Acta Anaesthesiol Scand 1992: 36: 6 10-6 14
The nursing care recording system. A preliminary study of a system for assessment of nursing care demands in the ICU E.
HJORTS0,
T. B U C H ’ , J. RYDING, K.
LUNDSTROM,
P. BARTRAM,L. DRAGSTED and J. QVIST
Department of Anesthesia and Intensive Care and ‘Intensive Care Unit, Herlev Hospital, University of Copenhagen, Herlev, Denmark
A new system, Nursing Care Recording (NCR), for the recording of nursing care in a general ICU is presented. NCR classifies I C U patients according to their need for intensive nursing care. Comparing the NCR with the Therapeutic Intervention Scoring System (TISS), a correlation coefficient of 0.60 was found. The main difference between the two systems was related to recording procedures allowing changes in nursing intensity within a 24-h period, reflecting patient improvement due to therapy, which was detected by NCR but not by ‘IISS.NCR can be used to estimate nursing capacity during different shifts and may be useful in the assessment of the total nursing staff necessary for a given ICU. It is suggested that NCR will allow detection of changes in the nursing care work load, whether this change is due to new activities in the unit or to alterations in the individual patient care.
Received 8 April 1991, accepted f o r publication I January 1992
K q words: Costs, intensive care, nursing care recording system; intensive care, quality of care, scoring system; patient care, nursing intensity, nursing capacity.
When the managers of a department request more nursing capacity, they should be able to document the underlying increase in the total nursing care work load. An increase in the amount of nursing care is more often caused by changes in thc severity of patients’ illness than by the introduction of new activities in the intensive care unit (ICU). The demand for documentation of the level and quality of nursing care has resulted in the development of different systems for the evaluation of nursing care in ordinary wards (1-3). These methods are based on either a factor evaluation, by which simple physical observations (“indicators”) are used to describe the different patient categories, or a prototype evaluation, by which a circumstantial description of the patients’ need for physical and psychological care is used to categorize the paticnts. The spccial circumstances in an ICU, characterized by rapid changes in the number of patients and in their condition, make it impossible to use the above-mentioned methods to document the nursing intensity in the ICU. The increasing need for comparative studies, both between different intensive care units and between different treatments, has emphasized the necessity of classifying paticnts according to their scverity of illncss and to the level of treatment. It is possible today to measure the severity of illness and thc level of therapy with the APACHE and TISS systems (4-8).Nursing intensity is depcndent on both severity of illness and
level of treatment. In brief, the special situation in the ICU has been considered in the following systems:
1. The Acute Physiology and Chronic Health Evaluation (APACHE) system measures the severity of illness by quantitating the acute changes of 14 physiological parameters and includes the patients’ chronic health status and the patient age ( 6 , 7 ) . The system is especially useful in mortality prediction in groups of patients. 2. The Therapeutic Intervention Scoring System (TISS) quantitates the type and the number of treatments. This system is therefore capable of indicating the need for nursing care or work load. It has been demonstrated that a nurse can handle the physical needs of a patient which are equivalent to 40 TISS points ( 5 ) . TISS measures the amount of time and effort used in treating critically ill patients. The system comprises 82 parameters collected after 24 h of therapy (4). 3. The Nursing Intervention Scoring System (NISS) (9) is similar to TISS, but the total number of items to be registered is 117. NISS was developed by nurses specifically for use in the ICU, whereas the primary objective of the TISS system, developed by physicians, was to evaluate the severity of illness of ICU patients. 4. A Swedish ICU nursing care system ( 10) developed by SAF (The Swedish Society of Anaesthesiologists)
T H E NCR SYSTEM
comprises eight easily observed “indicators” of care directly related to diagnosis and physical care. Having reviewed the above systems, we decided to develop our own system for measuring nursing care requirements because we want the system to focus on nursing activities in relation to therapeutic intensity and not to specific diagnosis. Furthermore, the system must continuously be used in each nursing shift to illustrate the sudden changes in nursing care demands seen in the ICU. Therefore, it is essential that the system describes the nursing activities with only a few parameters.
