AJCC
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

American Journal of Critical Care. 2008;17: 522-531
Copyright © 2008 by the American Association of Critical-Care Nurses.
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Respond to This Article
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kiekkas, P.
Right arrow Articles by Baltopoulos, G. I.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kiekkas, P.
Right arrow Articles by Baltopoulos, G. I.

Nursing Workload Associated With Fever in the General Intensive Care Unit

By Panagiotis Kiekkas, RN, MSc, PhD, George C. Sakellaropoulos, PhD, Hero Brokalaki, RN, PhD, Evangelos Manolis, MD, PhD, Adamantios Samios, RN, Chrisula Skartsani, RN and George I. Baltopoulos, MD, PhD. Panagiotis Kiekkas is a grade B nurse in the anesthesiology department, and Adamantios Samios and Chrisula Skartsani are grade A nurses in the intensive care unit at Patras University Hospital, Patras, Greece. George C. Sakellaropoulos is an assistant professor in the Department of Medical Physics, University of Patras, Greece. Hero Brokalaki is an assistant professor, Evangelos Manolis is an associate professor, and George I. Baltopoulos is a professor in the School of Nursing, University of Athens, Greece.

Corresponding author: Panagiotis Kiekkas, 76 Stratigou Konstantinopoulou St., Aroi, Patras 263-31, Greece (e-mail: kiekkpan{at}otenet.gr).


   Abstract
Top
 Abstract
Methods
 Results
 Discussion
 Conclusion
 References
 
Background Fever in a patient in the intensive care unit necessitates several nursing tasks. Moreover, factors associated with increased patient care needs may be associated with fever.

Objective To identify relationships between fever and characteristics of fever and nursing workload at the patient level.

Methods A prospective study was conducted in a medical-surgical intensive care unit. The sample consisted of 361 patients consecutively admitted from October 2005 to August 2006. Each patient’s body temperature was measured by using a tympanic membrane or an axillary thermometer. The Therapeutic Intervention Scoring System-28 was used to measure nursing workload.

Results A total of 188 patients (52.1%) had fever. Mean daily scores on the Therapeutic Intervention Scoring System and on 5 of its 7 categories were significantly higher for febrile patients than for nonfebrile patients. Fever was an independent predictor of the mean daily scores for all patients (P < .001). Peak body temperature but not duration of fever also was an independent predictor of mean daily scores for febrile patients (P < .001).

Conclusion In a general intensive care unit, fever in patients should be taken into consideration for the proper allocation of nursing personnel.

Nursing workload refers to nursing time and skills required for providing care to patients. At the patient level, the higher the demand for nursing care, described by the term patient dependency, the higher the nursing workload required for adequately meeting this demand.1 Systems for measuring nursing workload have been based on the systematic quantification of nursing activities. At the unit level, these systems have traditionally been used for staffing purposes; that is, the estimation of appropriate nurse to patient ratio and the number of nurses needed to care for patients on a daily basis.2 Several of these systems have been developed for and used in the intensive care environment.

Fever in patients in a general (medical-surgical) intensive care unit (ICU) may increase patient care demands associated with observation of patients or clinical interventions that either are performed by or require the assistance of nursing personnel. Determining the exact cause of fever, although important for planning care, is often difficult. However, the onset, magnitude, pattern, and duration of fever may suggest the cause.3 Diagnosis entails obtaining cultures of blood and other materials, ordering chest radiographs, or performing more invasive procedures (eg, bronchoalveolar lavage). The most common infectious causes of fever are ventilator-associated pneumonia and bloodstream, urinary tract, and intra-abdominal infections. Noninfectious processes associated with fever include postoperative stress, cerebral damage, administration of drugs, blood transfusions, and myocardial infarction.4

Options for treating elevation in body temperature include changing intravascular catheters to a new site and administering empiric antimicrobial therapy and antipyretic drugs. When response to drugs is limited or body temperature is high (>40°C), physical techniques can be used.5 These techniques mainly include placement of ice cool packs (on the neck, axilla, or groin), use of air-flow or water-flow cooling blankets, and body sponge baths with water or alcohol solution.6 Because fever provides an adaptive advantage to the host, mainly through generation of heat-shock protein, its suppression has been controversial.7 However, because even moderate temperature elevation can deleteriously affect brain tissue, fever should be treated in patients who have cerebral damage.8 Moreover, during a fever episode, heart and metabolic rate, cardiac output, and oxygen consumption markedly increase.9 In patients with preexisting cardiopulmonary disease or sepsis, inability to compensate for these increases can result in severely unstable hemodynamic status, which should be continually monitored and properly treated.


