RESEARCH doi: 10.1111/nicc.12018
Exploration and classification of intensive care nurses’ clinical decisions: a Greek perspective Vassiliki Karra, Elizabeth D. Papathanassoglou, Chryssoula Lemonidou, Panayota Sourtzi and Margarita Giannakopoulou ABSTRACT Aim: The recording, identification, coding and classification of clinical decisions by intensive care nurses. Background: Clinical decision-making is an essential dimension of nursing practice as through this process nurses make choices to meet the goals of patient care. Intensive care nurses’ decision-making has received attention because of the complexity and urgency associated with it, however, the types of nurses’ clinical decisions have not been described systematically. Methods: Qualitative content analysis of daily diaries of clinical decisions recorded during nursing work by 23 purposefully selected intensive care nurses from three major hospitals of Greece. The process of data collection and analysis continued until the point of theoretical saturation. Findings: Eight categories of nursing clinical decisions emerged including decisions related to: (1) evaluation, (2) diagnosis, (3) prevention, (4) intervention, (5) communication with patients, (6) clinical information seeking, (7) setting of clinical priorities and (8) communication with health care professionals. Psychological assessment and support decisions were scarce, whereas patient input in care decisions appeared to be limited. The most frequent types of decisions were regarding intervention (29%), evaluation (25%) and clinical setting of priorities (17%), while clinical information seeking (3%) and communication with patients decisions (2%) were the least frequent. Additionally, recorded decisions were ranked in order of degree of urgency and of dependency on medical order. Non-urgent decisions were 78% of the total and 60% of nurses’ intervention decisions were independent of medical order and were related to basic nursing care. Conclusions: Intensive care nurses make multiple decisions that seem to be in line with the nursing process, although the latter is not officially implemented in Greek ICUs. Relevance to clinical practice: The types and frequency of clinical decisions made by intensive care nurses are related to features of ICU work environment, their professional autonomy and accountability, as well as their perceptions of their clinical role. Key words: Clinical decisions • Content analysis • Intensive care • Nurses • Qualitative study • Types
INTRODUCTION Decision-making is a fundamental cognitive function, which involves deliberate choice between at least two alternative actions, recognizing triggers for
Authors: V. Karra, RN, MSN, MHCM, PhD(c), Faculty of Nursing, National and Kapodistrian University of Athens, Athens, Greece;ED Papathanassoglou, RN, MSc, PhD, Associate Professor, Department of Nursing, Cyprus University of Technology, Latsia, Nicosia, Cyprus; C Lemonidou, RN, MSc, PhD, Professor, Faculty of Nursing, National and Kapodistrian University of Athens, Athens, Greece;P Sourtzi, RN, PhD, Associate Professor (OHN), Faculty of Nursing, National and Kapodistrian University of Athens, Athens, Greece;M Giannakopoulou, RN, PhD, Assistant Professor, Faculty of Nursing, National and Kapodistrian University of Athens, Athens, Greece Address for correspondence: Dr Margarita Giannakopoulou, Faculty of Nursing, Internal Medicine-Nursing Sector, University of Athens, 123, Papadiamantopoulou str, 11527 Goudi, Athens, Greece E-mail: [email protected]
the initiation of an action, commitment to act in a particular way and the expectation for achieving goals (Noone, 2002). Clinical decisions and their underlying processes are an integral part of health care practice. Although, clinical decision-making – along with the uncertainty embedded in this process – may be viewed as situated largely within the scope of practice of physicians, Kitson (1999) suggested that clinical decision-making is an important skill for any health care professional. Clinical judgment and decision-making are inherent in nursing work and constitute indisputable components of the nursing process. In the context of intensive care units (ICU), the ability of nurses to make clinical decisions and to act upon them, is strongly linked with improved patient outcomes (Kollef et al., 1997; Curley, 2002) and the quality and safety of
© 2014 The Authors. Nursing in Critical Care © 2014 British Association of Critical Care Nurses • Vol 19 No 2
87
Exploration and classification of intensive care nurses’ clinical decisions
patient care (Bucknall and Thomas, 1997; Bucknall, 2003).
BACKGROUND Clinical decision-making is central to the science and practice of nursing (Hardy and Smith, 2008). Thompson et al. (2004) have estimated that emergency care nurses make one decision approximately every 10 min and Bucknall (2000) reported one decision every 30 s for nurses in intensive care. Research evidence on the types of decisions in nurses’ routine practice is scarce (McCaughan et al., 2005). Thompson et al. (2001) analysed qualitative data from observation and interviews with 240 nurses and they classified nursing decisions into eleven categories: decisions related to intervention/effectiveness, targeting, prevention, timing, referral, communication, service organization/delivery/management, assessment, diagnosis, information seeking and experiential/understanding/hermeneutic. McCaughan et al. (2005) used qualitative methods to explore clinical decision-making in a purposive sample of 29 nurse practitioners and developed a seven-fold typology capturing the types of decisions nurses made on a daily basis concerning assessment, diagnosis, intervention, referral, communication, service delivery and information seeking. Previously, Bucknall (2000) described the decision-making activities of 18 critical care nurses in the actual ICU setting using content analysis of observation transcripts. Decision-making activities were categorized into three core categories: intervention, communication and evaluation. The first two classifications mentioned above, are not specific for intensive care context, whereas the third, although it focuses on critical care nursing decisions, encompasses broad categories, making it difficult to highlight the range and complexity of potential clinical judgements. Thus, in order to improve the understanding of intensive care nurses’ clinical decisions and to establish an adequate framework for multi-centre research, a comprehensive classification scheme is necessary. Moreover, no research evidence exists with regard to Greek intensive care nurses’ clinical decision-making practices, despite reports of low decisional autonomy among Greek critical care nurses (Papathanassoglou et al., 2005, 2012).
given the debate over levels of education and staffing, understanding the types and complexity of clinical decisions made by critical care nurses is imperative in order to emphasize the importance of their work and establish standards for staffing and nurse/patient ratios, as well as in setting priorities for education, support and empowerment of nurses. Moreover, the development of an ICU-specific framework for clinical decision-making is pivotal for describing, comparing and auditing nursing clinical decisions across diverse critical care environments and over time, and for evaluating the effectiveness of educational interventions aiming to improve practice.
METHODS Design A naturalistic qualitative approach was employed and content analysis of clinical diaries was conducted. Clinical decisions made by ICU nurses were recorded by themselves during the course of their shift.
Participants Twenty-three nurses from general ICUs of three major Greek hospitals were purposefully selected, based on their educational level, previous nursing experience and intensive care experience in order to gather in-depth and rich information (Holloway, 2005). ICUs were selected on the criterion of reasonable representativeness of standard practice. The specific inclusion criteria for participants were (1) registered nurses, (2) at least 5-year nursing clinical experience and (3) at least 2-year ICU experience. Participants’ background, demographic and professional data are shown in Table 1.
ETHICAL APPROVAL The study received institutional review approval by the Scientific Review Board of each participating hospital. A written informed consent was obtained from every participant. Nurses were informed about the aims of the study, the voluntary nature of the participation, their right to privacy, anonymity and confidentiality, as well as the right to withdraw from the study at any time.
Data collection AIMS AND OBJECTIVES OF THE PROJECT The aims of this study were the recording, identification, coding and classification of clinical decisions made by ICU nurses in Greek hospitals. Considering the current state of nursing development in Greece, and 88
Data collection took place over a period of 8 months (October 2010 to May 2011). An original log sheet of decisions was designed specifically for the purpose of the study, through a panel of experts and pilot testing. The decisions’ log will be hereon referred to as the ‘diary of decisions’. It comprised of (1)
© 2014 The Authors. Nursing in Critical Care © 2014 British Association of Critical Care Nurses
Exploration and classification of intensive care nurses’ clinical decisions
Table 1 Participants’ demographic, background and professional data Attribute Age (years): mean ± SD Men/women (n) University/technological education (n) Postgraduate studies (n) Nursing specialty certificate Masters in nursing Total clinical experience (years): mean ± SD Experience in ICU (years): mean ± SD
33·6 ± 2·58 (range: 28–42) 6/17 6/17 16 1 15 10 ± 2·79 (range: 6–15) 6·26 ± 2·32 (range: 2·5–12)
ICU, intensive care unit.
a socio-demographic data questionnaire, including gender, age, educational background, overall nursing experience and ICU nursing experience, (2) two questions inquiring about the specific nursing shift and day of the week when recordings were made and (3) blank sheets for notes by the participants. The recordings took place during an 8-h shift. Each participant completed one diary, therefore 23 diaries were collected. Before commencement of data collection, it was clarified that the study did not explore whether participants’ decisions were correct or not, but rather described the types of decisions and the process of clinical decision-making. In order to overcome the obstacle of time delay between the making and recording of decisions and to avoid loss of significant information, the think-aloud technique was also suggested as an alternative approach to data collection (Erickson and Simon, 1993; Kushniruk, 2001; Lundgren-Laine and Salantera, 2010). The think-aloud approach, that is the simultaneous oral commentary and action, was not well accepted by participants, who complained that narration delayed and restricted their actions. Thus, the decision was made to use the manual hand-onto paper recordings of decisions and pertinent considerations after completion of a judgment (Rycroft-Malone et al., 2004). All recordings were completed by participants during the course of 7 mornings (07:00–15:00 h), 11 evenings (15:00–23:00 h) and 5 nights (23:00–07:00 h) shifts over a period of 17 weekdays and 6 holidays.
Data analysis Diary entries were analysed by content analysis. Two researchers read all the diaries of decisions several times to familiarize themselves with the data therein. Coding and categorizing were carried out independently by each researcher (Ritchie and Spencer, 1994). Frequent meetings were held to discuss the findings and to achieve consensus when disagreements occurred (Schilling, 2006). Whole diaries of decisions were contemplated as units of analysis. Words and
sentences were considered as meaning units which encapsulate the conceptual content of the notes (Zhang and Wildemuth, 2009). Codes were assigned to specific text excerpts (Rubin and Rubin, 1995) that represented a decision. Then an initial list of coding categories was generated. Although, there was no pre-specified coding scheme, after the initial analysis, it became obvious that the emerging coding categories largely reflected processes as described in the nursing process theory. The standard nursing process in Orlando’s Nursing Process Discipline Theory (1972) encompasses assessment, diagnosis, implementation and evaluation to meet patients’ immediate needs for help. Within the course of qualitative content analysis an additional dimension of codes emerged inductively. These additional codes referred to attributes of decisions, such as frequency, urgency and degree of dependence on medical orders and they spanned all previous coding categories. In addition, a quantitative analysis of demographic and professional information data was conducted. Moreover, the frequency with which specific categories of decisions were recorded was computed along with the mean values and standard deviations of the number of decisions per participant and per category using the statistical package SPSS 17.
FINDINGS Categories of clinical decisions made by ICU nurses Eight categories of decisions emerged from the data: evaluation, prevention, intervention, communication with patient, clinical information seeking, clinical priorities’ setting and communication with heath care personnel decisions.
