nursing CDSS was implemented. We did not know about this suggested order until the final weeks of data collection. Thus, evidence-based guidelines for one medical diagnosis may conflict with those of another; this possibility is an important area for further study. Guideline compliance appears to be related to the way the information is presented to the nurse. The AHRQ guideline indicated that HOB positioning should be lower, whereas the VAP guideline (nursing CDSS) indicated that HOB positioning should be higher. The AHRQ guideline was merely printed on the patients care plan. The VAP guideline (nursing CDSS) was integrated into the patients electronic flow sheet. Our study showed that HOB elevation increased, which strengthens the argument that presenting the guideline within the electronic flow sheet may be more effective than merely printing guidelines on a care plan. Another nurse-documented contraindication to HOB elevation was lack of tube feedings. This finding suggests that some nurses perceived tube feedings, but not mechanical ventilation, as a risk factor for aspiration. The finding is somewhat surprising because before the nurses participated in the study they were required to sign a consent form that indicated mechanical ventilation (alone) as a risk factor for aspiration. This finding supports the importance of continued nursing education on guidelines and evidence-based practice for VAP prevention. Time of Day Previous studies have indicated that patients are more likely to have higher HOB elevations during the day shift. Helman et al32 suggests that nurses may be inclined to place patients in lower HOB positions at night to facilitate sleep. However, our findings do not support time of day as a predictor of HOB elevation.

The presence of a feeding tube was a significant predictor for higher head-ofbed elevations.

Guidelines for head-of-bed positioning to reduce ventilator-associated pneumonia conflict with those to reduce pressure ulcers.

www.ajcconline.org

AJCC AMERICAN JOURNAL OF CRITICAL CARE, January 2010, Volume 19, No. 1 Downloaded from ajcc.aacnjournals.org by guest on February 8, 2011

45

Limitations Limitations of this study include the use of a small nonrandomized convenience sample at a single data collection site. Data were collected only once before implementation of the CDSS and only twice after implementation. Actual aspiration rates were not determined, and continuous HOB measurements were not collected. Continuous measurements would provide more precise HOB elevations than would single measurements obtained 1 to 3 times per day. The information system used to create the nursing CDSS had several limitations, including use of a character-limited text to remind nurses. In addition, the reminder message was limited to a preset font size and screen position. This study was limited in the data collected on participating nurses. Data were collected solely on nursing experience (years as a registered nurse and years working in critical care). Future investigators should consider incorporating education level and certification status.

eLetters
Now that youve read the article, create or contribute to an online discussion on this topic. Visit www.ajcconline.org and click Respond to This Article in either the full-text or PDF view of the article.

Conclusion
Our results indicate that a nursing CDSS in the form of a reminder can increase adherence to guidelines for prevention of such problems as VAP. Additionally, incorporating the system within a patients electronic flow sheet (accessed regularly by nurses) was effective. Future nursing reminder systems that allow the system to function at the time care is provided may also be effective. Involving nurses in the design and selection of hospital information systems to implement evidence-based practice is recommended. Although the time of day and nursing experience did not affect HOB elevation, several patient characteristics were predictive of HOB elevation. Tube-fed patients and patients with a primary pulmonary diagnosis had higher HOB elevations than did patients who were not tube fed or who had other diagnoses. However, patients with a primary gastrointestinal diagnosis had lower HOB elevations than did patients with other diagnoses. Additional studies on the effect of predictors of HOB elevation are needed.
ACKNOWLEDGMENTS We thank Norma A. Metheny, RN, PhD, professor, Saint Louis University, for her overall guidance, and Dana S. Oliver, biostatistician at Saint Louis University, for her assistance with the statistical analyses. The data were collected at Alton Memorial Hospital. FINANCIAL DISCLOSURES This study was funded by a grant from the American Association of Critical-Care Nurses, Aliso Viejo, California, and Edwards Life Sciences, Irvine, California.