Nursing workload systems are used to estimate nurse to patient ratios.

 

Noncausative associations between fever and patient acuity are also possible. Scores on the Acute Physiology and Chronic Health Evaluation (APACHE) II were significantly higher in febrile general ICU patients (mean, 15.8; SD, 6.1) than in nonfebrile patients (mean, 12.1; SD, 6.7; P = .04).10 Likewise, in a surgical ICU, patients with fever had significantly higher APACHE III scores (mean, 67.8; SE, 1.4) than did patients without fever (mean, 46.2; SE, 0.7; P < .001).11 These differences in clinical severity cannot be attributed to fever itself; rather, they seem to be a result of febrile patients’ comorbid conditions (eg, infection or cerebral damage). A strong positive correlation (Pearson correlation coefficient, r =0.54–0.68, P <.001) has been reported between clinical severity and ICU patient care demand measured by using the Therapeutic Intervention Scoring System (TISS)-28.12,13 Moreover, febrile ICU patients also have been significantly younger (mean, 56 years; SD, 17) than non-febrile ones (mean, 61 years; SD, 17; P < .05),14 and a weak but significant negative correlation has been found between age of these patients and TISS-76 score (r = –0.047; P = .02).15

Infection and cerebral damage not only are main causes of fever in general ICU patients, but also may increase nursing workload regardless of fever manifestation. Saulnier et al16 studied the alterations in daily nursing workload (measured by using the Project Research in Nursing score) in a medical ICU in relation to the presence of nosocomial infections attributed to multidrug-resistant bacteria. The daily score for nursing workload was significantly higher for infected patients (mean, 160; SD, 25) than for noninfected ones (mean, 146; SD, 34; P = .03). Infection control procedures, such as isolation precautions, were mainly responsible for this difference in nursing workload. Thus, the increased care demands of febrile patients may be due primarily to the underlying infectious process. Cerebral damage also can result in increases in patient acuity, but this effect has not been reported in the literature.

The aim of our study was to identify relationships between (1) manifestation and characteristics of fever in general ICU patients and (2) nursing workload at the patient level. We hypothesized that manifestation of fever, as well as high temperature and long duration of fever, would significantly increase patient care demands.


Daily nursing workload is higher in patients with nosocomial infections.

 


   Methods
Top
 Abstract
 Methods
Results
 Discussion
 Conclusion
 References
 
A prospective, descriptive approach was used. Permission to conduct this study was obtained by the nursing agency and the hospital ethics committee.

Sample and Setting
The study was conducted in a 14-bed, medical-surgical ICU at Patras University Hospital, Patras, Greece, a tertiary care, academic hospital. All patients who were consecutively admitted from October 2005 to August 2006 and stayed for at least 12 hours were included in the study. For patients readmitted to the ICU during the study period, only the data from the first admission were included. For each patient, age, sex, APACHE II score (calculated during the first 24 hours after ICU admission), type of admission (medical or surgical), duration of mechanical ventilation, and ICU length of stay were recorded.

The purpose and method of the study were explained to all nurses, and their verbal consent was obtained. Before enrollment in the study, each patient’s designated health care surrogate provided written informed consent. Use of a tympanic membrane thermometer is comfortable, and because it is not placed in direct contact with the membrane, patients are not at risk for perforation or trauma. Care was also taken to avoid spreading infection during tympanic measurements; the thermometer lens filter, which comes into contact with the patient, was replaced after each use.

Measurement of Core Temperature
Fever was defined as a core temperature of 38.3°C or greater, in accordance with the consensus statement of the Society of Critical Care Medicine and the Infectious Diseases Society of America.17 Two methods were used to measure patients’ temperature. Three of the researchers (P.K., A.S., and C.S.) used ThermoScan plus (IRT 3520, Braun GmbH, Kronberg, Germany), an infrared light thermometer that detects heat radiated by the tympanic membrane. The researchers self-trained in the use of this thermometer (according to official guidelines provided by Braun) on ICU patients for a 1-month period before the study began. Especially when measurements are made by highly trained personnel, temperatures measured via the tympanic membrane are strongly correlated with those measured via the pulmonary artery (r = 0.98), and the mean difference (0.11°C) is not clinically significant.18 Thus, tympanic membrane temperature was considered to properly represent core temperature. Axillary, quicksilver thermometers were used when measurement via the tympanic membrane was contraindicated (hemotympanum, cerebrospinal fluid otorrhea, occlusion of the auditory canal).