61 1
month (33 patients), comparing the calculated nursing intensity using NCR with the roster nursing capacity. Thus, we did not take into account the utilization of extra personnel called upon to cope with the very high work load. Qualitative assessment of the level of care was performed daily during this month. This “sufficiency estimate” was then compared to the nursing capacity utilization ratio, i.e. the calculated ratio between nursing intensity and nursing capacity. The study was approved by the Scientific Ethical Committee for the county of Copenhagen.
The Nursing Care Recording system (NCR) was developed in a general ICU treating only adult patients. The unit did not receive patients with acute myocardial infarctions, thermal burns or neurological diseases, nor did it serve as a recovery room for patients after uncomplicated surgical procedures. The basic concept of nursing care was modified primary nursing care, i.e. each nurse was responsible for all nursing activities concerning a single or a few patients and their relatives. The NCR system was based on indicators of nursing care uniquely related to intensive care and represented the most common procedures in intensive care nursing. They were related to monitoring and treatment within different organ systems (cardiovascular, pulmonary, renal systems, etc.), as shown in Table l. The indicators also reflected the nurses’ involvement in carrying out therapy and diagnostic procedures. Each indicator was assigned a weight from 1 to 3 points, depending upon the estimated duration of the procedures. To estimate the total time consumption, we used the time taken to meet the patients’ physical and psychological needs, and the time taken for care of the relatives. Expert opinion, based on a combination of the nurses’ theoretical knowledge and practical experience, was used in these estimates of nursing care time (3). The time estimates directly reflected the need for personnel. The 10 NCR indicators were scored at the end of each nursing shift, i.e. every 8 h. The score (the sum of 10 indicators) from a single shift allowed us to classify patients into one of three categories: Category no. A: Stable patients in need of intensive care monitoring and intensive nursing care (the patient “at risk”). Category no. B: Stable patients in need of intensive care therapy and specz$c intensive nursing cure (the patient’s condition is stabilized due to a specific treatment). Category no. C: Unstable patients in need of maximal intensive care therapy and spec;fic intensive nursing care (the patient’s condition is changing in spite of maximal therapeutic efforts).
RESULTS During 1 year, in 393 patients, 5733 NCR scores were recorded. The number of patient days was 191 1, and the mean length of stay was 4.9 days. The classification of the patients into three categories (A, B and C) by their NCR score is shown in Fig. 1. The number of patients in each category showed only minor variations over the months. The TISS scores varied from 11 to 66 points, and the NCR between 31 and 78. The correlation between the NCR and TISS was found to be significant, as shown in Fig. 2; the correlation coeffcient was 0.60 (s.d. = + 8.6 and P < 0.05). An even better correlation between the two systems was found by selecting the “worst” (highest) of the three NCR scores within each indicator during the first 24 h. The correlation coeffcient then increased to 0.66 and the s.d. decreased to 3.2. The three categories in the NCR system and their score ranges were established using the correlation to TISS (Table 2). Table 2 also depicts the estimated nursing time for each patient category per 8-h shift: A = 6 h, B = 12 h, and C = 16 h. In the 33 patients, we assessed nursing intensity by recording the nursing care during 489 patient-day shifts. The ratio between calculated nursing intensity and the available nursing capacity is shown in Fig. 3. The nursing intensity was almost always higher than the available nursing capacity. The “quality assessment study” in the same 33 patients showed good agreement for 28 out of 30 study days between nursing utilization capacity and the nurses’ perception of the adequacy of the available personnel. There was agreement that the unit was understaffed when the nursing utilization ratio was above 1.2.
study protoco~and patient selection The NCR system was evaluated for I year, comprising a total of 393 patients. A study was carried out comparing NCR with TISS in 100 consecutive patients during the first 4 months. The TISS scoring was performed by a group of physicians familiar with that system from previous studies (1 1) and completed at the conclusion of the first 24-h stay in the ICU. The NCR scoring was performed by the nurses. I n contrast to TISS, the NCR registration was continued for the remainder of the patients’ stay in the ICU. The NCR score for the first 24 h (the sum of three scores) was compared with the TISS score, using analysis ofcorrelation (a significance level of 5% was chosen). In addition, we studied nursing capacity utilization during a single
DISCUSSION ICU patients have been classified into three categories using TISS (5). We adopted this classification to NCR by extrapolation (Fig. 1), although the correlation between the two systems was found to be only fair (Fig. 2). The reason for the lack of a better correlation is probably due to the way different nursing events are registered. TISS describes the patient following 24 h therapy, while NCR is recorded three times during the