REFERENCES 1. Shaw MJ. ventilator-associated pneumonia. Curr Opin Pulm Med. 2005;11(3):236-241. 2. Chastre J, Fagon JY. ventilator-associated pneumonia. Am J Respir Crit Care Med. 2002;165(7):867-903. 3. Ricart M, Lorente C, Diaz E, Kollef MH, Rello J. Nursing adherence with evidence-based guidelines for preventing ventilator-associated pneumonia. Crit Care Med. 2003; 31(11):2693-2696. 4. Joshi N, Localio AR, Hamory BH. A predictive risk index for nosocomial pneumonia in the intensive care unit. Am J Med. 1992;93(2):135-142. 5. Healthcare Infection Control Practices Advisory Committee; Centers for Disease Control and Prevention. Guidelines for preventing health careassociated pneumonia, 2003 recommendations of the CDC and the Healthcare Infection Control Practices Advisory Committee. Respir Care. 2004;49(8):926-939. 6. Torres A, Serra-Batlles J, Ros E, et al. Pulmonary aspiration of gastric contents in patients receiving mechanical ventilation: the effect of body position. Ann Intern Med. 1992; 116(7):540-543. 7. Ibez J, Peafiel A, Raurich JM, Marse P, Jord R, Mata F. Gastroesophageal reflux in intubated patients receiving enteral nutrition: effect of supine and semirecumbent positions. JPEN J Parenter Enteral Nutr. 1992;16(5):419-422. 8. Orozco-Levi M, Torres A, Ferrer M, et al. Semirecumbent position protects from pulmonary aspiration but not completely from gastroesophageal reflux in mechanically ventilated patients. Am J Respir Crit Care Med. 1995;152(4, pt 1): 1387-1390. 9. Drakulovic MB, Torres A, Bauer TT, Nicolas JM, Nogu S, Ferrer M. Supine body position as a risk factor for nosocomial pneumonia in mechanically ventilated patients: a randomised trial. Lancet. 1999;354(9193):1851-1858. 10. Metheny NA, Chang YH, Ye JS, et al. Pepsin as a marker for pulmonary aspiration. Am J Crit Care. 2002;11:150-154. 11. Metheny NA, Clouse RE, Chang Y, Stewart BJ, Oliver DA, Kollef MH. Tracheobronchial aspiration of gastric contents in critically ill tube-fed patients: frequency, outcomes, and risk factors. Crit Care Med. 2006;34:1007-1015. 12. Grap MJ, Munro CL, Hummel RS III, Elswick RK Jr, McKinney JL, Sessler CN. Effect of backrest elevation on the development of ventilator-associated pneumonia. Am J Crit Care. 2005;14:325-332. 13. McKay CA, Speers M. Watch for AACN practice alerts: new communication to advise on clinical practice standards. AACN News. 2004;21(2). 14. McClave SA, DeMeo MT, DeLegge MH, et al. North American Summit on Aspiration in the Critically Ill Patient: consensus statement. JPEN J Parenter Enteral Nutr. 2002;26(6 suppl): S80-S85. 15. Institute for Healthcare Improvement. Implement the ventilator bundle: elevation of the head of the bed. Institute for Healthcare Improvement Web site. http://www.ihi.org/IHI /Topics/CriticalCare/IntensiveCare/Changes/IndividualChanges /Elevationoftheheadofthebed.htm. Accessed October 8, 2009. 16. American Association of Critical-Care Nurses. AACN practice alert: ventilator associated pneumonia. http://www.aacn.org /WD/Practice/Docs/Ventilator_Associated_Pneumonia_1-2008 .pdf. Issued January 2008. Accessed October 8, 2009. 17. Joint Commission. Specifications manual for national hospital quality measures-ICU. Joint Commission Web site. http://www.jointcommission.org/PerformanceMeasurement /MeasureReserveLibrary/Spec+Manual+-+ICU.htm. Published 2005. Accessed October 8, 2009. 18. Heyland DK, Cook DJ, Griffith L, Keenan SP, Brun-Buisson C. The attributable morbidity and mortality of ventilatorassociated pneumonia in the critically ill patient. The Canadian Critical Trials Group. Am J Respir Crit Care Med.