Evaluation decisions Participants made several distinct decisions to evaluate patient-specific parameters and data with minimum and maximum frequency of 4–37 decisions, respectively, per participant during an 8-h shift and an average frequency of 14 (±7·24) evaluation decisions per participant per an 8-h shift. Such evaluation decisions fell under the following three sub-categories:
Evaluation of physiological functions and symptoms (assessment of vital signs, interpretation of monitoring readings, use of clinical assessment scales and interpretation of laboratory test results). This sub-category encompassed the majority of recorded decisions. Assessment of pathological-physiological parameters of patients, included physiological functions such as respiratory (‘I assessed breathing frequency’, diary 1),
© 2014 The Authors. Nursing in Critical Care © 2014 British Association of Critical Care Nurses
89
Exploration and classification of intensive care nurses’ clinical decisions
cardiovascular (‘I measured the blood pressure’diary 3), renal (‘I checked the current levels of urea, creatinine’, diary 14), digestive (‘I had to monitor the residual gastric content’, diary 11), acid-base balance (‘his gases were good in the morning’, diary 5), the balance of water and electrolytes (‘she needs fluids’, diary 10, ‘I checked the potassium’, diary 17), neurological function and level of consciousness (‘he does not react to painful stimuli’, diary 21), pain (‘I asked (the patient) about the characteristics and intensity of the pain’, diary 10), sleep (‘The patient probably had a night with little or no sleep’, diary 19) and pressure ulcers (‘the bedsores have spread’, diary 16). Each participant recorded an average of 12 (±6·17) physiological assessment decisions with a minimum and maximum frequency of 3 and 33 decisions, respectively. Additionally, nurses rarely implemented physical examination techniques, such as auscultation or palpation (only four recordings), although physical examination is part of nursing school curricula in Greece.
Evaluation of effectiveness of therapeutic interventions. The evaluation of the effectiveness of therapeutic interventions was recorded several times (‘The patient needed extra analgesic’, diary 2 or ‘the patient slept well, after diming the lights’, diary 2). Participants recorded an average of 3(±2·43) decisions of this sub-category with a minimum and maximum frequency of 0 and 9 decisions per 8-h shift.
Monitoring responses to therapy. Patients’ response to treatment was found in entries, such as ‘I administered extra fluids and rechecked the central venous pressure’ (diary 10), ‘After suctioning patient’s secretions, his breath sounds were improved’(diary 1), ‘As the patient had not slept for one hour, I switched off the light to improve patient’s comfort’ (diary 2), ‘I reassessed the patient for pain within half an hour to titrate IV analgesic’ (diary 5). Participants recorded an average of three (±1·67) decisions with a minimum frequency of one and a maximum frequency of six decisions.
six (±6·76) diagnoses on average, with a minimum and maximum frequency of 1–35 decisions per 8-h shift. Diagnostic decisions were the fourth most common category of decisions made by nurses in ICU (10%, n = 128). These decisions included identification of respiratory disorders (i.e. ‘tachypnoea’, diary 13); alterations of cardiovascular (i.e. ‘supraventricular tachycardia’, diary 12), renal (i.e. ‘renal failure’, diary 17 ) and gastrointestinal (i.e. ‘gastrointestinal bleeding’, diary 3) function; disorders of fluid and acid-base balance [i.e. ‘acidosis’ (diary 15), ‘hyperpotassemia’ (diary 17)]; skin pathology [i.e. ‘cutaneous candidiasis’ (diary 23), ‘skin rash’ (diary 18)]; infections (i.e. ‘probable bacteraemia’, diary 17) and pain (i.e. ‘epigastric pain’, diary 16).
Prevention decisions Prevention decisions included (1) identification of potential threats, (2) assessment of patients’ vulnerability to potential threats, (3) identification of possible outcomes and (4) alternative plans to avoid the potential risks. Prevention decisions focused on infection control (i.e. ‘He is an active hepatitis C patient, so I must take protective measures’, diary 1), aspiration of gastric contents (i.e. ‘The patient is in great risk for aspiration due to tube feeding. He needs bed positioning >30◦ to prevent aspiration’, diary 12), development of pressure ulcers (i.e. ‘I had to move patient’s head in the middle position to relieve pressure’, diary 13), risk for obstruction of drainage devices and catheters (i.e. ‘the patient requires frequent suction of his thick bronchial secretions to prevent obstruction of the endotracheal tube’, diary 6) and risk of falls (i.e. ‘the patient was in a delirium and was likely to fall off the bed. So I had to put the bed rails up for his protection’, diary 12). Participants recorded four (±2·76) prevention decisions on average, with a minimum and maximum frequency of 1 and 10 decisions per shift. Most prevention decisions led to decisions to apply interventions to alleviate the threat for patient and staff safety.
Intervention decisions Diagnosis decisions Diagnosis decisions involved a process of interpreting signs and symptoms to reveal an underlying alteration in patient’s pathology (i.e. the combination of ‘thick urine, low central venous pressure and feeling thirsty’ were diagnosed as ‘hypovolaemia’, diary 10). The diagnoses made by nurses were framed mainly within a medical diagnosis context (i.e. use of term delirium (diary 19) rather than cognitive and psychomotor impairment). Each nurse recorded 90
During the first phase of open coding, clinical intervention decisions were classified according to the following categories of opposite pairs: (1) decision to take action/decision to take no action, (2) decisions independent of/dependent on medical orders, (3) decisions based or not based on scientific evidence (research evidence or clinical guidelines) and (4) decisions made with or without patient’s participation. These were ultimately combined to one single category of interventions decisions, which represents 28·5%
© 2014 The Authors. Nursing in Critical Care © 2014 British Association of Critical Care Nurses
Exploration and classification of intensive care nurses’ clinical decisions
(n = 368) of the total decisions made by ICU nurses. Intervention decisions exhibited a range of frequencies from 1 to 37, and an average of 16 (±8·41) distinct intervention decisions per participant per 8-h shift. Noaction decisions included instances where participants recorded and acknowledged a problem (i.e. ‘recording high temperature of 38·6’, diary 4), nonetheless, they did not take any direct action to deal with it. Intervention decisions based on medical orders (i.e. ‘administration of medication as per the nursing log sheet’ (diary 4), ‘weaning from the ventilator following a medical order’, diary 8) rarely seemed to be the result of reassessment of the patient. On the other hand, decisions for independent nursing interventions (i.e. ‘care of stage 3 pressure ulcer, changing of dressing and position’, diary 13) were made following evaluation of a variety of data (i.e. ‘return to the care plan of the previous days’ and ‘choice of a suitable dressing from the existing ones’ and ‘purulent secretion from the ulcer ’and ‘the area was exposed to moisture’, diary 13). At two instances, it was recorded that the intervention decision was supported by research evidence (‘change in the connections of the ventilator and the intravenous drug delivery devices according to the guidelines of the Centers of Disease and Control’ (diary 1) ‘treating hyperglycaemia according to the insulin infusion protocol’ (diary 7). At several instances, individualized patient-specific tolerances with regard to physiological alterations were taken into account (i.e. ‘A 120 pulse was normal for this patient’, diary 9), without any choice of action being documented. Few decisions were guided by the preferences of patients themselves (i.e. ‘I connected the T-Piece with a capnograph for monitoring because of the patient’s refusal for arterial blood gas sampling and measuring of PCO2 ’, diary 16). Most decisions were made solely by the nurse (i.e. ‘change position due to abdominal bloating and explain to patient why he should remain in that position’, diary 16) and following physicians’ orders (‘giving extra potassium following doctor’s orders’, diary 7).
Communication with patient and family decisions Communication decisions formed a separate group of clinical decisions which reflected patients’ participation in their care. Participants recorded zero to six communication decisions (mean 1 ± 1·53), when the patient was able to communicate during their shift. These decisions were the least frequent (2%, n = 25) and they focused on giving instructions to patients and their relatives about matters of care or procedural/practical matters (i.e. ‘informing the patient about the need to take fluids intravenously
rather than by the mouth’ (diary 16), ‘informing the relatives about patient’s discharge from the ICU and the time of transfer to the medical ward’ (diary 10), ‘informing the relatives about their obligations towards the Blood Transfusion Department’ (diary 4).
Patient information seeking decisions Decisions to seek details about patients’ background data, medical history, progress notes, warnings about allergies or infectious diseases and ICU handoff reports were classified in a separate category. Participants made one to five decisions to obtain additional information (an average of 1·5 ± 1·53 per 8-h shift per participant), which corresponds to 3% (n = 37) of the total recorded decisions. The main sources of additional information were oral and written reports of nurses (50%, n = 19) and physicians (32%, n = 11) and to a lesser degree the results of laboratory and imaging tests (14%, n = 5) and scientific literature evidence from print and electronic sources (4%, n = 2).
Priority setting decisions Participants prioritized individual activities frequently, but the criteria employed to determine the sequence of actions were not clear. Data show that prioritization was probably driven mainly by the need for effective management of nurses’ time. Although decisions on setting priorities made up 17% (n = 224) of all nursing decisions (mean 9·74 ± 5·94, range 0–24 decisions), approaches to setting priorities were not clear (e.g. from the most to the least important, from the very strenuous to the less strenuous or vice versa). The only clear finding that did emerge was that when two actions had to take place at the same time, the urgent one took precedence (i.e. ‘Start with the correction of hypoxaemia by increasing oxygen in the ventilator and call the doctor on call later’, diary 17).
Communication with health care personnel (HCP) decisions Participants recorded several decisions to inform other ICU colleagues about a patient’s condition and to assure continuity of care within the context of collaboration (i.e. ‘ . . . we increased the analgesic slightly. The patient responded well. I informed the doctor during his visit so that he would not lower the dose of the painkiller’, diary 23). Decisions of this type were made by the participants at an average rate of 3·7 (±2·87) per 8-h shift.
© 2014 The Authors. Nursing in Critical Care © 2014 British Association of Critical Care Nurses
91
Exploration and classification of intensive care nurses’ clinical decisions
Table 2 Classification of clinical nursing decisions in descending order of frequency Percentage (n = 1281)
Type of decisions Intervention decisions Evaluation decisions
Assessment of physiological functions and symptoms Assessment of effectiveness of therapeutic interventions Monitoring responses to therapy
28·5% (n = 368) 24·5% (n = 314)
16·7% (n = 214) 3·9% (n = 5) 3·9% (n = 5)
17·5% (n = 224) 10% (n = 128) 7·5% (n = 98) 7% (n = 87) 3% (n = 37) 2% (n = 25)
Priority setting decisions Diagnosis decisions Prevention decisions Communication with other HCP decisions Patient information seeking decisions Communication with patient and family decisions HCP, health care personnel.
Ranking and classification according to additional decision attributes (frequency, urgency, independence) Ranking by order of frequency Nurses’ clinical decisions were classified in descending order of frequency (Table 2). The most frequent decisions were regarding ‘interventions’ (28·5%, n = 368) and the less frequent ‘communication with patient and family’ (2%, n = 25).
Ranking by order of degree of urgency The categorization of clinical decisions as urgent/nonurgent was based on (1) the need for rapid intervention and (2) how threatening the condition was for the patient (Table 3). Non-urgent decisions were 75% (n = 962) of the total nursing decisions in this study. With regard to decisions related to interventions, evaluation, priority setting, patient information seeking and communication with HCP, the percentages of non-urgent decisions were three times higher than the urgent ones. With regard to diagnostic decisions the percentages of urgent and non-urgent decisions were almost equal (46%, n = 59 and 54%, n = 69, respectively). Nonetheless, in the category ‘prevention’ and ‘patient communication’ urgent decisions comprised only a small fraction of total decisions (3%, n = 3 and 4%, n = 1, respectively).
Classification according to degree of independence/dependence on medical orders Clinical decisions were further classified as independent or dependent on medical orders (Table 4). Approximately 40% (n = 148) of nursing intervention decisions made were triggered by a medical order, while the remaining 60% (n = 220) were independent. The majority of clinical decisions which were made independently were related to basic nurse care tasks 92
(22%, n = 48) (including nursing interventions related to body hygiene, positioning in bed, care of decubitus ulcers and dressing change), titration of infusion of medication based on a patient’s clinical picture (37%, n = 82), change in mechanical ventilation parameters (6%, n = 13) and renal replacement therapy (2%, n = 5).
