Nursing demand in intensive therapy units assessed by the Nursing Activities Score
*Edyta K. Cudak, Danuta Dyk
Department of Anaesthesiological and Intensive Care Nursing, Poznań University of Medical Sciences
Background. The cost of nursing in a modern hospital can take up to 50% of the total budget. Therefore, it is very important to use objective tools for assessment of the nursing workload and adjust staff requirements accordingly. The purpose of the study was to evaluate nursing workload using the Nursing Activities Score (NAS) in intensive care units.
Methods. This prospective analysis of nursing care of 314 adult patients was performed simultaneously in five intensive care units in Poland. NAS was used for the evaluation of nursing workload and the APACHE II score was used for assessment of the severity of cases. The APACHE score was calculated during the first 24 hours of ICU stay, while the NAS was recorded over an entire stay.
Results. The average age of the patients was 58.4±16.6 years. The length of stay was 8.7±11.4 days. The distribution between surgical (56%) and non-surgical cases (44%) was almost equal. Mortality was 21%. The median APACHE II score was 20 (1 - 42) and the NAS score 84.4%. There was no correlation between patients’ clinical condition and nursing workload (p>0.05), and the workload in surgical and non-surgical patients. The NAS score correlated well with the length of stay (p<0.05).
Conclusion. The study showed that ICU nurses spend 84.4% of their working time by the patient’s bed. The optimal ICU nurse: patient ratio was estimated to be 1:1.2.
Intensive therapy is an interdisciplinary field in which cooperation with other medical specialities ensures that patients are provided with all necessary aid during complex surgical and conservative procedures. ITUs, where highly specialized procedures of monitoring and treatment are used, generate extremely high costs [1, 2]. In some countries over 30% of hospital budget is allocated for this purpose . Moreover, ITU costs result from the employment of nursing personnel, which is the highest investment of hospitals, accounting for about 50% of total expenditure [2, 3]. The cost reductions unfavourably affect the quality of care as proper nursing staffing (the number of nurses in relation to the number of patients) and appropriate postgraduate qualifications of nurses guarantee good quality of nursing care [4, 5, 6].
In Poland, nursing staffing is assessed using only the TISS -28, which mainly calculates the ITU treatment costs . In some studies, the TISS-28 and Nine Equivalents of Nursing Manpower Use Score (NEMS) were applied to assess nursing workload [8, 9, 10]. Since the scales mentioned do not fully reflect the nursing activities provided, the aim of the present study was to determine nursing workload and assess nursing demand in ITUs using the Nursing Activities Score (NAS).
The study design was approved by the Bioethics Committee of Poznań University of Medical Sciences.
The prospective study was carried out using observations as well as analyses of medical records and nursing records of patients. The NAS  and APACHE II  classification systems were applied.
NAS describes the working time devoted by nurses to different nursing tasks and administrative activities. It consists of 23 therapeutic-nursing interventions in which nurses are involved. In our study, all activities during each ITU day were recorded. The clinical condition of patients was assessed during the first 24 hours of ITU stay based on APACHE II.
Additionally, the author’s information sheet was used to record demographic data, admission date, discharge/death date, the diagnosis directly resulting in ITU admission, coexisting chronic diseases, modes of ITU admission and discharge. Furthermore, the 24-hour NAS, number of patients and of nurses for an individual day were recorded.
The data were collected in four stages.
At stage I, after obtaining the appropriate approval, the NAS was adjusted for the study purposes with all the necessary annotations concerning its application according to the instructions of the author of the original version.
At stage II, the ITUs from hospitals of different levels of referral were selected and all the approval procedures for research were completed; in each of the hospitals, a coordinator of the study was appointed.
Stage III covered the period of 2.5 months during which nursing activities for patients and their families were recorded. The NAS for each patient was calculated on a daily basis; the scores were summed up for all ITU patients on a particular day and divided by the number of hospitalized patients. All information was recorded in the author’s overall record. Moreover, the number of nurses on duties and the number of patients on individual shifts were noted every day. The calculations of optimal nursing staffing were based on the scale assumption that one nurse should not work out more than the NAS of 100 pts (100%).
Furthermore, the APACHE II physiological variables were recorded; having accounted for age of respondents and chronic diseases, the overall score for clinical condition was calculated.
In the statistical analysis, basic descriptive measures adjusted to the type of variables were used. Correlations and inter-group differences were analysed using the following tests: the Mann-Whitney test to detect the differences in workload between patients treated surgically and conservatively as well as between survivors and non-survivors; the Spearman coefficient of correlation to determine the relations between the severity of cases and nursing workload as well as between the length of stay and nursing workload. The Student’s t-test was applied to check the significance of the coefficient of correlation. P<0.05 was considered as significant.