46

AJCC AMERICAN JOURNAL OF CRITICAL CARE, January 2010, Volume 19, No. 1 Downloaded from ajcc.aacnjournals.org by guest on February 8, 2011

www.ajcconline.org

1999;159(4, pt 1):1249-1256. 19. Osheroff JA, Pifer EA, Teich JM, Sittig DF, Jenders RA. Improving Outcomes With Clinical Decision Support: An Implementers Guide. Chicago, IL: Healthcare Information and Management Systems Society; 2005. 20. Shea S, DuMouchel W, Bahamonde L. A meta-analysis of 16 randomized controlled trials to evaluate computer-based clinical reminder systems for preventive care in the ambulatory setting. J Am Med Inform Assoc. 1996;3(6):399-409. 21. Hunt DL, Haynes RB, Hanna SE, Smith K. Effects of computer-based clinical decision support systems on physician performance and patient outcomes: a systematic review. JAMA. 1998;280:1339-1346. 22. Agrawal A, Mayo-Smith MF. Adherence to computerized clinical reminders in a large healthcare delivery network. Stud Health Technol Inform. 2004;107(pt 1):111-114. 23. Garg AX, Adhikari NK, McDonald H, et al. Effects of computerized clinical decision support systems on practitioner performance and patient outcomes: a systematic review. JAMA. 2005;293:1223-1238. 24. Randolph AG, Haynes RB, Wyatt JC, Cook DJ, Guyatt GH. Users guides to the medical literature, XVIII: how to use an article evaluating the clinical impact of a computer-based clinical decision support system. JAMA. 1999;282:67-74. 25. ONeil ES, Dluhy NM, Fortier PJ, Michel HE. Knowledge acquisition, synthesis, and validation: a model for decision support systems. J Adv Nurs. 2004;47:134-142. 26. Polit DF, Beck CT. Nursing Research: Principles and Methods. Philadelphia, PA: Lippincott Williams & Wilkins; 2003.

27. American Association of Critical-Care Nurses. AACN practice alert: ventilator-associated pneumonia. AACN Clin Issues Adv Pract Acute Crit Care. 2005;16(1):105-109. 28. Lyerla F. Design and implementation of a nursing clinical decision support system to promote guideline adherence. Comput Inform Nurs. 2008;26:227-233. 29. US Department of Health and Human Services, National Institutes of Health. Calculate your body mass index. http: //www.nhlbisupport.com/bmi/. Accessed October 8, 2009. 30. Potter PA, Perry AG. Fundamentals of Nursing. 6th ed. St Louis, MO: Mosby; 2005. 31. Panel for the Prediction and Prevention of Pressure Ulcers in Adults. Pressure Ulcers in Adults: Prediction and Prevention. Rockville, MD: Agency for Health Care Policy and Research, Public Health Service, US Department of Health and Human Services; May 1992. AHCPR Publication No. 92-0047. Clinical Practice Guideline No. 3. 32. Helman DL, Sherner JH, Fitzpatrick TM, Callender ME, Shorr AF. Effect of standardized orders and provider education on head-of-bed positioning in mechanically ventilated patients. Crit Care Med. 2003;31:2285-2290.

To purchase electronic or print reprints, contact The InnoVision Group, 101 Columbia, Aliso Viejo, CA 92656. Phone, (800) 899-1712 or (949) 362-2050 (ext 532); fax, (949) 362-2049; e-mail, reprints@aacn.org.

www.ajcconline.org

AJCC AMERICAN JOURNAL OF CRITICAL CARE, January 2010, Volume 19, No. 1 Downloaded from ajcc.aacnjournals.org by guest on February 8, 2011

47