MATERIAL AND METHODS
612
E. H J O R T S 0 E T AL.
Table 1 NCR record form. 1 to 3 points are given to each of 10 indicators of nursing care every 8 hours. The minimal score for an ICU patient is 1 0 points per work shift. Nursing Care Recording in the ICU
I ndicalors
2 points
3 points
1 point
~~
Frequency of monitoring
> I/h
1/h
< l/h
CNS
Neurolog. ex. > 1 /h N-M relax. convulsion Confusion, agitation
Neurolog. ex. l / h N-M relax., immobile Risk of convulsion
Conscious
Respiration
Max. vent. support
Ordinary support weaning
S-PEEP inhalation of drugs
~~
Cardiovascular
Unstable on pharmacol. sup- Stable on pharmacol. support No pharmacol. support port
Gastrointestinal
> 5 drains/stomies/ gastric tube surg. wound care > 3/day
3-4 drains/stomies/ gastric tube surg. wound care 3/day
< 2 drainslstomiesl gastric tube surg. wound care I/day
Renal
Patient unstable on dialysis
Patient stable on dialysis Urine output > 150 ml/h
Urine output/h
~~
~
~~
~
~
~~
> 5 parallel infusions injec- 3-4 parallel infusions injection < 2 parallel infusions on tions > IIh 1 Ih Deroral med.
Infusions and injections Patient care
Full assistance of 2-3 persons
Full assistance of 1-2 persons
Partly self-reliant
Other monitoring and treatment assist
S-G cath cardiac arrest intubation art-ven-shunt Sengstaken tube
Central venous line pleural drainage peritoneal drainage
iv-line urine cath gastric tube arterial-line
Hematology
Blood test >2/8h other tests > 2/8h
Blood test 2/8h other tests 2/8h
Blood test 1/8h other tests 1/8h
N-M relax = Neuromuscular relaxation; S-G = Swan Ganz catheter; S-PEEP = Spontaneous - positive end expiratory pressure
same time span. With the use of NCR, patients who responded to therapy demonstrated a decreasing score over time. When we compared the worst (highest) of the three NCR values for the 24-h period with the TISS score, we expected, and found, a better correlation, although i t was non-significant, We have used the patients' classification into categories, known from TISS, to describe the nursing intensity in our unit during 1 year (Fig. 1). The results demonstrate only minor variations. With a fixed num-
.DD KP. A
B C
n 300 250
200 150 100
50
Nov.
D.E.
JM.
W.
Mu.
Apr.
May
Jun.
JuI.
Aug.
S.P.
?iii
Fig. I . The number of registrations (dayleveninglnight) for categories A, B and C during the I-year study period.
ber of beds in a unit, the nursing intensity can be shown using the total number of NCR points per day/month/year, because increasing NCR points will represent change in nursing demands due to increase in severity of illness or intensity of therapy. This could also be seen as a shift in patient categories. When assessing the time needed for nursing care, one can either measure the actual time' taken or make estimates of the overall time needed for the care. Actual time studies are paramount to make certain that overall patient care and care for relatives is comprehensively described, but none have yet been carried out in an ICU setting. The nursing care time per 8-hour shift, per day or per month could be calculated for the three patient categories based upon the NCR score (Fig. 3 ) . We estimated that 18 NCR points were equivalent to 12 nursing hours within one 8-hour shift. As mentioned, this time consumption includes both physical and psychological care of the patient and the time spent supporting the relatives. The calculated nursing intensity exceeded the actual nursing capacity available on most days (Fig. 3 ) . Thcre are two possible explanations: 1) Thc avail-
613
THE NCR SYSTEM
NCR
0
"1
0
0
0 + 2 SD
NURSE UTILIZATION
0
0'
10
I
10
20
I
I
I
I
I
30
40
50
60
70
TISS Fig. 2. The correlation between NCR and TISS. The line of regression: Y=33.7+0.5x; r=0.60, SD= k8.6; P
The nursing care recording system. A preliminary study of a system for assessment of nursing care demands in the ICU E.