DISCUSSION On the basis of these findings, intensive care nurses in Greek hospitals appear to make clinical decisions of similar scope as those reported in the international literature (Holtzdaw, 1998; Helmrich et al., 2001; Hijahi and Al-Ansari, 2005; Nelson et al., 2006; Thomas et al., 2006; Rose et al., 2008; Corley et al., 2009; Dubose et al., 2009). The frequency of decisions recorded in this study was lower compared to a previous research report in intensive care nurses. Specifically, one decision every 30 s was reported by Bucknall (2000) compared to only 56 decisions per 8-h shift per participant in our study. Detailed recording in conjunction with observation conducted in Bucknall’s (2000) study is likely to outweigh written self-reporting diaries, because of the richer content of speech information. Although the findings show that ICU nurses make ample clinical decisions at their daily practice, the degree to which either nurses themselves or other clinicians acknowledge this decision-making load is doubtful, because nursing decisions are scarcely documented in Greek ICUs. The low percentage of communication decisions with other HCP observed in this study is remarkable. In the future, it may be worth exploring whether there is an association between poor nursing documentation and inter/intra-professional communication in Greek ICUs. Documentation is rare either because current charting systems in Greek hospitals are obsolete and inadequate as they do not reflect nursing activities, or because ICU nurses in Greece make the decision to save nursing time by
© 2014 The Authors. Nursing in Critical Care © 2014 British Association of Critical Care Nurses
Exploration and classification of intensive care nurses’ clinical decisions
Table 3 Classification of clinical nursing decisions according to time priority Urgent decisions
% (n)
Examples
Non-urgent decisions
Interventions
25 (92)
Interventions
75 (276)
Pressure ulcer care and treatment
Priority setting
25 (56)
Priority setting
75 (168)
Evaluation Prevention
25 (78) 3 (3)
Evaluation Prevention
75 (236) 97 (95)
Communication with other HCP Diagnoses
25 (21)
Communication with other HCP Diagnoses
75 (66)
Training of patient’s relatives in feeding their patient with tracheostomy Pressure ulcer risk assessment Precautions to prevent MRSA cross infection Report on an unplanned nasogastric tube removal Pressure ulcer staging
Communication with patient and family
4 (1)
Commencement of cardiopulmonary resuscitation Start correcting hypoxaemia and subsequently call the doctor on call Evaluation of vital signs Restraining measures in patients with agitation Emergency call for an unplanned extubation Identification of ventricular tachycardia Informing relatives about patient’s discharge from ICU
Patient information seeking
25 (9)
46 (59)
Search for information on blood type and Rhesus compatibility before a blood transfusion
% (n)
54 (69)
Communication with patient and family
96 (24)
Patient information seeking
75 (28)
Examples
Informing relatives about hospital policies related to blood transfusion for their patient Seeking information from patient’s medical record
HCP, health care personnel; ICU, intensive care unit; MRSA, methicillin-resistant Staphylococcus aureus. Table 4 Classification of nursing clinical decisions in intensive care units according to level of independence Independent decisions
% (n)
Interventions Priority setting
60 (220) 83 (186)
Evaluation Prevention
91 (286) 82 (80)
Communication with other HCP Diagnoses Communication with patient and family Patient information seeking
50 (44) 97 (124) 96 (23) 46 (17)
Examples
Dependent decisions
% (n)
Patient bed positioning Control of tracheal secretions before ventilator weaning Pressure ulcer staging Universal precautions for infection control
Interventions Priority setting
40 (148) 17 (38)
Evaluation Prevention
9 (28) 18 (18)
Handover report about a patient’s clinical condition Impaired gas exchange diagnosis Informing a patient’s relatives about visitation hours Retrieving clinical guidelines with regard to the control of blood glucose levels with continuous intravenous insulin infusion
Communication with other HCP Diagnoses Communication with patient and family Patient information seeking
50 (43) 3 (4) 4 (2) 54 (20)
Examples Correction of electrolytic disorders Commencing weaning from a ventilator Measuring haemodynamic parameters Transferring patients colonized or infected with multi-resistant bacteria to isolation Informing physiotherapists on a patient’s mobility needs Diagnosing decreased cardiac output Informing the relatives about a patient’s discharge from the ICU Seeking information from medical records or attending physicians
HCP, health care personnel; ICU, intensive care unit.
limiting written documentation. The latter might be a nursing response to inadequate staffing in Greek ICUs. Indeed, in a European survey, the lowest nurseto-patient ratios have been reported in Greek ICUs (Papathanassoglou et al., 2012). Decisions pertaining to the emotional responses and psychosocial well-being of patients and families were almost absent in contrast to previous reports (Titler et al., 1991; Nelson et al., 2005). Moreover, although in many cases, clinical decisions made by participants
were individualized to specific patients, as reported by others too (Klein, 1989), these could not be regarded as patient-centred decisions, because of minimal patient input to care decisions. Such failure to involve patients in their own care may be due to the dominance of the biomedical paradigm in the ICU culture. Such medicalized culture may lead nurses to construct representations of disease reproducing the basic principles of medical science, which is centred more on the disease, rather than human responses and care (Alexias,
© 2014 The Authors. Nursing in Critical Care © 2014 British Association of Critical Care Nurses
93
Exploration and classification of intensive care nurses’ clinical decisions
2001). This is in line with the finding that almost all diagnostic decisions made by nurses were medical diagnoses, whereas, they seem to make almost no use of nursing diagnoses of patients’ responses. Nonetheless, despite the aforementioned indications of a dominant biomedical paradigm, Greek ICU nurses appeared to make fewer decisions related to physical assessment findings. The low implementation of physical examination techniques to aid clinical judgements is worth exploring, and it may imply that Greek ICU nurses are constrained within the boundaries of a conventional and obsolete professional role. Within such traditional constraints, physical assessments and the ensuing clinical decisions are viewed as belonging to the domain of medicine. This study showed that nurses’ clinical reasoning was in line with the stages of the nursing process, despite the fact that there was no evidence of actual implementation of the nursing process itself. Participants seemed rather unaware that their clinical judgements corresponded to the nursing process, since the latter is not formally employed in Greek ICUs. These results are understandable within the context of nursing education and practice in Greece. Although, the nursing process is included in Greek nursing school curricula of both academic and technically oriented higher education institutions, actual implementation of the nursing process is not enforced in clinical practice. Moreover, the great majority of registered nurses are graduates of technically oriented institutions, where the emphasis of instruction has traditionally been on performance of nursing interventions rather than on clinical reasoning. The nursing process constitutes a scientific problem solving technique with multiple causal relationships at every stage of the process, requiring completion of one step before starting the next, dealing with one problem at a time and the conscious transition from one stage to another (Wilkinson, 1996). The process of clinical judgement employed by the participants did not appear to be linear. In this investigation as well as in previous studies, experienced ICU nurses appeared to approach clinical problems in a broad fashion and make their clinical decisions using a combination of analytical thinking and intuition (Klein, 1989; Aitken and Mardegan, 2000). The range of clinical decisions made by intensive care nurses in this study was indicative of critical thinking skills and of reflective, inductive and productive thinking and rational evaluation. However, critical thinking, may be ‘blocked’ when nurses’ judgments are viewed as dependent on medical orders. On the basis of these findings, it appears that when nurses were faced with doctors’ decisions, they suspended the process 94
of evaluating clinical information. When carrying out medical orders, nurses may be hindered to exercise their best judgment to reach valid conclusions, because they may feel that their accountability is limited and that medical orders may provide them with a firm foundation when acting in a tight time-frame (Benner et al., 2008). Nonetheless, the likely case of nurses accepting the traditional authority of physicians with a collective sense of a ‘learned inability’ (PatirakiKourbani, 2003) in the absence of decisional autonomy (Fagin and Garelick, 2004) cannot be excluded. Indeed, previous studies have shown low decisional and low overall autonomy in Greek ICU nurses compared to other European ICU nurses (Papathanassoglou et al., 2005, 2012). Furthermore, this study showed that the tendency of intensive care nurses to support their decisions with research data and scientific evidence is limited, which is a finding consistent with the gap between theory and practice recorded in previous studies (Parahoo, 2000). In this study, nurses appeared to rely mainly on interpersonal sources of information and patient records, which is in line with the results of other ethnographic studies (McKnight, 2006). The findings of this study show that intensive care nurses do not routinely elicit patient and family involvement in decisions about their care in ICU, except when handling procedural/practical matters, such as family’ s obligations towards the hospital. Other researchers have reached similar conclusions about the serious lack of communication between nurses and patients (Papathanassoglou et al., 2005). Instead intensive care nurses appear to communicate with their patients mainly with regard to administrative activities or operational tasks (Crotty, 1985). This may be partially explained by considering that the majority of Greek nurses are graduates of technically and not theoretically oriented nursing curricula (Papathanassoglou et al., 2005). The limitations in communication seem to extend among the members of the ICU team because decisions to communicate with other HCP were scarce compared to the volume of information managed in such a complex clinical environment. The relatively low incidence of HCP communication decisions may imply either (1) that nurses function in a mechanistic framework where each employee is acting in isolation, or (2) that they are under a complex system where each nurse ‘filters’ the information and only transmits the important messages or unresolved issues to the rest or (3) that the context of vertical information flow is not conducive to an interactive discussion about the patient, but rather the traditional transposition of
© 2014 The Authors. Nursing in Critical Care © 2014 British Association of Critical Care Nurses
Exploration and classification of intensive care nurses’ clinical decisions
medical orders from the physician to the nurse and ‘major’ comments from the nurse to physician (ColonEmeric et al., 2006). Nonetheless, the grounds for the scarce communication decisions with other HCP are still unclear. In some cases, lack of communication appears to constrain the nurse within the boundaries of strict implementation of medical orders, whereas in other cases, nurses intervene directly to manage a clinical problem. It needs to be noted that the titration of pharmacological agents, adjustments of ventilator settings and autonomous management of renal dialysis are performed by nurses in Greek ICUs unofficially, because the legislative framework for the scope of nursing practice is obsolete and does not include provisions for such interventions (Papathanassoglou et al., 2005). Therefore, Greek ICU nurses may perform such tasks either within the context of a rapidly changing clinical situation or by delegation of tasks by physicians (Papathanassoglou et al., 2005). Although in this study, the researchers used qualitative approach to improve understanding of nurses’ decision-making in critical care natural settings, the use of ‘diary of decisions’ may have compromised the richness of the data obtained, because during the process of recording, participants had the opportunity to reflect on and to probably filter their recorded decisions. Moreover, although these results cannot be deemed representative of the entire ICU nurses’ population due to qualitative nature of the design, they may portray just about accurately the types of clinical ICU nurses’ decisions, because the selection of participants was based on their rich experience and the recordings were made at three different major ICUs.
CONCLUSIONS Clinical decisions made by Greek ICU nurses appear to span a wide array of judgments and are basically consistent with the stages of the nursing process. Although nurses individualized their decisions by taking into account patient-specific data, they rarely allowed patients to participate in and guide their care. Diagnostic decisions mainly involved medical diagnoses, whereas use of nursing diagnoses was almost absent. Moreover, ICU nurses in this study rarely made decisions regarding matters of psychosocial care for either patients or families. Furthermore, decisions relating to physical assessment findings were very scarce, whereas decisions dependent to medical orders often seemed to be made without assessing the reliability of the medical order.