The study involved 7 intensive therapy units in Poland; 5 ITUs (3 from teaching hospitals and 2 from regional or community hospitals) completed the study. During the study period, 314 patients were hospitalized, including 186 (69.2%) men and 128 (40.8%) women.
The mean age of all patients was 58.4±16.6 years. The biggest age group (31.1%) consisted of patients >70 years of age whereas the least numerous one of those < 30 years of age.
The length of ITU stay was 8.7±11.4 days; 155 patients were treated <4 days while 80 patients >10 days; 176 (56%) patients were qualified for surgical procedures, 138 (43.9%) for conservative treatment. The mortality during the study period was 21%. Demographic characteristics of patients, their mode of admission and discharge from ITU are presented in Table 1.
Diseases resulting in ITU admissions are listed in Table 2.
Based on APACHE II, the condition of 38 patients was considered moderate, of 127 – medium and of 149 – severe. The median of clinical status was 20 (1-42); for survivors - 20 (1-38) and for non-survivors – 25 (10-42).
Nursing workload expressed as a median based on NAS for an individual patient was 71.2 (12-150.3). In survivors, this score was 68.3 (53.4-128.9), in non-survivors – 107 (69.8-150.3).
Workload did not exceed 72% in any patient. In the group treated surgically, workload was lower than 69.7% whereas in the group treated conservatively – 74.1%.
The workload score (according to NAS) was 421.8 (115.3-924.4). The median of nursing workload for each patient did not exceed 84.4; thus, nurses used only 84.4% of the working time. The optimal number of nurses for this score was 4.4 nurses for a day and night shift, which was not markedly different from the actual number being 4 per duty. The actual nurse-patient ratio estimated according to NAS ranged from 1:1.3 to 1:2 (norm 1:1.2).
No significant correlation was found between the severity of patients` conditions and nursing workload comparing the groups treated surgically vs. conservatively or survivors vs non-survivors.
A significant relation was demonstrated between the ITU stay and nursing workload, which concerned patients surgically (Rs=-0.28) or conservatively treated (Rs=-0.4) as well as survivors and non-survivors (Rs=-0.6). The weakest correlation coefficient was found in the group of survivors: Rs=-0.2.
The non-parametric Mann-Whitney test for intergroup differences did not demonstrate differences in NAS workload according to the methods of treatment applied in ITUs. However, a significant difference in workload (p<0.0001) was found between survivors and non-survivors.
The Nursing Activities Scale published in 2003 has become a valuable tool for assessment of nursing demand and workload in ITUs .
According to the study by Padilha and co-workers , carried out in the group of elderly patients, the median of workload did not exceed 75% and was higher amongst patients treated surgically . The study using different scoring systems, e.g. TISS-28 or NEMS, confirms that workload is higher in the group treated surgically . The difference between surgical and non-surgical patients is 1.6 pts according to NEMS and 4.4 pts according to TISS-28, and is higher in surgically treated patients. Our earlier studies with NEMS did not demonstrate significant differences between the groups of patients mentioned , which was accounted for by the fact that our analysis did not covered first days of stay when the intensity of therapeutic activities might have been higher in relation to this group of patients . The study by Lundgren-Laine  did not show the differences in nursing workload between patients treated surgically and conservatively. This is likely to result from the fact that NAS consists of universal interventions in severely ill patients. Moreover, the scale was supplemented with the nursing activities that were not taken into consideration earlier and that are common to the majority of situations regarding nursing of patients.
The comparison of survivors and non-survivors revealed higher medians of workload as well as maximum and minimum values. Our earlier study in the group of 60 patients using TISS-28 confirms these observations . Similar conclusions were also presented by other authors [17, 18, 19]. The differences in workload are related to more advanced methods of diagnosis, monitoring and treatment of patients as well as patients` clinical conditions. Thus, both the number of nursing-therapeutic activities is higher and more time is needed for complex procedures. The nursing workload is likely to depend on the severity of patient`s condition. Such a relation was demonstrated in our earlier studies with TISS-28 and NEMS [8, 9, 10]. The more severe the clinical condition of patients, the higher TISS-28 and NEMS scores, which is confirmed by the positive correlation between APACHE II vs TISS-28 and TISS-76  and by study findings demonstrating direct correlation between the severity of patients` condition and TISS-28 score . Similar results were obtained in the study comparing SAPS II and TISS-28 .