Nursing personnel employed in the ICU used axillary, quicksilver thermometers. Axillary temperature has been reported to be considerably lower than pulmonary artery temperature (mean difference, 0.27°C–0.68°C)19,20 and thus does not represent core temperature.

Because the purpose of the study was to use axillary measurements of temperature to predict the values of tympanic membrane measurements, simple linear regression was used to determine the correlation between the two; tympanic membrane temperature was the dependent variable and axillary temperature was the explanatory (independent) variable. A pilot study was conducted during October 2005. A total of 30 pairs of simultaneously measured axillary and tympanic membrane temperatures were obtained from 30 ICU patients; in 15 of the pairs, the tympanic membrane value was 38.3°C or greater. For both sites, 2 sequential measurements were obtained and the higher value was recorded, because the lower value could be due to poor technique or insufficient measurement time.

The mean difference between tympanic membrane and axillary temperature was 0.21°C (SE, 0.02°C) and did significantly differ between febrile and nonfebrile patients, so use of 2 different regression equations was not considered necessary. For the whole range of temperature values, the Pearson correlation coefficient was 0.98 (R2 = 0.96; P < .001). The estimated regression equation was tympanic membrane temperature (°C) = 0.889 + 0.982 xaxillary temperature (°C).

Temperatures of all patients included in the study were measured and recorded at 1-hour intervals, by using either of the 2 described methods, on a 24-hour basis.

Measurement of Nursing Workload
The TISS-28 score,21 a simplified version of the original TISS score (of 76 items), was used to measure nursing workload. The TISS-76 was developed by an expert panel who selected items and attributed weights to the items on the basis of the philosophy that the type and number of therapeutic interventions in the ICU are related to severity of illness. Scores on the TISS-28 are strongly correlated with scores on the TISS-76 (r = 0.93), and its reliability and validity have been supported through research.13 One TISS-28 point equals 10.6 minutes of each nurse’s shift; a typical nurse is capable of delivering activities equal to 46.35 TISS-28 points per shift.21 Items included in TISS-28 are divided into 7 categories (Table 1Go). The total TISS-28 score and the category scores for each patient were estimated and recorded on a daily basis. Data were collected at the same time each day by one of the researchers (G.C.S.) not involved in measuring patients’ temperatures, reflecting patient care needs and respective interventions performed during the previous 24-hour period. When a patient was discharged from the ICU (or died), the TISS-28 score reflected the previous shift or 8-hour period.


View this table:
[in this window]
[in a new window]

  		Table 1 Therapeutic Intervention Scoring System-28

 
In order to avoid the Hawthorne effect,22 that is, bias introduced into a study when personnel modify their behavior because they know a study is being performed, all clinical and therapeutic decisions were made solely by ICU medical and nursing staff on the basis of patient care needs, and no additional procedures were carried out because patients were followed up as part of this study. Measuring temperature at 1-hour intervals on a 24-hour basis is a routine clinical nursing practice in the ICU at Patras University Hospital.

When cumulative TISS scores for a patient’s entire ICU stay are calculated, variations of length of stay result in significant differences in overall nursing workload per patient. The total score may be extremely high for patients who have long ICU stays but relatively low daily care demands. In addition, both fever and nosocomial infections in ICU patients are associated with a significant prolongation of ICU stay.14,23 In order to compare nursing workload independent of ICU length of stay, the mean daily TISS-28 score (or mean daily score of each category) of each patient was used. For each patient, this score was defined as the sum of daily TISS-28 scores (or daily scores of each category) for the days the patient stayed in the ICU, divided by the patient’s ICU length of stay.

Documentation of the Cause of Fever
Isolation of at least one pathogen from a clinically relevant sample in combination with Centers for Disease Control and Prevention definitions of nosocomial infections24 was used to confirm the presence of infection, with the exception of pneumonia. Pneumonia was documented as the cause of fever when culture of a patient’s endotracheal aspirate was positive for a pathogen (>106 colony-forming units/mL) and the patient had fever or leukocytosis and clinical or radiographic evidence of pneumonia.