IMPLICATIONS FOR PRACTICE Within the course of critical care, clinical nursing decisions are ample and they affect patient outcomes significantly. Therefore, the current practice in Greek ICUs, where the clinical reasoning of nurses and nursing decisions and assessments are scarcely documented may result in underestimation of the significance and intensity of nursing work, with unfavourable implications for continuity of care, quality care and patient safety (Huffman, 2004, Johnson et al., 2006, Keenan et al., 2008), as well as staffing levels and nurse-patient ratios (Buchan, 2006). On the basis of these results, ICU nurses need to claim tasks such as physical assessment, assessment of emotional responses and psychological support and nursing diagnosis as integral parts of nursing care. Active continuing education of nursing personnel on these issues could aid this process. Furthermore, it appears that patients and their family are led to adopt passive roles and accept nursing authority. Hence, one big challenge for ICU nursing in Greece is to redefine the concepts of holistic and patient-centred care. An important consideration for future research would be to develop a quantitative tool based on the findings of qualitative studies, so that nurses’ clinical decisions can be surveyed and compared among different clinical settings and countries. Such tools would also be valuable in assessing the effectiveness of educational interventions aiming to improve the quality of critical care nursing.
ACKNOWLEDGEMENTS This research has been co-financed by the European Union (European Social Fund – ESF) and Greek national funds through the Operational Program ‘Education and Lifelong Learning’ of the National Strategic Reference Framework (NSRF) – Research Funding Program: Heracleitus II. Investing in knowledge society through the European Social Fund.
CONFLICT OF INTEREST V. K. was involved in study design, data collection, analysis process and drafting of the manuscript. E. P. was involved in study design, data analysis process and critical revision of the manuscript. C. L. was involved in study design and critical revision of the manuscript. P. S. was involved in critical revision of the manuscript. M. G. was involved in study design, supervision of the research process and critical revision of the manuscript. All authors contributed to the writing and approval of the final version of the manuscript submitted for publication.
© 2014 The Authors. Nursing in Critical Care © 2014 British Association of Critical Care Nurses
95
Exploration and classification of intensive care nurses’ clinical decisions
WHAT IS KNOWN ABOUT THE TOPIC • Clinical decision-making is an integral part of nursing care in critical care settings. • ICU nurses’ clinical decisions affect patient outcomes, quality and safety of care, as well as they involve commitment of large amounts of nursing time, effort and resources. WHAT THIS PAPER ADDS • An ICU-specific typology of critical care nurses’ clinical decisions. • A framework for evaluating and comparing nursing clinical decisions from different critical care environments. • Qualitative and quantitative data about different types of clinical decisions, providing a better understanding of the dominant scientific paradigm of critical care nursing in Greece, and nurses’ perceptions of their clinical role.
REFERENCES Aitken LM, Mardegan KJ. (2000). ‘‘Thinking-aloud’’: data collection in the natural setting. Western Journal of Nursing Research; 22: 841–853. Alexias G. (2001). The Social Construction of Medical Discourse in Intensive Care Units (in Greek). Athens: Panteion University Publication. Benner P, Hughes RG, Sutphen M. (2008). Clinical reasoning, decision making and action: thinking critically and clinically. In: Hughes RG, (ed), Patient Safety and Quality: An EvidenceBased Handbook for Nurses. Rockville: Agency for Healthcare Research and Quality. Buchan J. (2006). Evidence of nursing shortages or a shortage of evidence? Journal of Advanced Nursing; 56: 457–458. Bucknall TK. (2000). Critical care nurses’ decision making activities in the natural clinical setting. Journal of Clinical Nursing; 9: 25–36. Bucknall T. (2003). The clinical landscape of critical care: nurses’ decision making. Journal of Advanced Nursing; 43: 310–319. Bucknall T, Thomas S. (1997). Nurses’ reflections on problems associated with decision-making in critical care settings. Journal of Advanced Nursing; 25: 229–237. Colon-Emeric CS, Ammarell N, Bailey D, Corazzini K, UtleySmith Q, Lekan-Rutledge D, Anderson RA, Piven ML. (2006). Patterns of medical and nursing staff communicating in nursing homes: implications and insights from complexity science. Qualitative Health Research; 16: 173–190. Corley A, Barnett AG, Mullany D, Fraser JF. (2009). Nurse determined assessment of cardiac output. Comparing a noninvasive cardiac output device and pulmonary artery catheter: a prospective observational study. International Journal of Nursing Studies; 46: 1291–1297. Crotty M. (1985). Communication between nurses and their patients. Nurse Education Today; 5: 130–134. Curley MA. (2002). Experienced nurses + autonomy = better patient outcomes. Pediatric Critical Care Medicine; 3: 385–386. Dubose JJ, Nomoto S, Higa L, Paolim R, Teixeira PG, Inaba K, Demetriades D, Belzberg H. (2009). Nursing involvement improves compliance with tight blood glucose in the trauma ICU: a prospective observational study. International Critical Care Nursing; 25: 101–107. Erickson KA, Simon HA. (1993). Protocol Analysis. Verbal Report as Data (Revised Edition). Cambridge: MIT Press. Fagin L, Garelick A. (2004). The doctor–nurse relationship. Advances in Psychiatric Treatment; 10: 277–286. Hardy D, Smith B. (2008). Decision making in clinical practice. British Journal of Anaesthetic and Recovery Nursing; 9: 19–21.
96
Helmrich S, Yates P, Nash R, Hobman A, Poulton V, Berggren L. (2001). Factors influencing nurses’ decisions to use nonpharmacological therapies to manage patients’ pain. Australian Journal of Advanced Nursing; 19: 27–35. Hijahi M, Al-Ansari M. (2005). Protocol-driven vs. physiciandriven electrolyte replacement in adult critically ill patients. Annals of Saudi Medicine; 25: 105–110. Holloway I. (2005). Qualitative Research in Health Care. Maidenhead: Open University Press, Mc-Graw-Hill Education. Holtzdaw BJ. (1998). New trends in thermometry for the patient in the ICU. Critical Care Nursing Quarterly; 21: 12–25. Huffman M. (2004). Redefine care delivery and documentation. Nursing Management; 35: 34–38. Johnson K, Hallsey D, Meredith RL, Warden EJ. (2006). A nursedriven system for improving patient quality outcomes. Journal of Nursing Care Quality; 21: 168–175. Keenan GM, Yakel E, Tschannen D, Mandeville M. (2008). Documentation and the Nurse Care Planning Process. In: Hughes RG, (ed), Patient Safety and Quality: An EvidenceBased Handbook for Nurses. Rockville: Agency for Healthcare Research and Quality. Kitson A. (1999). Evidence-Based Practice: A Primer for Health Care Professionals. London: Churchill Livingstone. Klein G. (1989). Advances in Man-Machine-Systems Research. Greenwich: JAI. Kollef MH, Shapiro SD, Silver P, John RE, Prentice D, Sauer S, Ahrens TS, Shannon W, Baker-Clinkscale D. (1997). A randomized trial of protocol-directed versus physiciandirected weaning from mechanical ventilation. Critical Care Medicine; 25: 567–574. Kushniruk AW. (2001). Analysis of complex decision making processes in health care: cognitive approaches to health informatics. Journal of Biomedical Informatics; 34: 365–376. Lundgren-Laine H, Salantera S. (2010). Think-aloud technique and protocol analysis in clinical decision making research. Qualitative Health Research; 20: 565–575. McCaughan D, Thompson C, Cullum N, Sheldon T, Raynor P. (2005). Nurse practitioner and practice nurses’ use of research information in clinical decision making: findings from an exploratory study. Family Practice; 22: 490–497. McKnight M. (2006). The information seeking of on-duty critical care nurses: evidence from participant observation and incontext interviews. Journal of the Medical Library Association; 94: 145–151. Nelson G, Kinjo K, Meier DE, Ahmad K, Morrison RS. (2005). When critical illness becomes chronic: informational needs of patients and families. Journal of Critical Care; 20: 79–89.
© 2014 The Authors. Nursing in Critical Care © 2014 British Association of Critical Care Nurses
Exploration and classification of intensive care nurses’ clinical decisions
Nelson JE, Mulkerin CM, Adams LL, Pronovost PJ. (2006). Improving comfort and communication in the ICU: a practical new tool for palliative care performance measurement and feedback. Quality & Safety in Health Care; 15: 264–271. Noone J. (2002). Concept analysis of decision making. Nursing Forum; 37: 21–32. Orlando IJ. (1972). The Discipline and Teaching of Nursing Process (an Evaluative Study). New York: G.P. Putman’s Sons. Papathanassoglou ED, Karanikola MN, Kalafati M, Giannakopoulou M, Lemonidou C, Albarran JW. (2012). Professional autonomy, collaboration with physicians, and moral distress among European intensive care nurses. American Journal of Critical Care; 21: e41–e52. Papathanassoglou ED, Tseroni M, Karydaki A, Vazaiou G, Kassikou J, Lavdaniti M. (2005). Practice and clinical decision making autonomy among Hellenic critical care nurses. Journal of Nursing Management; 13: 154–164. Parahoo K. (2000). Barriers to and facilitators of, research utilization among nurses in Northern Ireland. Journal of Advanced Nursing; 31: 89–98. Patiraki-Kourbani E. (2003). Greece (Hellenic Republic). In: Erickson D’Avanzo C, Geissler EM, (eds), Cultural Health Assessment. St Louis: Mosby. Ritchie J, Spencer L. (1994). Qualitative data analysis for applied policy research. In: Bryman A, Burgess R, (eds), Analyzing Qualitative Data. London: Routlege. Rose L, Nelson S, Johnston L, Presneill JJ. (2008). Workforce profile, organization structure and role responsibility for ventilation and weaning practices in Australia and New Zealand intensive care units. Journal of Clinical Nursing; 17: 1035–1043.
Rubin HJ, Rubin IS. (1995). Qualitative Interviewing: The Art of Hearing Data. Thousand Oaks: Sage Publications. Rycroft-Malone J, Seers K, Titchen A, Harvey G, Kitson A, McCormack B. (2004). What counts as evidence in evidence-based practice? Journal of Advanced Nursing; 47: 81–90. Schilling J. (2006). On the pragmatics of qualitative assessment: designing the process for content analysis. European Journal of Psychological Assessment; 22: 28–37. Thomas PJ, Paratz JD, Stanton WR, Deans R, Lipman J. (2006). Positioning practices for ventilated intensive care patients: current practice indications and contraindications. Australian Critical Care; 19: 122–126 128, 130–132. Thompson C, Cullum N, McCaughan D, Sheldon T, Raynor P. (2004). Nurses, information use, and clinical decision makingthe real world potential for evidence-based decisions in nursing. Evidence-Based Nursing; 7: 68–72. Thompson C, McCaughan D, Cullum N, Sheldon TA, Mulhall A, Thompson DR. (2001). Research information in nurses’ clinical decision-making: what is useful? Journal of Advanced Nursing; 36: 376–388. Titler MG, Cohen MZ, Craft MJ. (1991). Impact of adult critical care hospitalization: perceptions of patients, spouses, children, and nurses. Heart & Lung; 20: 174–182. Wilkinson JM. (1996). Nursing Process: A Critical Thinking Approach. Menlo Park: Addison–Wesley. Zhang Y, Wildemuth BM. (2009). Qualitative analysis of content. In: Wildemuth B, (ed), Applications of Social Research Methods to Questions Information and Library Science. Westport: Libraries Unlimited.