In our study no significant correlation was found between nursing workload and severity of patient`s condition. The only correlation was observed between subgroups of such individual interventions as respiratory support, communication, diagnostic and therapeutic methods, which accounts for about 69% of total workload. This seems obvious – the more severe the clinical condition of a patient, the more attention and time is needed due to additional complex diagnostic and therapeutic procedures that have to be instituted. However, Miranda and co-workers  did not confirm that workload measured by NAS was related to the clinical condition of patients. What might explain the lack of relation between these two relevant variables? In many cases, severely ill patients undergoing mechanical ventilation and sedation do not require higher workload [11, 21, 22]. Thus, patients without any chances for survival do not require nursing activities due to limited options of treatment . On the other hand, in cases of unconscious patients and their concerned families, when communication is difficult or some anxiety develops, the presence of a nurse is invaluable and the care of such patients is more preoccupying.
Our findings show that in ITUs the working time of nurses is not used in 100%. Yet, it should be remembered that NAS involving administrative activities does not cover the breaks, which are considered the auxiliary time (together with administrative activities). Moreover, NAS lists individual interventions performed only if need be. They are scored one time, irrespective of whether e.g. patients had to be transported to the operating suite twice, required bronchoscopy or additional endotracheal intubation.
The analysis of demands for nursing care in ITUs did not disclose high staff shortage. According to NAS, the optimal demand was estimated based on the nurse:patient ratio of 1:1.2. Similar ratios were calculated in 2000; on average, the ratio found was 1:1.6 .
The length of ITU stay in Poland is comparable with that in other countries. Once the patient`s condition improves, the range and intensity of nursing activities decrease. Adell and co-workers  observed lower NAS workload on hospitalization day 1 compared to our results (<40.8%); this workload decreased reaching 39.3% of nursing time on discharge. Our findings of 2004 show that the duration of ITU stay was 14±17 days . The present findings demonstrate the tendency to shorten the ITU treatment time, which is confirmed by numerous other reports [25, 26, 27, 28].
The tendency to shorten the ITU stay is also visible in the report of the Agency for Healthcare Research and Quality, which indicates that the ITU stay decreased from 7.5 to 4.9 days over the period of 20 years . This may be explained by better procedures of diagnosis, therapy, nursing and rehabilitation of patients. Our analysis revealed that the longer the ITU stay the lower nursing workload. The weak correlation between the duration of ITU stay and TISS interventions was demonstrated by Lundgren-Laine and colleagues .
The limitations of NAS are relevant and to some extent may reduce the actual usage of nursing working time. They are mainly related to individual interventions, which are recorded one time although their frequency during a given duty varies. The same limitations concern TISS-28 and NEMS. On the other hand, NAS enables to describe by about 40% more of the time devoted to nursing activities compared to TISS-28. Moreover, it is important that the scale includes typical nursing activities, which earlier scoring systems lacked [7, 18].
The use of NAS is not time-consuming – 5-6 min., on average. Therefore, in everyday clinical practice NAS may be a useful tool for assessment of the demand for nursing staff.
Workload of nursing staff in Polish ITUs calculated by NAS is not optimal.
According to NAS, the level of nursing staffing in Polish ITUs should be 1:1.2.
No correlation was observed between the clinical condition of patients and nursing workload in ITUs.
NAS enables the assessment of more time devoted to nursing activities compared to other scoring systems.
1. Sznajder M, Leleu G, Buonamico G: Estimation of direct cost and resource allocation in intensive care: correlation with Omega system. Int Care Med 1998; 24: 582-589.
2. Beattie J, Calpin-Davies PJ: Workforce dilemmas: a comparison of staffing in a generalist and a specialist intensive care unit. Intensive and Critical Care Nursing 1999; 15: 52-57.
3. Stricker K, Rothen HU, Takala J: Resource use in the ICU: short – vs long-term patients. Acta Anaesthesiol Scand 2003; 47: 508-515.
4. Ambrose C: Recruitment problem in intensive care: a solution. Nursing Standard 2002; 17: 39-40.
5. Fawzi Abualrub Raeda: Nursing shortage in Jordan: what is the solution? J Prof Nurs 2007; 23: 117-120.
6. Volpatti C, Leathley M, Walley KR: Time-weighted nursing demand is a better predictor than midnight census of nursing supply in an intensive care unit. J Crit Care 2000; 15: 147-150.
7. Miranda DR, 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.
8. Cudak E, Dyk D: Evaluation of nursing workload in ICU based on NEMS (Nine Equivalents of Nursing Manpower Use Score). Problemy Pielęgniarstwa 2007; 1: 7-12.