A patient was considered to have cerebral damage when appropriate diagnostic procedures (ie, computed tomography) confirmed the presence of traumatic brain injury, cerebral or subarachnoid hemorrhage, or ischemic stroke. When cerebral damage was detected, fever was considered to be of central origin if the patient’s core temperature was high (>40°C) and no infection was detected. In surgical patients, fever was considered to be postoperative when it occurred within the first 4 days after surgery and no infection was detected.


Ventilatory, neurologic, and metabolic support were significantly higher in febrile patients.

 

Statistical Analysis
SPSS, version 15.0 (SPSS Inc, Chicago, Illinois), was used for statistical analysis of collected data; significance was set at P < .05. The Kolmogorov-Smirnov test was used to check normality of distribution of continuous variables. A t test was used for normally distributed variables (APACHE II and mean daily TISS-28 scores). The Mann-Whitney test was used for nonnormally distributed variables (age, duration of mechanical ventilation, ICU length of stay, and mean daily TISS-28 categories scores). A {chi}2 test was used for categorical variables. Multivariate linear regression (backward elimination) was used to check whether fever (dichotomous variable) was an independent predictor of mean daily TISS-28 score and to check whether peak temperature (highest temperature measured) and duration of fever (in days) were independent predictors of mean daily TISS-28 scores for febrile patients. Age, APACHE II score, and presence of infection or cerebral damage were included in the regression models because these factors were considered possible predictors of nursing workload.


   Results
Top
 Abstract
 Methods
 Results
Discussion
 Conclusion
 References
 
A total of 361 patients were included in the study; 7 patients were ineligible because their ICU stay was less than 12 hours. The sample included 245 men (67.9%) and 173 surgical patients (47.9%). Median age was 56.0 years (interquartile range [IQR], 37.0–71.0) and the mean APACHE II score was 13.4 (SD, 5.5). Median duration of mechanical ventilation was 3.0 days (IQR, 1.0–9.0) and median length of ICU stay was 4.0 days (IQR, 2.0–10.0). A total of 188 patients (52.1%) had fever. A total of 91 patients (25.2%) had infection. The most common sites of infection were the lower respiratory tract (79.1% of infected patients), blood (12.1%), and abdomen (6.6%). Cerebral damage was detected in 69 patients (19.1%), and 57 of these had fever. As shown in Table 2Go, compared with nonfebrile patients, febrile patients were younger (P = .048), had higher APACHE II scores (P = .003), had a longer duration of mechanical ventilation (P < .001), and stayed longer in the ICU (P = .001).


View this table:
[in this window]
[in a new window]

  		Table 2 Differences between febrile and nonfebrile intensive care unit patients

 
Mean daily TISS-28 scores were significantly higher for febrile patients than for nonfebrile patients (P < .001, Table 3Go). Table 4Go gives the multivariate linear regression analysis with mean daily TISS-28 score of all ICU patients as the dependent variable. Manifestation of fever and APACHE II score were the only 2 significant predictors in the model (P = .003 and P < .001, respectively) and explained 25.4% of the variance in mean daily TISS-28 scores (adjusted R2 = 0.254, P < .001).


View this table:
[in this window]
[in a new window]

  		Table 3 Mean daily scores on the Therapeutic Intervention Scoring System-28 (TISS-28) and its categories: differences between febrile and nonfebrile patients

 

View this table:
[in this window]
[in a new window]

  		Table 4 Multivariate linear regression for all patients with mean daily Therapeutic Intervention Scoring System-28 score as the dependent variable

 
Among the 188 patients who had fever, the fever was present at the time of ICU admission in 16 (8.5%), manifested until ICU day 4 in 132 (70.2%), and manifested after ICU day 4 in 40 (21.3%). Median fever duration was 3.0 days (IQR, 1.0–6.0), median peak temperature was 39.1°C (IQR, 38.7°C–39.6°C), and median number of fever episodes was 3.0 (IQR, 1.0–7.0). Fever was considered infectious in 86 patients (45.7%), central in 34 (18.1%), and postoperative in 40 (21.3%). Mean daily TISS-28 scores were significantly higher in patients with central fever (mean, 34; SD, 3.9) than in patients with postoperative fever (mean, 31.8; SD, 4.1; P =.03). The scores for these 2 groups did not differ significantly from those of patients with infectious fever (mean score, 32.9; SD, 3.7).