© 2014 The Authors. Nursing in Critical Care © 2014 British Association of Critical Care Nurses
97
Exploration and classification of intensive care nurses’ clinical decisions: a Greek perspective Vassiliki Karra, Elizabeth D. Papathanassoglou, Chryssoula Lemonidou, Panayota Sourtzi and Margarita Giannakopoulou ABSTRACT Aim: The recording, identification, coding and classification of clinical decisions by intensive care nurses. Background: Clinical decision-making is an essential dimension of nursing practice as through this process nurses make choices to meet the goals of patient care. Intensive care nurses’ decision-making has received attention because of the complexity and urgency associated with it, however, the types of nurses’ clinical decisions have not been described systematically. Methods: Qualitative content analysis of daily diaries of clinical decisions recorded during nursing work by 23 purposefully selected intensive care nurses from three major hospitals of Greece. The process of data collection and analysis continued until the point of theoretical saturation. Findings: Eight categories of nursing clinical decisions emerged including decisions related to: (1) evaluation, (2) diagnosis, (3) prevention, (4) intervention, (5) communication with patients, (6) clinical information seeking, (7) setting of clinical priorities and (8) communication with health care professionals. Psychological assessment and support decisions were scarce, whereas patient input in care decisions appeared to be limited. The most frequent types of decisions were regarding intervention (29%), evaluation (25%) and clinical setting of priorities (17%), while clinical information seeking (3%) and communication with patients decisions (2%) were the least frequent. Additionally, recorded decisions were ranked in order of degree of urgency and of dependency on medical order. Non-urgent decisions were 78% of the total and 60% of nurses’ intervention decisions were independent of medical order and were related to basic nursing care. Conclusions: Intensive care nurses make multiple decisions that seem to be in line with the nursing process, although the latter is not officially implemented in Greek ICUs. Relevance to clinical practice: The types and frequency of clinical decisions made by intensive care nurses are related to features of ICU work environment, their professional autonomy and accountability, as well as their perceptions of their clinical role. Key words: Clinical decisions • Content analysis • Intensive care • Nurses • Qualitative study • Types
INTRODUCTION Decision-making is a fundamental cognitive function, which involves deliberate choice between at least two alternative actions, recognizing triggers for
Authors: V. Karra, RN, MSN, MHCM, PhD(c), Faculty of Nursing, National and Kapodistrian University of Athens, Athens, Greece;ED Papathanassoglou, RN, MSc, PhD, Associate Professor, Department of Nursing, Cyprus University of Technology, Latsia, Nicosia, Cyprus; C Lemonidou, RN, MSc, PhD, Professor, Faculty of Nursing, National and Kapodistrian University of Athens, Athens, Greece;P Sourtzi, RN, PhD, Associate Professor (OHN), Faculty of Nursing, National and Kapodistrian University of Athens, Athens, Greece;M Giannakopoulou, RN, PhD, Assistant Professor, Faculty of Nursing, National and Kapodistrian University of Athens, Athens, Greece Address for correspondence: Dr Margarita Giannakopoulou, Faculty of Nursing, Internal Medicine-Nursing Sector, University of Athens, 123, Papadiamantopoulou str, 11527 Goudi, Athens, Greece E-mail: [email protected]
the initiation of an action, commitment to act in a particular way and the expectation for achieving goals (Noone, 2002). Clinical decisions and their underlying processes are an integral part of health care practice. Although, clinical decision-making – along with the uncertainty embedded in this process – may be viewed as situated largely within the scope of practice of physicians, Kitson (1999) suggested that clinical decision-making is an important skill for any health care professional. Clinical judgment and decision-making are inherent in nursing work and constitute indisputable components of the nursing process. In the context of intensive care units (ICU), the ability of nurses to make clinical decisions and to act upon them, is strongly linked with improved patient outcomes (Kollef et al., 1997; Curley, 2002) and the quality and safety of
© 2014 The Authors. Nursing in Critical Care © 2014 British Association of Critical Care Nurses • Vol 19 No 2
87
Exploration and classification of intensive care nurses’ clinical decisions
patient care (Bucknall and Thomas, 1997; Bucknall, 2003).
BACKGROUND Clinical decision-making is central to the science and practice of nursing (Hardy and Smith, 2008). Thompson et al. (2004) have estimated that emergency care nurses make one decision approximately every 10 min and Bucknall (2000) reported one decision every 30 s for nurses in intensive care. Research evidence on the types of decisions in nurses’ routine practice is scarce (McCaughan et al., 2005). Thompson et al. (2001) analysed qualitative data from observation and interviews with 240 nurses and they classified nursing decisions into eleven categories: decisions related to intervention/effectiveness, targeting, prevention, timing, referral, communication, service organization/delivery/management, assessment, diagnosis, information seeking and experiential/understanding/hermeneutic. McCaughan et al. (2005) used qualitative methods to explore clinical decision-making in a purposive sample of 29 nurse practitioners and developed a seven-fold typology capturing the types of decisions nurses made on a daily basis concerning assessment, diagnosis, intervention, referral, communication, service delivery and information seeking. Previously, Bucknall (2000) described the decision-making activities of 18 critical care nurses in the actual ICU setting using content analysis of observation transcripts. Decision-making activities were categorized into three core categories: intervention, communication and evaluation. The first two classifications mentioned above, are not specific for intensive care context, whereas the third, although it focuses on critical care nursing decisions, encompasses broad categories, making it difficult to highlight the range and complexity of potential clinical judgements. Thus, in order to improve the understanding of intensive care nurses’ clinical decisions and to establish an adequate framework for multi-centre research, a comprehensive classification scheme is necessary. Moreover, no research evidence exists with regard to Greek intensive care nurses’ clinical decision-making practices, despite reports of low decisional autonomy among Greek critical care nurses (Papathanassoglou et al., 2005, 2012).
given the debate over levels of education and staffing, understanding the types and complexity of clinical decisions made by critical care nurses is imperative in order to emphasize the importance of their work and establish standards for staffing and nurse/patient ratios, as well as in setting priorities for education, support and empowerment of nurses. Moreover, the development of an ICU-specific framework for clinical decision-making is pivotal for describing, comparing and auditing nursing clinical decisions across diverse critical care environments and over time, and for evaluating the effectiveness of educational interventions aiming to improve practice.
METHODS Design A naturalistic qualitative approach was employed and content analysis of clinical diaries was conducted. Clinical decisions made by ICU nurses were recorded by themselves during the course of their shift.
Participants Twenty-three nurses from general ICUs of three major Greek hospitals were purposefully selected, based on their educational level, previous nursing experience and intensive care experience in order to gather in-depth and rich information (Holloway, 2005). ICUs were selected on the criterion of reasonable representativeness of standard practice. The specific inclusion criteria for participants were (1) registered nurses, (2) at least 5-year nursing clinical experience and (3) at least 2-year ICU experience. Participants’ background, demographic and professional data are shown in Table 1.
ETHICAL APPROVAL The study received institutional review approval by the Scientific Review Board of each participating hospital. A written informed consent was obtained from every participant. Nurses were informed about the aims of the study, the voluntary nature of the participation, their right to privacy, anonymity and confidentiality, as well as the right to withdraw from the study at any time.
Data collection AIMS AND OBJECTIVES OF THE PROJECT The aims of this study were the recording, identification, coding and classification of clinical decisions made by ICU nurses in Greek hospitals. Considering the current state of nursing development in Greece, and 88
Data collection took place over a period of 8 months (October 2010 to May 2011). An original log sheet of decisions was designed specifically for the purpose of the study, through a panel of experts and pilot testing. The decisions’ log will be hereon referred to as the ‘diary of decisions’. It comprised of (1)
© 2014 The Authors. Nursing in Critical Care © 2014 British Association of Critical Care Nurses
Exploration and classification of intensive care nurses’ clinical decisions
Table 1 Participants’ demographic, background and professional data Attribute Age (years): mean ± SD Men/women (n) University/technological education (n) Postgraduate studies (n) Nursing specialty certificate Masters in nursing Total clinical experience (years): mean ± SD Experience in ICU (years): mean ± SD
33·6 ± 2·58 (range: 28–42) 6/17 6/17 16 1 15 10 ± 2·79 (range: 6–15) 6·26 ± 2·32 (range: 2·5–12)
ICU, intensive care unit.
a socio-demographic data questionnaire, including gender, age, educational background, overall nursing experience and ICU nursing experience, (2) two questions inquiring about the specific nursing shift and day of the week when recordings were made and (3) blank sheets for notes by the participants. The recordings took place during an 8-h shift. Each participant completed one diary, therefore 23 diaries were collected. Before commencement of data collection, it was clarified that the study did not explore whether participants’ decisions were correct or not, but rather described the types of decisions and the process of clinical decision-making. In order to overcome the obstacle of time delay between the making and recording of decisions and to avoid loss of significant information, the think-aloud technique was also suggested as an alternative approach to data collection (Erickson and Simon, 1993; Kushniruk, 2001; Lundgren-Laine and Salantera, 2010). The think-aloud approach, that is the simultaneous oral commentary and action, was not well accepted by participants, who complained that narration delayed and restricted their actions. Thus, the decision was made to use the manual hand-onto paper recordings of decisions and pertinent considerations after completion of a judgment (Rycroft-Malone et al., 2004). All recordings were completed by participants during the course of 7 mornings (07:00–15:00 h), 11 evenings (15:00–23:00 h) and 5 nights (23:00–07:00 h) shifts over a period of 17 weekdays and 6 holidays.
Data analysis Diary entries were analysed by content analysis. Two researchers read all the diaries of decisions several times to familiarize themselves with the data therein. Coding and categorizing were carried out independently by each researcher (Ritchie and Spencer, 1994). Frequent meetings were held to discuss the findings and to achieve consensus when disagreements occurred (Schilling, 2006). Whole diaries of decisions were contemplated as units of analysis. Words and
sentences were considered as meaning units which encapsulate the conceptual content of the notes (Zhang and Wildemuth, 2009). Codes were assigned to specific text excerpts (Rubin and Rubin, 1995) that represented a decision. Then an initial list of coding categories was generated. Although, there was no pre-specified coding scheme, after the initial analysis, it became obvious that the emerging coding categories largely reflected processes as described in the nursing process theory. The standard nursing process in Orlando’s Nursing Process Discipline Theory (1972) encompasses assessment, diagnosis, implementation and evaluation to meet patients’ immediate needs for help. Within the course of qualitative content analysis an additional dimension of codes emerged inductively. These additional codes referred to attributes of decisions, such as frequency, urgency and degree of dependence on medical orders and they spanned all previous coding categories. In addition, a quantitative analysis of demographic and professional information data was conducted. Moreover, the frequency with which specific categories of decisions were recorded was computed along with the mean values and standard deviations of the number of decisions per participant and per category using the statistical package SPSS 17.
FINDINGS Categories of clinical decisions made by ICU nurses Eight categories of decisions emerged from the data: evaluation, prevention, intervention, communication with patient, clinical information seeking, clinical priorities’ setting and communication with heath care personnel decisions.