HJORTS0,
T. B U C H ’ , J. RYDING, K.
LUNDSTROM,
P. BARTRAM,L. DRAGSTED and J. QVIST
Department of Anesthesia and Intensive Care and ‘Intensive Care Unit, Herlev Hospital, University of Copenhagen, Herlev, Denmark
A new system, Nursing Care Recording (NCR), for the recording of nursing care in a general ICU is presented. NCR classifies I C U patients according to their need for intensive nursing care. Comparing the NCR with the Therapeutic Intervention Scoring System (TISS), a correlation coefficient of 0.60 was found. The main difference between the two systems was related to recording procedures allowing changes in nursing intensity within a 24-h period, reflecting patient improvement due to therapy, which was detected by NCR but not by ‘IISS.NCR can be used to estimate nursing capacity during different shifts and may be useful in the assessment of the total nursing staff necessary for a given ICU. It is suggested that NCR will allow detection of changes in the nursing care work load, whether this change is due to new activities in the unit or to alterations in the individual patient care.
Received 8 April 1991, accepted f o r publication I January 1992
K q words: Costs, intensive care, nursing care recording system; intensive care, quality of care, scoring system; patient care, nursing intensity, nursing capacity.
When the managers of a department request more nursing capacity, they should be able to document the underlying increase in the total nursing care work load. An increase in the amount of nursing care is more often caused by changes in thc severity of patients’ illness than by the introduction of new activities in the intensive care unit (ICU). The demand for documentation of the level and quality of nursing care has resulted in the development of different systems for the evaluation of nursing care in ordinary wards (1-3). These methods are based on either a factor evaluation, by which simple physical observations (“indicators”) are used to describe the different patient categories, or a prototype evaluation, by which a circumstantial description of the patients’ need for physical and psychological care is used to categorize the paticnts. The spccial circumstances in an ICU, characterized by rapid changes in the number of patients and in their condition, make it impossible to use the above-mentioned methods to document the nursing intensity in the ICU. The increasing need for comparative studies, both between different intensive care units and between different treatments, has emphasized the necessity of classifying paticnts according to their scverity of illncss and to the level of treatment. It is possible today to measure the severity of illness and thc level of therapy with the APACHE and TISS systems (4-8).Nursing intensity is depcndent on both severity of illness and
level of treatment. In brief, the special situation in the ICU has been considered in the following systems:
1. The Acute Physiology and Chronic Health Evaluation (APACHE) system measures the severity of illness by quantitating the acute changes of 14 physiological parameters and includes the patients’ chronic health status and the patient age ( 6 , 7 ) . The system is especially useful in mortality prediction in groups of patients. 2. The Therapeutic Intervention Scoring System (TISS) quantitates the type and the number of treatments. This system is therefore capable of indicating the need for nursing care or work load. It has been demonstrated that a nurse can handle the physical needs of a patient which are equivalent to 40 TISS points ( 5 ) . TISS measures the amount of time and effort used in treating critically ill patients. The system comprises 82 parameters collected after 24 h of therapy (4). 3. The Nursing Intervention Scoring System (NISS) (9) is similar to TISS, but the total number of items to be registered is 117. NISS was developed by nurses specifically for use in the ICU, whereas the primary objective of the TISS system, developed by physicians, was to evaluate the severity of illness of ICU patients. 4. A Swedish ICU nursing care system ( 10) developed by SAF (The Swedish Society of Anaesthesiologists)
T H E NCR SYSTEM
comprises eight easily observed “indicators” of care directly related to diagnosis and physical care. Having reviewed the above systems, we decided to develop our own system for measuring nursing care requirements because we want the system to focus on nursing activities in relation to therapeutic intensity and not to specific diagnosis. Furthermore, the system must continuously be used in each nursing shift to illustrate the sudden changes in nursing care demands seen in the ICU. Therefore, it is essential that the system describes the nursing activities with only a few parameters.