The multivariate linear regression analyses with mean daily TISS-28 score for febrile ICU patients as the dependent variable are presented in Tables 5Go and 6Go. In the first analysis, peak temperature was included and was the only significant predictor of mean daily TISS-28 score (P < .001). This model explained 31.9% of the variance in mean daily TISS-28 scores (adjusted R2 = 0.319; P < .001). In the second analysis, duration of fever was included, but it was not a significant predictor in the model, which explained 19.6% of the variance in mean daily TISS-28 scores (adjusted R2 = 0.196; P < .001). In this model, APACHE II score was the only significant predictor (P < .001).


View this table:
[in this window]
[in a new window]

  		Table 5 Multivariate linear regression for febrile patients with mean daily Therapeutic Intervention Scoring System-28 score as the dependent variable and peak temperature as the independent variable

 

View this table:
[in this window]
[in a new window]

  		Table 6 Multivariate linear regression for febrile patients with mean daily Therapeutic Intervention Scoring System-28 score as the dependent variable and duration of fever as the independent variable

 

   Discussion
Top
 Abstract
 Methods
 Results
 Discussion
Conclusion
 References
 
The possibility that TISS-28 scores were underestimations of differences in nursing workload between patients with and without fever cannot be ruled out because some tasks associated with fever are not considered in the TISS-28 score (eg, physical antipyretic interventions, isolation techniques). However, nursing workload has no widely accepted definition, and its concepts have been defined and measured in different and often contradictory ways.25 A perfect measurement of nursing workload, one that includes the entirety of nursing activities, will probably never be developed.26 Thus, underestimation of the relationship between fever and nursing workload is a possible limitation for all existing measurement systems, because no system is superior to the others. A further criticism of nursing workload measures is that tasks not directly associated with patient care (eg, organizational and administrative tasks) are not addressed by most of the current measurement systems.26 However, this criticism is valid only when workload is estimated at the unit level. In our study, because nursing workload was estimated at the patient level, only direct care activities (which reflect patient care needs) must be considered.

Although we thought that using mean daily TISS-28 scores allowed for comparisons independent of the ICU length of stay, the effect of the variable lengths of stay might not have been completely eliminated if length of stay was significantly correlated with mean daily TISS-28 scores. If a significant correlation existed, then mixed linear modeling might be necessary to incorporate the longitudinal character of data (daily TISS-28 score and daily manifestation of fever). However, mean daily TISS-28 score and ICU length of stay were not significantly correlated in our study (Spearman rank correlation coefficient = 0.057; P = .28). Because of this lack of correlation and the extremely large range of ICU length of stay (1–68 days, which would result in too many subgroups in mixed linear modeling), we preferred to use mean daily TISS-28 scores and more common statistical methods, even though this choice might be suboptimal.

Our study is the first in which mean daily nursing workload in a general ICU was significantly higher for febrile patients than for nonfebrile patients. Even more important was the finding that fever was a significant predictor of mean daily nursing workload when all factors associated with nursing workload were included in the regression model. This result suggests an independent association between fever and nursing workload at the patient level, once clinical severity, age of patients, and presence of infection and cerebral damage are accounted for.

Scores for the 7 categories of TISS-28 were significantly higher for febrile patients for 5 categories: basic activities, ventilatory support, metabolic support, neurological support, and specific interventions. As previously discussed, high numbers of diagnostic procedures and drugs administered are common in patients with fever and may explain the high basic activities score. The score for ventilatory support is higher for patients receiving mechanical ventilation than for patients who are breathing spontaneously. Duration of mechanical ventilation is significantly longer in patients with fever than in patients without fever because fever increases metabolic rate and ventilatory needs, thus considerably delaying the weaning process.14

The score for metabolic support increases when a patient receives parenteral nutrition. In patients with fever, the parenteral route is often preferred for feeding (instead of the enteral route), because the increased metabolic rate associated with fever decreases intestinal blood flow and absorption of nutrients.27 Scores for neurological support and specific interventions are generally higher in patients with cerebral damage than in patients without such damage because of measurements of intracranial pressure and surgical or diagnostic procedures performed outside the ICU (mainly craniotomy or computed tomography). Many patients with cerebral damage have central fever, and these patients may account for the association of fever with higher scores for neurological support and specific interventions.28