Evaluation decisions Participants made several distinct decisions to evaluate patient-specific parameters and data with minimum and maximum frequency of 4–37 decisions, respectively, per participant during an 8-h shift and an average frequency of 14 (±7·24) evaluation decisions per participant per an 8-h shift. Such evaluation decisions fell under the following three sub-categories:
Evaluation of physiological functions and symptoms (assessment of vital signs, interpretation of monitoring readings, use of clinical assessment scales and interpretation of laboratory test results). This sub-category encompassed the majority of recorded decisions. Assessment of pathological-physiological parameters of patients, included physiological functions such as respiratory (‘I assessed breathing frequency’, diary 1),
© 2014 The Authors. Nursing in Critical Care © 2014 British Association of Critical Care Nurses
89
Exploration and classification of intensive care nurses’ clinical decisions
cardiovascular (‘I measured the blood pressure’diary 3), renal (‘I checked the current levels of urea, creatinine’, diary 14), digestive (‘I had to monitor the residual gastric content’, diary 11), acid-base balance (‘his gases were good in the morning’, diary 5), the balance of water and electrolytes (‘she needs fluids’, diary 10, ‘I checked the potassium’, diary 17), neurological function and level of consciousness (‘he does not react to painful stimuli’, diary 21), pain (‘I asked (the patient) about the characteristics and intensity of the pain’, diary 10), sleep (‘The patient probably had a night with little or no sleep’, diary 19) and pressure ulcers (‘the bedsores have spread’, diary 16). Each participant recorded an average of 12 (±6·17) physiological assessment decisions with a minimum and maximum frequency of 3 and 33 decisions, respectively. Additionally, nurses rarely implemented physical examination techniques, such as auscultation or palpation (only four recordings), although physical examination is part of nursing school curricula in Greece.
Evaluation of effectiveness of therapeutic interventions. The evaluation of the effectiveness of therapeutic interventions was recorded several times (‘The patient needed extra analgesic’, diary 2 or ‘the patient slept well, after diming the lights’, diary 2). Participants recorded an average of 3(±2·43) decisions of this sub-category with a minimum and maximum frequency of 0 and 9 decisions per 8-h shift.
Monitoring responses to therapy. Patients’ response to treatment was found in entries, such as ‘I administered extra fluids and rechecked the central venous pressure’ (diary 10), ‘After suctioning patient’s secretions, his breath sounds were improved’(diary 1), ‘As the patient had not slept for one hour, I switched off the light to improve patient’s comfort’ (diary 2), ‘I reassessed the patient for pain within half an hour to titrate IV analgesic’ (diary 5). Participants recorded an average of three (±1·67) decisions with a minimum frequency of one and a maximum frequency of six decisions.
six (±6·76) diagnoses on average, with a minimum and maximum frequency of 1–35 decisions per 8-h shift. Diagnostic decisions were the fourth most common category of decisions made by nurses in ICU (10%, n = 128). These decisions included identification of respiratory disorders (i.e. ‘tachypnoea’, diary 13); alterations of cardiovascular (i.e. ‘supraventricular tachycardia’, diary 12), renal (i.e. ‘renal failure’, diary 17 ) and gastrointestinal (i.e. ‘gastrointestinal bleeding’, diary 3) function; disorders of fluid and acid-base balance [i.e. ‘acidosis’ (diary 15), ‘hyperpotassemia’ (diary 17)]; skin pathology [i.e. ‘cutaneous candidiasis’ (diary 23), ‘skin rash’ (diary 18)]; infections (i.e. ‘probable bacteraemia’, diary 17) and pain (i.e. ‘epigastric pain’, diary 16).
Prevention decisions Prevention decisions included (1) identification of potential threats, (2) assessment of patients’ vulnerability to potential threats, (3) identification of possible outcomes and (4) alternative plans to avoid the potential risks. Prevention decisions focused on infection control (i.e. ‘He is an active hepatitis C patient, so I must take protective measures’, diary 1), aspiration of gastric contents (i.e. ‘The patient is in great risk for aspiration due to tube feeding. He needs bed positioning >30◦ to prevent aspiration’, diary 12), development of pressure ulcers (i.e. ‘I had to move patient’s head in the middle position to relieve pressure’, diary 13), risk for obstruction of drainage devices and catheters (i.e. ‘the patient requires frequent suction of his thick bronchial secretions to prevent obstruction of the endotracheal tube’, diary 6) and risk of falls (i.e. ‘the patient was in a delirium and was likely to fall off the bed. So I had to put the bed rails up for his protection’, diary 12). Participants recorded four (±2·76) prevention decisions on average, with a minimum and maximum frequency of 1 and 10 decisions per shift. Most prevention decisions led to decisions to apply interventions to alleviate the threat for patient and staff safety.
Intervention decisions Diagnosis decisions Diagnosis decisions involved a process of interpreting signs and symptoms to reveal an underlying alteration in patient’s pathology (i.e. the combination of ‘thick urine, low central venous pressure and feeling thirsty’ were diagnosed as ‘hypovolaemia’, diary 10). The diagnoses made by nurses were framed mainly within a medical diagnosis context (i.e. use of term delirium (diary 19) rather than cognitive and psychomotor impairment). Each nurse recorded 90
During the first phase of open coding, clinical intervention decisions were classified according to the following categories of opposite pairs: (1) decision to take action/decision to take no action, (2) decisions independent of/dependent on medical orders, (3) decisions based or not based on scientific evidence (research evidence or clinical guidelines) and (4) decisions made with or without patient’s participation. These were ultimately combined to one single category of interventions decisions, which represents 28·5%
© 2014 The Authors. Nursing in Critical Care © 2014 British Association of Critical Care Nurses
Exploration and classification of intensive care nurses’ clinical decisions
(n = 368) of the total decisions made by ICU nurses. Intervention decisions exhibited a range of frequencies from 1 to 37, and an average of 16 (±8·41) distinct intervention decisions per participant per 8-h shift. Noaction decisions included instances where participants recorded and acknowledged a problem (i.e. ‘recording high temperature of 38·6’, diary 4), nonetheless, they did not take any direct action to deal with it. Intervention decisions based on medical orders (i.e. ‘administration of medication as per the nursing log sheet’ (diary 4), ‘weaning from the ventilator following a medical order’, diary 8) rarely seemed to be the result of reassessment of the patient. On the other hand, decisions for independent nursing interventions (i.e. ‘care of stage 3 pressure ulcer, changing of dressing and position’, diary 13) were made following evaluation of a variety of data (i.e. ‘return to the care plan of the previous days’ and ‘choice of a suitable dressing from the existing ones’ and ‘purulent secretion from the ulcer ’and ‘the area was exposed to moisture’, diary 13). At two instances, it was recorded that the intervention decision was supported by research evidence (‘change in the connections of the ventilator and the intravenous drug delivery devices according to the guidelines of the Centers of Disease and Control’ (diary 1) ‘treating hyperglycaemia according to the insulin infusion protocol’ (diary 7). At several instances, individualized patient-specific tolerances with regard to physiological alterations were taken into account (i.e. ‘A 120 pulse was normal for this patient’, diary 9), without any choice of action being documented. Few decisions were guided by the preferences of patients themselves (i.e. ‘I connected the T-Piece with a capnograph for monitoring because of the patient’s refusal for arterial blood gas sampling and measuring of PCO2 ’, diary 16). Most decisions were made solely by the nurse (i.e. ‘change position due to abdominal bloating and explain to patient why he should remain in that position’, diary 16) and following physicians’ orders (‘giving extra potassium following doctor’s orders’, diary 7).
Communication with patient and family decisions Communication decisions formed a separate group of clinical decisions which reflected patients’ participation in their care. Participants recorded zero to six communication decisions (mean 1 ± 1·53), when the patient was able to communicate during their shift. These decisions were the least frequent (2%, n = 25) and they focused on giving instructions to patients and their relatives about matters of care or procedural/practical matters (i.e. ‘informing the patient about the need to take fluids intravenously
rather than by the mouth’ (diary 16), ‘informing the relatives about patient’s discharge from the ICU and the time of transfer to the medical ward’ (diary 10), ‘informing the relatives about their obligations towards the Blood Transfusion Department’ (diary 4).
Patient information seeking decisions Decisions to seek details about patients’ background data, medical history, progress notes, warnings about allergies or infectious diseases and ICU handoff reports were classified in a separate category. Participants made one to five decisions to obtain additional information (an average of 1·5 ± 1·53 per 8-h shift per participant), which corresponds to 3% (n = 37) of the total recorded decisions. The main sources of additional information were oral and written reports of nurses (50%, n = 19) and physicians (32%, n = 11) and to a lesser degree the results of laboratory and imaging tests (14%, n = 5) and scientific literature evidence from print and electronic sources (4%, n = 2).
Priority setting decisions Participants prioritized individual activities frequently, but the criteria employed to determine the sequence of actions were not clear. Data show that prioritization was probably driven mainly by the need for effective management of nurses’ time. Although decisions on setting priorities made up 17% (n = 224) of all nursing decisions (mean 9·74 ± 5·94, range 0–24 decisions), approaches to setting priorities were not clear (e.g. from the most to the least important, from the very strenuous to the less strenuous or vice versa). The only clear finding that did emerge was that when two actions had to take place at the same time, the urgent one took precedence (i.e. ‘Start with the correction of hypoxaemia by increasing oxygen in the ventilator and call the doctor on call later’, diary 17).
Communication with health care personnel (HCP) decisions Participants recorded several decisions to inform other ICU colleagues about a patient’s condition and to assure continuity of care within the context of collaboration (i.e. ‘ . . . we increased the analgesic slightly. The patient responded well. I informed the doctor during his visit so that he would not lower the dose of the painkiller’, diary 23). Decisions of this type were made by the participants at an average rate of 3·7 (±2·87) per 8-h shift.
© 2014 The Authors. Nursing in Critical Care © 2014 British Association of Critical Care Nurses
91
Exploration and classification of intensive care nurses’ clinical decisions
Table 2 Classification of clinical nursing decisions in descending order of frequency Percentage (n = 1281)
Type of decisions Intervention decisions Evaluation decisions
Assessment of physiological functions and symptoms Assessment of effectiveness of therapeutic interventions Monitoring responses to therapy
28·5% (n = 368) 24·5% (n = 314)
16·7% (n = 214) 3·9% (n = 5) 3·9% (n = 5)
17·5% (n = 224) 10% (n = 128) 7·5% (n = 98) 7% (n = 87) 3% (n = 37) 2% (n = 25)
Priority setting decisions Diagnosis decisions Prevention decisions Communication with other HCP decisions Patient information seeking decisions Communication with patient and family decisions HCP, health care personnel.
Ranking and classification according to additional decision attributes (frequency, urgency, independence) Ranking by order of frequency Nurses’ clinical decisions were classified in descending order of frequency (Table 2). The most frequent decisions were regarding ‘interventions’ (28·5%, n = 368) and the less frequent ‘communication with patient and family’ (2%, n = 25).
Ranking by order of degree of urgency The categorization of clinical decisions as urgent/nonurgent was based on (1) the need for rapid intervention and (2) how threatening the condition was for the patient (Table 3). Non-urgent decisions were 75% (n = 962) of the total nursing decisions in this study. With regard to decisions related to interventions, evaluation, priority setting, patient information seeking and communication with HCP, the percentages of non-urgent decisions were three times higher than the urgent ones. With regard to diagnostic decisions the percentages of urgent and non-urgent decisions were almost equal (46%, n = 59 and 54%, n = 69, respectively). Nonetheless, in the category ‘prevention’ and ‘patient communication’ urgent decisions comprised only a small fraction of total decisions (3%, n = 3 and 4%, n = 1, respectively).
Classification according to degree of independence/dependence on medical orders Clinical decisions were further classified as independent or dependent on medical orders (Table 4). Approximately 40% (n = 148) of nursing intervention decisions made were triggered by a medical order, while the remaining 60% (n = 220) were independent. The majority of clinical decisions which were made independently were related to basic nurse care tasks 92
(22%, n = 48) (including nursing interventions related to body hygiene, positioning in bed, care of decubitus ulcers and dressing change), titration of infusion of medication based on a patient’s clinical picture (37%, n = 82), change in mechanical ventilation parameters (6%, n = 13) and renal replacement therapy (2%, n = 5).