9. Dyk D, Cudak-Bańska E: Effect of nurse workload on the outcomes of ICU patients – preliminary results. Intensive Care Med 2004; 30 (Suppl. 1): 12.
10. Cudak-Bańska E, Dyk D, Zadroga M: Quantitative evaluation of nursing workload in ICU and application of TISS-28 and NEMS. Medycyna Intensywna i Ratunkowa 2005; 8: 137-143.
11. Miranda DR, Nap R, de Rijk A: Nursing Activities Score. Crit Care Med 2003; 31: 374-382.
12. Knaus WA, Dreper E, Wagner DP: APACHE II: a severity of disease classification system. Crit Care Med 1985; 13: 818-829.
13. Padilha KG: Elderly patients and nursing workload in intensive care units. Crit Care 2007; 11 (Suppl. 3): 37.
14. Rothen HU, Kung V, Ryser DH: Validation of „nine equivalents of nursing manpower use score“ on an independent data sample. Int Care Med 1999; 25: 606-611.
15. Miranda DR, Moreno R, Iapichino G: Nine equivalents of nursing manpower use score (NEMS). Int Care Med 1997; 23: 760-765.
16. Lundgren-Laine H, Suominen T: Nursing intensity and patient classification at an adult intensive care unit (ICU). Intensive Crit Care Nurs 2007; 23: 97-103.
17. Castillo-Lorente E, Rivera-Fernandez R, Rodriguez-Elvira M et al: Tiss 76 and Tiss 28: correlation of two therapeutic activity indices on a Spanish multicenter ICU database. Int Care Med 2000; 26: 57-61.
18. Padilha KG, Cardoso Sousa R, Kimura M: Nursing workload in intensive care units: A study using the Therapeutic Intervention Scoring System -28 (TISS-28). Intensive Crit Care Nurs 2007; 23: 162-169.
19. Lefering R, Zart M, Neugebauer E: Retrospective evaluation of the Simplified Therapeutic Scoring System (TISS-28) in a surgical intensive care unit. Int Care Med 2000; 26: 1794-1802 .
20. Wing Yee Kwok W, Pak Chun Chau J, Pau Le Low L et al: The reliability and validity of the therapeutic activity index. J Crit Care 2005; 20: 257-263.
21. Kiekkas P, Brokalaki H, Manolis E: Patient severity as an indicator of nursing workload in the intensive care unit. Nursing in Critical Care 2007; 12: 34-41.
22. Pyykko AK, Ala Koko TI: Nursing staff resources in direct patient care: comparison of TISS and ICNSS. Acta Anaesthesiol Scand 2004; 48: 1003-1005.
23. Wojtaś M: Evaluation of scoring systems for measuring nursing workload in ICU. Master’s thesis. Akademia Medyczna, Poznań. 2002.
24. Bernat Adel A, Campos R: Nursing Activities Score (NAS). Our experience with nursing load calculation system based on times. Enfermeria Intensiva 2005; 16: 164-173.
25. Kväle R, Flaatten H: Changes in intensive care from 1987 to 1997 – has outcome improved? A single centre study. Int Care Med 2002; 28: 1110-1116.
26. Jakob SM, Rothen HU: Intensive care 1980-1995: change in patient characteristics, nursing workload and outcome. Int Care Med 1997; 23: 1165-1170.
27. Iapichino G, Gattioni L, Radrizzani D: Volume of activity and occupancy rate in intensive care units. Association with mortality. Int Care Med 2004; 30: 290-297.
28. Hugonnet S, Uckay I, Pittet D: Staffing level: a determinant of late onset ventilator – associated pneumonia, online: http://ccforum.com/content/111/4/R80.
29. Stanton MW: Hospital nurse staffing and quality of care. Agency for Healthcare Research and Quality. Research in action 2004; 14: AHRQ 04 – 029.
The authors are grateful to and thank Dorota Ozga, Department of Nursing, University of Rzeszów, Sylwia Miętkiewicz, Department of Intensive Therapy and Pain Management, Teaching Hospital no 2 in Poznań, Barbara Frymorgen, Department of Anaesthesiology and Intensive Therapy in Bielsko-Biała and Agnieszka Wiesiołowska, PhD, Department of Medical Statistics, Poznań University of Medical Sciences.
*Edyta K. Cudak
Zakład Pielęgniarstwa Anestezjologicznego
i Intensywnej Opieki
ul. Smoluchowskiego 11, 60-179 Poznań
tel. kom.: 0-509 105 563
fax: 0-61 655 92 66