Among fever characteristics, peak temperature was a significant predictor of the mean daily TISS-28 score of febrile patients. A high fever can have deleterious effects, such as increased metabolic rate, neurological deterioration, suppression of immune responses, and impaired oxygen release to tissues.29 The positive association between peak temperature and patient care needs may be attributed to the aggravation of patients’ clinical status and the intensification of antipyretic interventions necessary to prevent dangerously high core temperatures. Moreover, the contribution of patients with cerebral damage to this association may be important, because they require a high level of therapeutic activity (according to our findings), and central fever (often a combination of fever and hyperthermia) is generally characterized by high temperature elevations.

Besides its statistical significance, what is the importance of the association between fever and patient care demand from a clinical or administrative point of view? In the Western world, the demand for qualified nursing personnel in ICUs has increased, mainly because of the increasing number and acuity of patients, but the supply of nurses has declined.30,31 Because of this imbalance, the best possible allocation of existing personnel among ICU patients requires an accurate prediction of patient care needs, which are expected to vary significantly (the daily TISS-28 score ranged from 15 to 52 in our study). The importance of identifying predictors of these needs has been supported by research findings, which confirm that properly matching patient care demands with nursing personnel is a main determinant of patients’ safety.32,33 When the nursing workload in the ICU increases, patients’ probabilities of dying and of receiving a compromised quality of care increase significantly. Use of fever as a predictor of nursing workload can be further supported by the characteristics of temperature measurement; the measurement is easy to perform, is relatively accurate, is not time-consuming for nurses, and is convenient for patients.


Nursing intervention scores were higher in patients with central fever versus postoperative fever.

 

In our study, manifestation of fever was followed by a 9.7% mean increase of mean daily nursing workload at the patient level; the increases were 11.4% and 25.9% when peak temperature exceeded 39.2°C and 40.2°C, respectively. Thus, at the patient level, the proper nurse to patient ratio is expected to be higher for patients with fever than for those without fever, and especially higher for those who have high fever. At the ICU level, the higher the proportion of febrile patients and the higher the patients’ temperatures, the higher the expected nursing workload. More specifically, a daily increase of A percentage units in febrile patients would lead to a mean nursing workload increase at the ICU level of (A/100) x N x 0.097 x (mean daily TISS-28 score), where N is the total number of patients. Although mean daily TISS-28 scores of patients and number of beds may differ markedly among ICUs, this increase seems to be small to justify differences at the daily nurse staffing level, except for ICUs with a large number of beds or for extreme fluctuations in the percentage of febrile patients. However, a single variable probably is not the explanation for a large part of the variance in nursing workload. Instead, fever should be included in prediction models, along with other factors known to contribute to increases in nursing workload.


Daily nursing workload was significantly higher in febrile general ICU patients.

 

Besides the possible underestimation of differences in nursing workload between patients with and without fever, a further important limitation of our study was the use of 2 methods for measuring patients’ temperature in order to combine a satisfactory accuracy of core temperature measurement (tympanic membrane temperature) with a high frequency of measurements (axillary temperature) so that the best possible detection of fever could be achieved. However, although a regression equation was estimated for converting axillary temperature values to tympanic membrane values, in some patients the actual difference between measurements obtained at these 2 sites was undoubtedly greater or smaller than the difference estimated by using the regression equation, and conversion could not always be accurate. A third limitation was the lack of determination of interrater reliability both among researchers who measured tympanic membrane temperature and among nurses who measured axillary temperature. Fourth, the accuracy and precision of both tympanic membrane and axillary temperature measurements have recently been challenged.34 Fifth, although we tried to determine the most likely cause of fever in our patients, in some patients temperature elevation might have been attributed to more than a single cause (eg, infection coexisted with cerebral damage, or postoperative fever was followed by infectious fever). A final limitation is the high heterogeneity of patients in a general ICU. Our study was done at a single center, and our sample cannot be considered representative of all general ICU patients, because differences in the mix of patients are expected among general ICUs of different hospitals. Thus, the degree of generalizability of these findings may be limited.


The TISS may underestimate nursing workload differences in febrile and nonfebrile patients.