61 1
month (33 patients), comparing the calculated nursing intensity using NCR with the roster nursing capacity. Thus, we did not take into account the utilization of extra personnel called upon to cope with the very high work load. Qualitative assessment of the level of care was performed daily during this month. This “sufficiency estimate” was then compared to the nursing capacity utilization ratio, i.e. the calculated ratio between nursing intensity and nursing capacity. The study was approved by the Scientific Ethical Committee for the county of Copenhagen.
The Nursing Care Recording system (NCR) was developed in a general ICU treating only adult patients. The unit did not receive patients with acute myocardial infarctions, thermal burns or neurological diseases, nor did it serve as a recovery room for patients after uncomplicated surgical procedures. The basic concept of nursing care was modified primary nursing care, i.e. each nurse was responsible for all nursing activities concerning a single or a few patients and their relatives. The NCR system was based on indicators of nursing care uniquely related to intensive care and represented the most common procedures in intensive care nursing. They were related to monitoring and treatment within different organ systems (cardiovascular, pulmonary, renal systems, etc.), as shown in Table l. The indicators also reflected the nurses’ involvement in carrying out therapy and diagnostic procedures. Each indicator was assigned a weight from 1 to 3 points, depending upon the estimated duration of the procedures. To estimate the total time consumption, we used the time taken to meet the patients’ physical and psychological needs, and the time taken for care of the relatives. Expert opinion, based on a combination of the nurses’ theoretical knowledge and practical experience, was used in these estimates of nursing care time (3). The time estimates directly reflected the need for personnel. The 10 NCR indicators were scored at the end of each nursing shift, i.e. every 8 h. The score (the sum of 10 indicators) from a single shift allowed us to classify patients into one of three categories: Category no. A: Stable patients in need of intensive care monitoring and intensive nursing care (the patient “at risk”). Category no. B: Stable patients in need of intensive care therapy and specz$c intensive nursing cure (the patient’s condition is stabilized due to a specific treatment). Category no. C: Unstable patients in need of maximal intensive care therapy and spec;fic intensive nursing care (the patient’s condition is changing in spite of maximal therapeutic efforts).
RESULTS During 1 year, in 393 patients, 5733 NCR scores were recorded. The number of patient days was 191 1, and the mean length of stay was 4.9 days. The classification of the patients into three categories (A, B and C) by their NCR score is shown in Fig. 1. The number of patients in each category showed only minor variations over the months. The TISS scores varied from 11 to 66 points, and the NCR between 31 and 78. The correlation between the NCR and TISS was found to be significant, as shown in Fig. 2; the correlation coeffcient was 0.60 (s.d. = + 8.6 and P < 0.05). An even better correlation between the two systems was found by selecting the “worst” (highest) of the three NCR scores within each indicator during the first 24 h. The correlation coeffcient then increased to 0.66 and the s.d. decreased to 3.2. The three categories in the NCR system and their score ranges were established using the correlation to TISS (Table 2). Table 2 also depicts the estimated nursing time for each patient category per 8-h shift: A = 6 h, B = 12 h, and C = 16 h. In the 33 patients, we assessed nursing intensity by recording the nursing care during 489 patient-day shifts. The ratio between calculated nursing intensity and the available nursing capacity is shown in Fig. 3. The nursing intensity was almost always higher than the available nursing capacity. The “quality assessment study” in the same 33 patients showed good agreement for 28 out of 30 study days between nursing utilization capacity and the nurses’ perception of the adequacy of the available personnel. There was agreement that the unit was understaffed when the nursing utilization ratio was above 1.2.
study protoco~and patient selection The NCR system was evaluated for I year, comprising a total of 393 patients. A study was carried out comparing NCR with TISS in 100 consecutive patients during the first 4 months. The TISS scoring was performed by a group of physicians familiar with that system from previous studies (1 1) and completed at the conclusion of the first 24-h stay in the ICU. The NCR scoring was performed by the nurses. I n contrast to TISS, the NCR registration was continued for the remainder of the patients’ stay in the ICU. The NCR score for the first 24 h (the sum of three scores) was compared with the TISS score, using analysis ofcorrelation (a significance level of 5% was chosen). In addition, we studied nursing capacity utilization during a single
DISCUSSION ICU patients have been classified into three categories using TISS (5). We adopted this classification to NCR by extrapolation (Fig. 1), although the correlation between the two systems was found to be only fair (Fig. 2). The reason for the lack of a better correlation is probably due to the way different nursing events are registered. TISS describes the patient following 24 h therapy, while NCR is recorded three times during the
MATERIAL AND METHODS
612
E. H J O R T S 0 E T AL.
Table 1 NCR record form. 1 to 3 points are given to each of 10 indicators of nursing care every 8 hours. The minimal score for an ICU patient is 1 0 points per work shift. Nursing Care Recording in the ICU
I ndicalors
2 points
3 points
1 point
~~
Frequency of monitoring
> I/h
1/h
< l/h
CNS
Neurolog. ex. > 1 /h N-M relax. convulsion Confusion, agitation
Neurolog. ex. l / h N-M relax., immobile Risk of convulsion
Conscious
Respiration
Max. vent. support
Ordinary support weaning
S-PEEP inhalation of drugs
~~
Cardiovascular
Unstable on pharmacol. sup- Stable on pharmacol. support No pharmacol. support port
Gastrointestinal
> 5 drains/stomies/ gastric tube surg. wound care > 3/day
3-4 drains/stomies/ gastric tube surg. wound care 3/day
< 2 drainslstomiesl gastric tube surg. wound care I/day
Renal
Patient unstable on dialysis
Patient stable on dialysis Urine output > 150 ml/h
Urine output/h
~~
~
~~
~
~
~~
> 5 parallel infusions injec- 3-4 parallel infusions injection < 2 parallel infusions on tions > IIh 1 Ih Deroral med.
Infusions and injections Patient care
Full assistance of 2-3 persons
Full assistance of 1-2 persons
Partly self-reliant
Other monitoring and treatment assist
S-G cath cardiac arrest intubation art-ven-shunt Sengstaken tube
Central venous line pleural drainage peritoneal drainage
iv-line urine cath gastric tube arterial-line
Hematology
Blood test >2/8h other tests > 2/8h
Blood test 2/8h other tests 2/8h
Blood test 1/8h other tests 1/8h
N-M relax = Neuromuscular relaxation; S-G = Swan Ganz catheter; S-PEEP = Spontaneous - positive end expiratory pressure
same time span. With the use of NCR, patients who responded to therapy demonstrated a decreasing score over time. When we compared the worst (highest) of the three NCR values for the 24-h period with the TISS score, we expected, and found, a better correlation, although i t was non-significant, We have used the patients' classification into categories, known from TISS, to describe the nursing intensity in our unit during 1 year (Fig. 1). The results demonstrate only minor variations. With a fixed num-
.DD KP. A
B C
n 300 250
200 150 100
50
Nov.
D.E.
JM.
W.
Mu.
Apr.
May
Jun.
JuI.
Aug.
S.P.
?iii
Fig. I . The number of registrations (dayleveninglnight) for categories A, B and C during the I-year study period.
ber of beds in a unit, the nursing intensity can be shown using the total number of NCR points per day/month/year, because increasing NCR points will represent change in nursing demands due to increase in severity of illness or intensity of therapy. This could also be seen as a shift in patient categories. When assessing the time needed for nursing care, one can either measure the actual time' taken or make estimates of the overall time needed for the care. Actual time studies are paramount to make certain that overall patient care and care for relatives is comprehensively described, but none have yet been carried out in an ICU setting. The nursing care time per 8-hour shift, per day or per month could be calculated for the three patient categories based upon the NCR score (Fig. 3 ) . We estimated that 18 NCR points were equivalent to 12 nursing hours within one 8-hour shift. As mentioned, this time consumption includes both physical and psychological care of the patient and the time spent supporting the relatives. The calculated nursing intensity exceeded the actual nursing capacity available on most days (Fig. 3 ) . Thcre are two possible explanations: 1) Thc avail-
613
THE NCR SYSTEM
NCR
0
"1
0
0
0 + 2 SD
NURSE UTILIZATION
0
0'
10
I
10
20
I
I
I
I
I
30
40
50
60
70
TISS Fig. 2. The correlation between NCR and TISS. The line of regression: Y=33.7+0.5x; r=0.60, SD= k8.6; P