DISCUSSION On the basis of these findings, intensive care nurses in Greek hospitals appear to make clinical decisions of similar scope as those reported in the international literature (Holtzdaw, 1998; Helmrich et al., 2001; Hijahi and Al-Ansari, 2005; Nelson et al., 2006; Thomas et al., 2006; Rose et al., 2008; Corley et al., 2009; Dubose et al., 2009). The frequency of decisions recorded in this study was lower compared to a previous research report in intensive care nurses. Specifically, one decision every 30 s was reported by Bucknall (2000) compared to only 56 decisions per 8-h shift per participant in our study. Detailed recording in conjunction with observation conducted in Bucknall’s (2000) study is likely to outweigh written self-reporting diaries, because of the richer content of speech information. Although the findings show that ICU nurses make ample clinical decisions at their daily practice, the degree to which either nurses themselves or other clinicians acknowledge this decision-making load is doubtful, because nursing decisions are scarcely documented in Greek ICUs. The low percentage of communication decisions with other HCP observed in this study is remarkable. In the future, it may be worth exploring whether there is an association between poor nursing documentation and inter/intra-professional communication in Greek ICUs. Documentation is rare either because current charting systems in Greek hospitals are obsolete and inadequate as they do not reflect nursing activities, or because ICU nurses in Greece make the decision to save nursing time by
© 2014 The Authors. Nursing in Critical Care © 2014 British Association of Critical Care Nurses
Exploration and classification of intensive care nurses’ clinical decisions
Table 3 Classification of clinical nursing decisions according to time priority Urgent decisions
% (n)
Examples
Non-urgent decisions
Interventions
25 (92)
Interventions
75 (276)
Pressure ulcer care and treatment
Priority setting
25 (56)
Priority setting
75 (168)
Evaluation Prevention
25 (78) 3 (3)
Evaluation Prevention
75 (236) 97 (95)
Communication with other HCP Diagnoses
25 (21)
Communication with other HCP Diagnoses
75 (66)
Training of patient’s relatives in feeding their patient with tracheostomy Pressure ulcer risk assessment Precautions to prevent MRSA cross infection Report on an unplanned nasogastric tube removal Pressure ulcer staging
Communication with patient and family
4 (1)
Commencement of cardiopulmonary resuscitation Start correcting hypoxaemia and subsequently call the doctor on call Evaluation of vital signs Restraining measures in patients with agitation Emergency call for an unplanned extubation Identification of ventricular tachycardia Informing relatives about patient’s discharge from ICU
Patient information seeking
25 (9)
46 (59)
Search for information on blood type and Rhesus compatibility before a blood transfusion
% (n)
54 (69)
Communication with patient and family
96 (24)
Patient information seeking
75 (28)
Examples
Informing relatives about hospital policies related to blood transfusion for their patient Seeking information from patient’s medical record
HCP, health care personnel; ICU, intensive care unit; MRSA, methicillin-resistant Staphylococcus aureus. Table 4 Classification of nursing clinical decisions in intensive care units according to level of independence Independent decisions
% (n)
Interventions Priority setting
60 (220) 83 (186)
Evaluation Prevention
91 (286) 82 (80)
Communication with other HCP Diagnoses Communication with patient and family Patient information seeking
50 (44) 97 (124) 96 (23) 46 (17)
Examples
Dependent decisions
% (n)
Patient bed positioning Control of tracheal secretions before ventilator weaning Pressure ulcer staging Universal precautions for infection control
Interventions Priority setting
40 (148) 17 (38)
Evaluation Prevention
9 (28) 18 (18)
Handover report about a patient’s clinical condition Impaired gas exchange diagnosis Informing a patient’s relatives about visitation hours Retrieving clinical guidelines with regard to the control of blood glucose levels with continuous intravenous insulin infusion
Communication with other HCP Diagnoses Communication with patient and family Patient information seeking
50 (43) 3 (4) 4 (2) 54 (20)
Examples Correction of electrolytic disorders Commencing weaning from a ventilator Measuring haemodynamic parameters Transferring patients colonized or infected with multi-resistant bacteria to isolation Informing physiotherapists on a patient’s mobility needs Diagnosing decreased cardiac output Informing the relatives about a patient’s discharge from the ICU Seeking information from medical records or attending physicians
HCP, health care personnel; ICU, intensive care unit.
limiting written documentation. The latter might be a nursing response to inadequate staffing in Greek ICUs. Indeed, in a European survey, the lowest nurseto-patient ratios have been reported in Greek ICUs (Papathanassoglou et al., 2012). Decisions pertaining to the emotional responses and psychosocial well-being of patients and families were almost absent in contrast to previous reports (Titler et al., 1991; Nelson et al., 2005). Moreover, although in many cases, clinical decisions made by participants
were individualized to specific patients, as reported by others too (Klein, 1989), these could not be regarded as patient-centred decisions, because of minimal patient input to care decisions. Such failure to involve patients in their own care may be due to the dominance of the biomedical paradigm in the ICU culture. Such medicalized culture may lead nurses to construct representations of disease reproducing the basic principles of medical science, which is centred more on the disease, rather than human responses and care (Alexias,
© 2014 The Authors. Nursing in Critical Care © 2014 British Association of Critical Care Nurses
93
Exploration and classification of intensive care nurses’ clinical decisions
2001). This is in line with the finding that almost all diagnostic decisions made by nurses were medical diagnoses, whereas, they seem to make almost no use of nursing diagnoses of patients’ responses. Nonetheless, despite the aforementioned indications of a dominant biomedical paradigm, Greek ICU nurses appeared to make fewer decisions related to physical assessment findings. The low implementation of physical examination techniques to aid clinical judgements is worth exploring, and it may imply that Greek ICU nurses are constrained within the boundaries of a conventional and obsolete professional role. Within such traditional constraints, physical assessments and the ensuing clinical decisions are viewed as belonging to the domain of medicine. This study showed that nurses’ clinical reasoning was in line with the stages of the nursing process, despite the fact that there was no evidence of actual implementation of the nursing process itself. Participants seemed rather unaware that their clinical judgements corresponded to the nursing process, since the latter is not formally employed in Greek ICUs. These results are understandable within the context of nursing education and practice in Greece. Although, the nursing process is included in Greek nursing school curricula of both academic and technically oriented higher education institutions, actual implementation of the nursing process is not enforced in clinical practice. Moreover, the great majority of registered nurses are graduates of technically oriented institutions, where the emphasis of instruction has traditionally been on performance of nursing interventions rather than on clinical reasoning. The nursing process constitutes a scientific problem solving technique with multiple causal relationships at every stage of the process, requiring completion of one step before starting the next, dealing with one problem at a time and the conscious transition from one stage to another (Wilkinson, 1996). The process of clinical judgement employed by the participants did not appear to be linear. In this investigation as well as in previous studies, experienced ICU nurses appeared to approach clinical problems in a broad fashion and make their clinical decisions using a combination of analytical thinking and intuition (Klein, 1989; Aitken and Mardegan, 2000). The range of clinical decisions made by intensive care nurses in this study was indicative of critical thinking skills and of reflective, inductive and productive thinking and rational evaluation. However, critical thinking, may be ‘blocked’ when nurses’ judgments are viewed as dependent on medical orders. On the basis of these findings, it appears that when nurses were faced with doctors’ decisions, they suspended the process 94
of evaluating clinical information. When carrying out medical orders, nurses may be hindered to exercise their best judgment to reach valid conclusions, because they may feel that their accountability is limited and that medical orders may provide them with a firm foundation when acting in a tight time-frame (Benner et al., 2008). Nonetheless, the likely case of nurses accepting the traditional authority of physicians with a collective sense of a ‘learned inability’ (PatirakiKourbani, 2003) in the absence of decisional autonomy (Fagin and Garelick, 2004) cannot be excluded. Indeed, previous studies have shown low decisional and low overall autonomy in Greek ICU nurses compared to other European ICU nurses (Papathanassoglou et al., 2005, 2012). Furthermore, this study showed that the tendency of intensive care nurses to support their decisions with research data and scientific evidence is limited, which is a finding consistent with the gap between theory and practice recorded in previous studies (Parahoo, 2000). In this study, nurses appeared to rely mainly on interpersonal sources of information and patient records, which is in line with the results of other ethnographic studies (McKnight, 2006). The findings of this study show that intensive care nurses do not routinely elicit patient and family involvement in decisions about their care in ICU, except when handling procedural/practical matters, such as family’ s obligations towards the hospital. Other researchers have reached similar conclusions about the serious lack of communication between nurses and patients (Papathanassoglou et al., 2005). Instead intensive care nurses appear to communicate with their patients mainly with regard to administrative activities or operational tasks (Crotty, 1985). This may be partially explained by considering that the majority of Greek nurses are graduates of technically and not theoretically oriented nursing curricula (Papathanassoglou et al., 2005). The limitations in communication seem to extend among the members of the ICU team because decisions to communicate with other HCP were scarce compared to the volume of information managed in such a complex clinical environment. The relatively low incidence of HCP communication decisions may imply either (1) that nurses function in a mechanistic framework where each employee is acting in isolation, or (2) that they are under a complex system where each nurse ‘filters’ the information and only transmits the important messages or unresolved issues to the rest or (3) that the context of vertical information flow is not conducive to an interactive discussion about the patient, but rather the traditional transposition of
© 2014 The Authors. Nursing in Critical Care © 2014 British Association of Critical Care Nurses
Exploration and classification of intensive care nurses’ clinical decisions
medical orders from the physician to the nurse and ‘major’ comments from the nurse to physician (ColonEmeric et al., 2006). Nonetheless, the grounds for the scarce communication decisions with other HCP are still unclear. In some cases, lack of communication appears to constrain the nurse within the boundaries of strict implementation of medical orders, whereas in other cases, nurses intervene directly to manage a clinical problem. It needs to be noted that the titration of pharmacological agents, adjustments of ventilator settings and autonomous management of renal dialysis are performed by nurses in Greek ICUs unofficially, because the legislative framework for the scope of nursing practice is obsolete and does not include provisions for such interventions (Papathanassoglou et al., 2005). Therefore, Greek ICU nurses may perform such tasks either within the context of a rapidly changing clinical situation or by delegation of tasks by physicians (Papathanassoglou et al., 2005). Although in this study, the researchers used qualitative approach to improve understanding of nurses’ decision-making in critical care natural settings, the use of ‘diary of decisions’ may have compromised the richness of the data obtained, because during the process of recording, participants had the opportunity to reflect on and to probably filter their recorded decisions. Moreover, although these results cannot be deemed representative of the entire ICU nurses’ population due to qualitative nature of the design, they may portray just about accurately the types of clinical ICU nurses’ decisions, because the selection of participants was based on their rich experience and the recordings were made at three different major ICUs.
CONCLUSIONS Clinical decisions made by Greek ICU nurses appear to span a wide array of judgments and are basically consistent with the stages of the nursing process. Although nurses individualized their decisions by taking into account patient-specific data, they rarely allowed patients to participate in and guide their care. Diagnostic decisions mainly involved medical diagnoses, whereas use of nursing diagnoses was almost absent. Moreover, ICU nurses in this study rarely made decisions regarding matters of psychosocial care for either patients or families. Furthermore, decisions relating to physical assessment findings were very scarce, whereas decisions dependent to medical orders often seemed to be made without assessing the reliability of the medical order.
IMPLICATIONS FOR PRACTICE Within the course of critical care, clinical nursing decisions are ample and they affect patient outcomes significantly. Therefore, the current practice in Greek ICUs, where the clinical reasoning of nurses and nursing decisions and assessments are scarcely documented may result in underestimation of the significance and intensity of nursing work, with unfavourable implications for continuity of care, quality care and patient safety (Huffman, 2004, Johnson et al., 2006, Keenan et al., 2008), as well as staffing levels and nurse-patient ratios (Buchan, 2006). On the basis of these results, ICU nurses need to claim tasks such as physical assessment, assessment of emotional responses and psychological support and nursing diagnosis as integral parts of nursing care. Active continuing education of nursing personnel on these issues could aid this process. Furthermore, it appears that patients and their family are led to adopt passive roles and accept nursing authority. Hence, one big challenge for ICU nursing in Greece is to redefine the concepts of holistic and patient-centred care. An important consideration for future research would be to develop a quantitative tool based on the findings of qualitative studies, so that nurses’ clinical decisions can be surveyed and compared among different clinical settings and countries. Such tools would also be valuable in assessing the effectiveness of educational interventions aiming to improve the quality of critical care nursing.
ACKNOWLEDGEMENTS This research has been co-financed by the European Union (European Social Fund – ESF) and Greek national funds through the Operational Program ‘Education and Lifelong Learning’ of the National Strategic Reference Framework (NSRF) – Research Funding Program: Heracleitus II. Investing in knowledge society through the European Social Fund.
CONFLICT OF INTEREST V. K. was involved in study design, data collection, analysis process and drafting of the manuscript. E. P. was involved in study design, data analysis process and critical revision of the manuscript. C. L. was involved in study design and critical revision of the manuscript. P. S. was involved in critical revision of the manuscript. M. G. was involved in study design, supervision of the research process and critical revision of the manuscript. All authors contributed to the writing and approval of the final version of the manuscript submitted for publication.
© 2014 The Authors. Nursing in Critical Care © 2014 British Association of Critical Care Nurses
95
Exploration and classification of intensive care nurses’ clinical decisions
WHAT IS KNOWN ABOUT THE TOPIC • Clinical decision-making is an integral part of nursing care in critical care settings. • ICU nurses’ clinical decisions affect patient outcomes, quality and safety of care, as well as they involve commitment of large amounts of nursing time, effort and resources. WHAT THIS PAPER ADDS • An ICU-specific typology of critical care nurses’ clinical decisions. • A framework for evaluating and comparing nursing clinical decisions from different critical care environments. • Qualitative and quantitative data about different types of clinical decisions, providing a better understanding of the dominant scientific paradigm of critical care nursing in Greece, and nurses’ perceptions of their clinical role.
REFERENCES Aitken LM, Mardegan KJ. (2000). ‘‘Thinking-aloud’’: data collection in the natural setting. Western Journal of Nursing Research; 22: 841–853. Alexias G. (2001). The Social Construction of Medical Discourse in Intensive Care Units (in Greek). Athens: Panteion University Publication. Benner P, Hughes RG, Sutphen M. (2008). Clinical reasoning, decision making and action: thinking critically and clinically. In: Hughes RG, (ed), Patient Safety and Quality: An EvidenceBased Handbook for Nurses. Rockville: Agency for Healthcare Research and Quality. Buchan J. (2006). Evidence of nursing shortages or a shortage of evidence? Journal of Advanced Nursing; 56: 457–458. Bucknall TK. (2000). Critical care nurses’ decision making activities in the natural clinical setting. Journal of Clinical Nursing; 9: 25–36. Bucknall T. (2003). The clinical landscape of critical care: nurses’ decision making. Journal of Advanced Nursing; 43: 310–319. Bucknall T, Thomas S. (1997). Nurses’ reflections on problems associated with decision-making in critical care settings. Journal of Advanced Nursing; 25: 229–237. Colon-Emeric CS, Ammarell N, Bailey D, Corazzini K, UtleySmith Q, Lekan-Rutledge D, Anderson RA, Piven ML. (2006). Patterns of medical and nursing staff communicating in nursing homes: implications and insights from complexity science. Qualitative Health Research; 16: 173–190. Corley A, Barnett AG, Mullany D, Fraser JF. (2009). Nurse determined assessment of cardiac output. Comparing a noninvasive cardiac output device and pulmonary artery catheter: a prospective observational study. International Journal of Nursing Studies; 46: 1291–1297. Crotty M. (1985). Communication between nurses and their patients. Nurse Education Today; 5: 130–134. Curley MA. (2002). Experienced nurses + autonomy = better patient outcomes. Pediatric Critical Care Medicine; 3: 385–386. Dubose JJ, Nomoto S, Higa L, Paolim R, Teixeira PG, Inaba K, Demetriades D, Belzberg H. (2009). Nursing involvement improves compliance with tight blood glucose in the trauma ICU: a prospective observational study. International Critical Care Nursing; 25: 101–107. Erickson KA, Simon HA. (1993). Protocol Analysis. Verbal Report as Data (Revised Edition). Cambridge: MIT Press. Fagin L, Garelick A. (2004). The doctor–nurse relationship. Advances in Psychiatric Treatment; 10: 277–286. Hardy D, Smith B. (2008). Decision making in clinical practice. British Journal of Anaesthetic and Recovery Nursing; 9: 19–21.
96
Helmrich S, Yates P, Nash R, Hobman A, Poulton V, Berggren L. (2001). Factors influencing nurses’ decisions to use nonpharmacological therapies to manage patients’ pain. Australian Journal of Advanced Nursing; 19: 27–35. Hijahi M, Al-Ansari M. (2005). Protocol-driven vs. physiciandriven electrolyte replacement in adult critically ill patients. Annals of Saudi Medicine; 25: 105–110. Holloway I. (2005). Qualitative Research in Health Care. Maidenhead: Open University Press, Mc-Graw-Hill Education. Holtzdaw BJ. (1998). New trends in thermometry for the patient in the ICU. Critical Care Nursing Quarterly; 21: 12–25. Huffman M. (2004). Redefine care delivery and documentation. Nursing Management; 35: 34–38. Johnson K, Hallsey D, Meredith RL, Warden EJ. (2006). A nursedriven system for improving patient quality outcomes. Journal of Nursing Care Quality; 21: 168–175. Keenan GM, Yakel E, Tschannen D, Mandeville M. (2008). Documentation and the Nurse Care Planning Process. In: Hughes RG, (ed), Patient Safety and Quality: An EvidenceBased Handbook for Nurses. Rockville: Agency for Healthcare Research and Quality. Kitson A. (1999). Evidence-Based Practice: A Primer for Health Care Professionals. London: Churchill Livingstone. Klein G. (1989). Advances in Man-Machine-Systems Research. Greenwich: JAI. Kollef MH, Shapiro SD, Silver P, John RE, Prentice D, Sauer S, Ahrens TS, Shannon W, Baker-Clinkscale D. (1997). A randomized trial of protocol-directed versus physiciandirected weaning from mechanical ventilation. Critical Care Medicine; 25: 567–574. Kushniruk AW. (2001). Analysis of complex decision making processes in health care: cognitive approaches to health informatics. Journal of Biomedical Informatics; 34: 365–376. Lundgren-Laine H, Salantera S. (2010). Think-aloud technique and protocol analysis in clinical decision making research. Qualitative Health Research; 20: 565–575. McCaughan D, Thompson C, Cullum N, Sheldon T, Raynor P. (2005). Nurse practitioner and practice nurses’ use of research information in clinical decision making: findings from an exploratory study. Family Practice; 22: 490–497. McKnight M. (2006). The information seeking of on-duty critical care nurses: evidence from participant observation and incontext interviews. Journal of the Medical Library Association; 94: 145–151. Nelson G, Kinjo K, Meier DE, Ahmad K, Morrison RS. (2005). When critical illness becomes chronic: informational needs of patients and families. Journal of Critical Care; 20: 79–89.
© 2014 The Authors. Nursing in Critical Care © 2014 British Association of Critical Care Nurses
Exploration and classification of intensive care nurses’ clinical decisions
Nelson JE, Mulkerin CM, Adams LL, Pronovost PJ. (2006). Improving comfort and communication in the ICU: a practical new tool for palliative care performance measurement and feedback. Quality & Safety in Health Care; 15: 264–271. Noone J. (2002). Concept analysis of decision making. Nursing Forum; 37: 21–32. Orlando IJ. (1972). The Discipline and Teaching of Nursing Process (an Evaluative Study). New York: G.P. Putman’s Sons. Papathanassoglou ED, Karanikola MN, Kalafati M, Giannakopoulou M, Lemonidou C, Albarran JW. (2012). Professional autonomy, collaboration with physicians, and moral distress among European intensive care nurses. American Journal of Critical Care; 21: e41–e52. Papathanassoglou ED, Tseroni M, Karydaki A, Vazaiou G, Kassikou J, Lavdaniti M. (2005). Practice and clinical decision making autonomy among Hellenic critical care nurses. Journal of Nursing Management; 13: 154–164. Parahoo K. (2000). Barriers to and facilitators of, research utilization among nurses in Northern Ireland. Journal of Advanced Nursing; 31: 89–98. Patiraki-Kourbani E. (2003). Greece (Hellenic Republic). In: Erickson D’Avanzo C, Geissler EM, (eds), Cultural Health Assessment. St Louis: Mosby. Ritchie J, Spencer L. (1994). Qualitative data analysis for applied policy research. In: Bryman A, Burgess R, (eds), Analyzing Qualitative Data. London: Routlege. Rose L, Nelson S, Johnston L, Presneill JJ. (2008). Workforce profile, organization structure and role responsibility for ventilation and weaning practices in Australia and New Zealand intensive care units. Journal of Clinical Nursing; 17: 1035–1043.
Rubin HJ, Rubin IS. (1995). Qualitative Interviewing: The Art of Hearing Data. Thousand Oaks: Sage Publications. Rycroft-Malone J, Seers K, Titchen A, Harvey G, Kitson A, McCormack B. (2004). What counts as evidence in evidence-based practice? Journal of Advanced Nursing; 47: 81–90. Schilling J. (2006). On the pragmatics of qualitative assessment: designing the process for content analysis. European Journal of Psychological Assessment; 22: 28–37. Thomas PJ, Paratz JD, Stanton WR, Deans R, Lipman J. (2006). Positioning practices for ventilated intensive care patients: current practice indications and contraindications. Australian Critical Care; 19: 122–126 128, 130–132. Thompson C, Cullum N, McCaughan D, Sheldon T, Raynor P. (2004). Nurses, information use, and clinical decision makingthe real world potential for evidence-based decisions in nursing. Evidence-Based Nursing; 7: 68–72. Thompson C, McCaughan D, Cullum N, Sheldon TA, Mulhall A, Thompson DR. (2001). Research information in nurses’ clinical decision-making: what is useful? Journal of Advanced Nursing; 36: 376–388. Titler MG, Cohen MZ, Craft MJ. (1991). Impact of adult critical care hospitalization: perceptions of patients, spouses, children, and nurses. Heart & Lung; 20: 174–182. Wilkinson JM. (1996). Nursing Process: A Critical Thinking Approach. Menlo Park: Addison–Wesley. Zhang Y, Wildemuth BM. (2009). Qualitative analysis of content. In: Wildemuth B, (ed), Applications of Social Research Methods to Questions Information and Library Science. Westport: Libraries Unlimited.
© 2014 The Authors. Nursing in Critical Care © 2014 British Association of Critical Care Nurses
97