 


   Conclusion
Top
 Abstract
 Methods
 Results
 Discussion
 Conclusion
References
 
Manifestation of fever in general ICU patients is an independent predictor of nursing workload at the patient level. Patient care needs associated with basic activities, ventilatory support, metabolic support, neurological support, and specific interventions were higher in febrile patients than in nonfebrile patients. Peak temperature in patients with fever was also an independent predictor of mean daily patient care demands. These findings can be considered a reminder that regardless of the increasing number of clinical procedures and more complex technology used in the ICU, patients’ clinical characteristics and signs and symptoms remain important predictors of nursing intensity. If the association between fever and nursing workload is confirmed through the use of other measurement systems, fever should be considered when nurse managers try to predict patient care needs and properly allocate nursing personnel in the ICU. Modifying the existing systems used to measure nursing workload by including a fever weighting is suggested.


   ACKNOWLEDGMENTS
 
We thank the medical and nursing personnel of the Patras University Hospital ICU for their valuable assistance during data collection.

To purchase electronic or print reprints, contact The InnoVision Group, 101 Columbia, Aliso Viejo, CA 92656. Phone, (800) 809-2273 or (949) 362-2050 (ext 532); fax, (949) 362-2049; e-mail, reprints{at}aacn.org.

FINANCIAL DISCLOSURES
None reported

eLetters
Now that you’ve 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.


   REFERENCES
Top
 Abstract
 Methods
 Results
 Discussion
 Conclusion
 References
 

  1. Hurst K. Relationships between patient dependency, nursing workload and quality. Int J Nurs Stud. 2005;42(1):75–84.[Medline]
  2. Adomat R, Hewison A. Assessing patient category/dependence systems for determining the nurse/patient ratio in ICU and HDU: a review of approaches. J Nurs Manag. 2004;12(5):299–308.[Medline]
  3. Cunha BA. Fever in the intensive care unit [editorial]. Intensive Care Med. 1999;25(7):648–651.[Medline]
  4. Marik PE. Fever in the ICU. Chest. 2000;117(3):855–869.[Medline]
  5. Plaisance KI, Mackowiak PA. Antipyretic therapy: physiologic rationale, diagnostic implications, and clinical consequences. Arch Intern Med. 2000;160:449–456.[Abstract/Free Full Text]
  6. Price T, McGloin S. A review of cooling patients with severe cerebral insult in ICU, I. Nurs Crit Care. 2003;8(1):30–36.[Medline]
  7. Hasday JD, Singh IS. Fever and the heat shock response: distinct, partially overlapping processes. Cell Stress Chaperones. 2000;5(5):471–480.[Medline]
  8. Cairns CJS, Andrews PJD. Management of hyperthermia in traumatic brain injury. Curr Opin Crit Care. 2002;8:106–110.[Medline]
  9. Bruder N, Raynal M, Pellissier D, Courtinat C, Francois G. Influence of body temperature, with or without sedation, on energy expenditure in severe head-injured patients. Crit Care Med. 1998;26:568–572.[Medline]
  10. Circiumaru B, Baldock G, Cohen J. A prospective study of fever in the intensive care unit. Intensive Care Med. 1999; 25:668–673.[Medline]
  11. Barie PS, Hydo LJ, Eachempati SR. Causes and consequences of fever complicating critical surgical illness. Surg Infect (Larchmt). 2004;5(2):145–159.
  12. Castillo-Lorente E, Rivera-Fernández R, Rodriguez-Elvira M, Vázquez-Mata G. TISS 76 and TISS 28: correlation of two therapeutic activity indices on a Spanish ICU database. Intensive Care Med. 2000;26(1):57–61.[Medline]
  13. Yee Kwok WW, Chun Chau JP, Pau Le Low L, Thompson DR. The reliability and validity of the therapeutic activity index. J Crit Care. 2005;20:257–263.[Medline]
  14. Peres Bota D, Lopes Ferreira F, Melot C, Vincent JL. Body temperature alterations in the critically ill. Intensive Care Med. 2004;30:811–816.[Medline]
  15. Castillo-Lorente E, Rivera-Fernández R, Vázquez-Mata G. Limitation of therapeutic activity in elderly critically ill patients. Project for the Epidemiological Analysis of Critical Care Patients. Crit Care Med. 1997;25(10):1643–1648.[Medline]
  16. Saulnier FF, Hubert H, Onimus TM, et al. Assessing excess nurse work load generated by multiresistant nosocomial bacteria in intensive care. Infect Control Hosp Epidemiol. 2001;22(5):273–278.[Medline]
  17. O’Grady NP, Barie PS, Bartlett J, et al. Practice guidelines for evaluating new fever in critically ill adult patients. Task Force of the Society of Critical Care Medicine and the Infectious Diseases Society of America. Clin Infect Dis. 1998; 26(5):1042–1059.[Medline]
  18. Amoateng-Adjepong Y, Del Mundo J, Manthous KA. Accuracy of an infrared tympanic thermometer. Chest. 1999;115: 1002–1005.[Medline]
  19. Erickson RS, Kirklin SK. Comparison of ear-based, bladder, oral, and axillary methods for core temperature measurement. Crit Care Med. 1993;21:1528–1534.[Medline]
  20. Lefrant J-Y, Muller L, de la Coussaye JE, et al. Temperature measurements in intensive care patients: comparison of urinary bladder, oesophageal, rectal, axillary, and inguinal methods versus pulmonary artery core method. Intensive Care Med. 2003;29(3):414–418.[Medline]
  21. Reis Miranda D, de Rijk A, Schaufeli W. Simplified therapeutic intervention scoring system: the TISS-28 items—results from a multicenter study. Crit Care Med. 1996;24:64–73.[Medline]
  22. Wickström G, Bendix T. The "Hawthorne effect"—what did the original Hawthorne studies actually show? Scand J Work Environ Health. 2000;26(4):363–367.[Medline]
  23. Olaechea PM, Ulibarrena MA, Alvarez-Lerma F, Insausti J, Palomar M, De la Cal MA. Factors related to hospital stay among patients with nosocomial infection acquired in the intensive care unit. Infect Control Hosp Epidemiol. 2003; 24:207–213.[Medline]
  24. Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC definitions for nosocomial infections. Am J Infect Control. 1988;16:128–140.[Medline]
  25. Morris R, MacNeela P, Scott A, Treacy P, Hyde A. Reconsidering the conceptualization of nursing workload: literature review. J Adv Nurs. 2007;57:463–471.[Medline]
  26. Ball C. Patient: nurse ratios in critical care—time for some radical thinking [editorial]. Intensive Crit Care Nurs. 2001; 17(3):125–127.[Medline]
  27. Chan S, McCowen KC, Blackburn GL. Nutrition management in the ICU. Chest. 1999;115(5 suppl):145S–148S.[Medline]
  28. Stoccheti N, Rossi S, Zanier ER, Colombo A, Beretta L, Citerio G. Pyrexia in head-injured patients admitted to intensive care. Intensive Care Med. 2002;28:1555–1562.[Medline]
  29. Hasday JD, Fairchild KD, Shanholtz C. The role of fever in the infected host. Microbes Infect. 2000;2:1891–1904.[Medline]
  30. Harrison L, Nixon G. Nursing activity in general intensive care. J Clin Nurs. 2002;11:158–167.[Medline]
  31. Ewart GW, Marcus L, Gaba MM, Bradner RH, Medina JL, Chandler EB. The critical care medicine crisis: a call for federal action: a white paper from the critical care professional societies. Chest. 2004;125(4):1518–1521.[Medline]
  32. Carayon P, Gurses AP. A human factors engineering conceptual framework of nursing workload and patient safety in intensive care units. Intensive Crit Care Nurs. 2005;21:284–301.[Medline]
  33. Tarnow-Mordi W, Hau C, Warden A, Shearer AJ. Hospital mortality in relation to staff workload: a 4-year study in an adult intensive-care unit. Lancet. 2000;356:185–189.[Medline]
  34. Lawson L, Bridges EJ, Ballou I, et al. Accuracy and precision of noninvasive temperature measurement in adult intensive care patients. Am J Crit Care. 2007;16(5):485–496.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Am J Crit CareHome page
J. Pickett
Critical Thinking a Necessary Factor in Nursing Workload
Am. J. Crit. Care., March 1, 2009; 18(2): 101 - 101.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Respond to This Article
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kiekkas, P.
Right arrow Articles by Baltopoulos, G. I.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kiekkas, P.
Right arrow Articles by Baltopoulos, G. I.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS