Nursing Diagnoses (NANDA-I) in Hematology Oncology: A Delphi-Studyijnt_ Highly complex nursing care is necessary for the

Transcription

1 International Journal of Nursing Terminologies and Classifications Nursing Diagnoses (NANDA-I) in Hematology Oncology: A Delphi-Studyijnt_ Herma T. Speksnijder, MSc, RN, Arno P. Mank, RN, and Theo van Achterberg, PhD, RN PURPOSE: To identify NANDA-I diagnoses that are most relevant to hematology oncology nursing in Europe. METHODS: In a two-round, electronic, quantitative Delphi study, 28 experts from nine European countries assessed the relevance of NANDA-I diagnoses and health problems. FINDINGS: This study identified 64 relevant diagnoses and three health problems. All experts listed 11 diagnoses: imbalanced nutrition: less than body requirements, diarrhea, fatigue, risk for bleeding, risk for infection, impaired oral mucous membrane, risk for impaired skin integrity, impaired skin integrity, hyperthermia, nausea, acute pain, and the health problem pruritis. CONCLUSIONS AND IMPLICATIONS FOR PRACTICE: The NANDA-I classification describes, in almost all disease- and treatment-related problems, nursing diagnoses as relevant to the adult patient with hematological malignancy. These diagnoses are therefore recommended. Search terms: Nursing diagnoses, hematology-oncology, classification Dutch Version Abstract DOEL: Onderzoeken welke NANDA-I diagnoses het meest relevant zijn voor de hemato-oncologische verpleegkundige zorgverlening in Europa. METHODEN: In een elektronische, kwantitatieve Delphistudie (twee rondes) beoordeelden 28 experts uit negen Europese landen de relevantie van NANDA-I diagnoses en gezondheidsproblemen. RESULTATEN: Er werden 64 relevante diagnoses en drie gezondheidsproblemen geïdentificeerd. Alle experts beoordeelden 11 diagnoses relevant: Voedingstekort, Diarree, Oververmoeidheid, Bloedinggevaar, Infectiegevaar, Beschadigd mondslijmvlies, Dreigend huiddefect, Huiddefect, Hyperthermie, Misselijkheid, Acute pijn en het gezondheidsprobleem Jeuk. CONCLUSIES EN AANBEVELINGEN: De NANDA-I classificatie beschrijft in bijna alle ziekte- en behandelingsgerelateerde situaties verpleegkundige diagnoses voor de volwassen patiënt met een hematologische maligniteit. NANDA-I diagnoses worden om deze reden aanbevolen. Search terms: Verpleegkundige diagnoses, hematologische oncologie, classificatie 2011, The Authors International Journal of Nursing Terminologies and Classifications 2011, NANDA International doi: /j X x Herma T. Speksnijder MSc, RN, is a Nurse and Nurse Scientist, Erasmus MC, Rotterdam, the Netherlands, Arno P. Mank, RN, is a Nurse Researcher, Academic Medical Centre, Amsterdam, the Netherlands, and Theo van Achterberg, PhD, RN, is a Professor of Nursing Science, Radboud University Nijmegen Medical Centre, Scientific Institute for Quality of Care, the Netherlands. Highly complex nursing care is necessary for the disease- and treatment-related health problems in patients with a hematological malignancy. Hematological malignancies account for approximately 7% of all new malignant tumors and cancer deaths in Europe (Löwenberg, 2008). Diseases in the field of hematology are diverse but are generally associated with specific disease- and treatment-related symptoms. Hematological disorders can be benign but are more often malignant, as with myeloid or lymphocytic leukemia, lymphoma, and multiple myeloma. Treatment is typically characterized by high doses of chemotherapy and is sometimes finalized with autologous or allogeneic stem cell transplantation, contributing to symptoms such as fatigue, fever, risk for infection, a tendency toward increased bleeding, or toward thrombosis (Kluin-Nelemans, de Brouwer, & Roodbol, 2006). In the last several decades, targeted therapy has been introduced in combination with traditional therapies in these patients. Nurses play a key role in ensuring the proper and safe administration of these therapies and are often the first to identify the signs of side effects. Patient education about anticipated side effects and close monitoring of patients can lead to symptom management interventions that are essential to patient comfort and safety (Colson, Doss, Swift, Tariman, & Thomas, 2004). As part of the multidisciplinary hematology oncology team, nurses are ideally positioned to play a key role in the identification and management of hematologic toxicities and act as patient advocates of supportive care. Nurses possess the expert knowledge of the day-to-day responses of their patients. Consistent and frequent clinical assessment International Journal of Nursing Terminologies and Classifications Volume 22, No. 2, April-June,

2 Nursing Diagnoses (NANDA-I) in Hematology Oncology: A Delphi Study helps to identify signs and symptoms early (Montoya, 2007). To identify, collect data, communicate, and organize nursing care, nursing diagnoses play an important role (Pesut & Herman, 1999). In a nursing language, like nursing diagnoses, it is defined as what belongs to the field of nursing, what kind of care is given, why it is given, and what results are expected. NANDA-I defined the first classification of nursing diagnoses, which is the most frequently used international nursing diagnoses classification (Müller-Staub, 2004). A nursing diagnosis contains a definition, defining characteristics, and related or risk factors. NANDA-I defines a nursing diagnosis as a clinical judgment about individual, family, or community responses to actual or potential health problems/life processes. A nursing diagnosis provides the basis for selection of nursing interventions to achieve outcomes for which the nurse is accountable (NANDA International, 2009, p. 419). NANDA-I taxonomy II divided nursing diagnoses into 13 domains. Roget, 1980 (as cited in NANDA International, 2009) defines a domain as a sphere of activity, study, or interest. Müller-Staub, Lavin, Needham, and van Achterberg (2006) indicate that by using standard terms from the NANDA-I classification, nursing care will be well documented, and data collection will be facilitated. It also makes the quality and cost-effectiveness of nursing care possible. Hoyt (1997) underlines the importance of a nursing language: If you can t name it, you can t control it, finance it, teach it, or put it into policy. To identify nursing diagnoses in hematology oncology nursing, a systematic literature review was conducted. International databases (PubMed, The Cochrane Library, Embase, EBSCO, Cummulative Index to Nursing and Allied Health Literature, and PsycINFO) were searched ( ) with the following Medical Subject Headings (MeSH) terms: nursing diagnosis, hematology, nursing diagnosis/ classification, and nursing/classification. The terms NANDA taxonomy, functional health pattern, and ICIDH classification were also used. Only four studies addressing nursing diagnoses in patients with a hematological malignancy were found. De Souza and de Gorini (2006) identified 32 nursing diagnoses in six adult patients with acute myeloid leukemia by interviewing patients and consulting medical dossiers. In a study by Courtens and Huijer Abu-Saad (1998), seven nurses were interviewed and 15 nursing records of leukemia patients (adults) were analyzed. De Jesus and de Carvalho (1997) identified nursing diagnoses in 14 patients (14 84 years old) with a hematological malignancy like leukemia, multiple myeloma, or lymphoma by interviewing and observing patients, and carrying out physical examinations on them. Finally, Magalhăes, Matzenbacher, and Pacheco (2005) conducted a case study to analyze nursing diagnoses occurring in a patient with a lymphoma undergoing allogeneic stem cell transplantation. Nursing diagnoses identified in the literature regarding hematology oncology nursing care mainly focused on patients with leukemia. These four studies identified the following nursing diagnoses: effective control of the therapeutic regimen: individual, ineffective management of therapeutic regimen: family, impaired adjustment, imbalanced nutrition: less than body requirements, excess fluid volume, risk for deficient fluid volume, constipation, diarrhea, disturbed sleep pattern, sleep disturbance, altered tissue perfusion, fatigue, activity intolerance, risk for bleeding, self-care deficit related to hygienic care, knowledge deficit family, knowledge deficit, uncertainty, disturbed sensory perception, decisional conflict, denial, disturbed body image, change in family role, change in parenting, disturbed family processes, anxiety about hospital, impaired skin integrity, hyperthermia, risk for imbalanced body temperature, risk for infection, altered protection, impaired oral mucous membrane, altered oral mucous membrane, risk for injury, skin problems, nausea, anticipatory nausea, acute pain, and pain (Courtens & Huijer Abu-Saad, 1998; De Jesus & de Carvalho, 1997; De 78 International Journal of Nursing Terminologies and Classifications Volume 22, No. 2, April-June, 2011

3 Souza & de Gorini, 2006; Magalhăes et al., 2005). Two studies (Courtens & Huijer Abu-Saad, 1998; De Jesus & de Carvalho, 1997) used the NANDA-I classification and one (De Souza & de Gorini, 2006) used Maslow s hierarchy of needs. In the latter study, patients needs were categorized as: self-actualization, esteem needs, belonging needs, safety needs, and physiological needs. Magalhăes et al. (2005) did not describe a classification at all. The four studies together did not give an up-to-date overview of relevant nursing diagnoses in hematology oncology nursing. Furthermore, the use of classifications varied, and some studies were rather old and possibly outdated. The researchers therefore decided to add studies addressing nursing diagnoses in oncology nursing because oncology patients often share the same disease- and treatment-related problems. An overview of relevant nursing diagnoses from these studies might be helpful in identifying nursing diagnoses in hematology oncology nursing. The added terms in the search were oncology and cancer. This resulted in 10 articles addressing nursing diagnoses in oncology patients (Antall, 1989; Chang, Vredevoe, & Hirsch, 1995; Dougherty, 2007; Lee, 2006; Lopes, Macedo, & de Moraes Lopes, 1997; MacAvoy & Moritz, 1992; Miaskowski & Garofallou, 1986; Ogasawara et al., 2005; Woodtli & van Ort, 1991, 1993). To get a complete overview of the most relevant NANDA-I diagnoses to hematology oncology nursing, a Delphi study was conducted. The question under study was: Which nursing diagnoses, formulated by the NANDA-I classification , and health problems (not named in nursing diagnoses) are assessed as relevant by experts in hematology oncology for the adult patient with a hematological malignancy in Europe? More up-to-date data regarding NANDA-I diagnoses and added health problems of relevance to hematology oncology nursing are needed so that hematology oncology nurses have current, evidence-based data on which to base care. Methods A quantitative, descriptive, two-round Delphi study was used to identify NANDA-I nursing diagnoses and health problems relevant to hematology oncology nursing. The Delphi technique is a group facilitation technique that seeks to obtain consensus on the opinion of experts through several series of structured questionnaires/rounds (Hasson, Keeney, & McKenna, 2000). One of the basic principles underpinning the Delphi technique is to have as many rounds as are required to achieve consensus or until the law of diminishing returns occurs. The classic Delphi study suggests four rounds (Young & Hogben, 1978), but more recent evidence suggests that either two or three rounds are preferred (Green, Jones, Hughes, & Willimas, 1999). The number of rounds depends on the topic of interest. In this study the number of rounds was determined by saturation of data. When consensus was reached and no new main topics were added, the researchers decided that the point of data saturation had been reached. Sample A network sample, a sampling of participants based on referrals from others already in the sample (Polit & Beck, 2008), was drawn from the accessible population who were members of the European Blood and Marrow Transplantation Nurses Group (EBMT-NG), recruited through advertisement. When experts agreed to participate in the study, they were informed about the goal of the study, their contribution, and how much time it would take. Studies employing the Delphi method use experts with knowledge of the topic of interest (Hasson et al., 2000). There are no universally agreed criteria for the selection of experts, and no guidance on the minimum or maximum number of experts on a panel; rather, the method appears to be related to common sense and practical logistics (Keeney, Hasson, & McKenna, 2006). The researchers deemed a number of experts would be needed to International Journal of Nursing Terminologies and Classifications Volume 22, No. 2, April-June,

4 Nursing Diagnoses (NANDA-I) in Hematology Oncology: A Delphi Study assure that enough experts from European countries would be included and to assure sufficient numbers should attrition occur. Eligibility criteria for expert panelists included being active in adult hematology oncology nursing for a minimum of 3 years, able to understand Dutch or English, having Internet access, and, preferably, being familiar with the knowledge of the components of nursing diagnoses and the usage of diagnoses in nursing practice. Consensus, Iteration, and Controlled Feedback In the Delphi technique, multiple iterations are used to achieve consensus, without the necessity of face-toface discussion (Polit & Beck, 2008). There are no recognized guidelines on an appropriate level of consensus. Keeney et al. (2006) stated that 75% appears to be the minimal level, but there is no obvious scientific rationale for this. In the present study, absolute consensus (100%) was considered when all experts agreed on the relevance of the nursing diagnosis. A consensus level of 80 99% was considered as high consensus, 65% 79% was considered as moderate consensus, and levels below 65% were considered as no consensus. The Delphi technique is an iterative multistage process designed to combine expert opinion into group consensus. The status of the group s collective opinion was fed back after both rounds to help identify items that experts could have missed or thought unimportant. That is why there was the opportunity to change opinions in the second round. This enabled experts to see their response in relation to that of the group. For ethical considerations, quasi-anonymity was used. Experts were known to the researcher, but their judgments and opinions remained strictly anonymous to others (Hasson et al., 2000). Instrument, Data Collection, and Data Analysis The Delphi study was performed using electronic questionnaires (Dutch and English) on the Internet. Experts received a log-in code to access the questionnaires. Data were collected from January to March To achieve a firm basis for the first Delphi round, a systematic literature review was conducted. Nursing diagnoses identified in these studies were compared with nursing diagnoses in the NANDA-I classification The instrument for the first Delphi round was designed using 62 nursing diagnoses identified from 14 studies (Antall, 1989; Chang et al., 1995; Courtens & Huijer Abu-Saad, 1998; De Jesus & de Carvalho, 1997; De Souza & de Gorini, 2006; Dougherty, 2007; Lee, 2006; Lopes, Macedo, & de Moraes Lopes, 1997; MacAvoy & Moritz, 1992; Magalhăes et al., 2005; Miaskowski & Garofallou, 1986; Ogasawara et al., 2005; Woodtli & van Ort, 1991, 1993). Experts were asked to list the relevance of diagnoses named. Diagnoses were assessed quantitatively with a nominal variable ( yes = relevant, no = not relevant). For every item under study, outcomes were combined using frequencies in the Statistical Package for the Social Sciences 15.0 (SPSS, Inc., Chicago, IL, USA). Frequencies were expressed in four groups of percentages: absolute consensus (100%), high consensus (80 99%), moderate consensus (65 79%), and no consensus (< 65%) diagnoses. Absolute consensus and high consensus diagnoses were directly put on the list with nursing diagnoses relevant to hematology oncology nursing. Moderate consensus diagnoses were studied again in the second round and no consensus diagnoses were excluded. In the first round, there was also the opportunity to add nursing diagnoses from the NANDA-I Nursing diagnoses: Definitions and classification or health problems (not named in NANDA-I). The second round consisted of diagnoses with a moderate consensus level from the first round to give experts the opportunity to reconsider or modify their responses in light of the overall response from the first round. Apart from these moderate consensus-level diagnoses from the first round, nursing diagnoses added by five or more experts in the first round, and health problems named twice or 80 International Journal of Nursing Terminologies and Classifications Volume 22, No. 2, April-June, 2011

5 Figure 1. Composition Diagram of the Delphi Rounds 1 st Delphi-round n=62 Result: 5 no consensus, 15 moderate consensus, 31 high consensus, 11 absolute consensus Added in 1 st Delphi-round 20 nursing diagnoses (NANDA-I) 3 health problems 2 nd Delphi-round n=38 (15 moderate consensus 1 st round + 20 added nursing diagnoses + 3 health problems) Result: 13 no consensus, 12 moderate consensus, 12 high consensus, 1 absolute consensus" Results of both Delphi rounds: 12 moderate consensus (10 nursing diagnoses, 2 health problems) 55 high consensus" (54 nursing diagnoses, 1 health problem) more were studied in the second round. The Delphi rounds are shown schematically in Figure 1. Data from both rounds were analyzed by two independent scientific researchers. Health problems that were named by two or more experts in the first round were given an expert-based definition (Table 1) and were introduced in the next Delphi round. Results Experts were invited for this study by advertisement. Forty-one experts responded to the advertisement. Thirty experts met the eligibility criteria and were included; 14 were EMBT-NG Table 1. Label Pruritis Excessive sweating Risk for or actual financial instability Definition of Health Problems Added by Experts Definition A sensation in the skin that induces the desire to scratch (itch) The condition of excreting excessive perspiration (> ml/day) Financial instability caused by the inability or reduced ability to be active as an employee during or after illness International Journal of Nursing Terminologies and Classifications Volume 22, No. 2, April-June,

6 Nursing Diagnoses (NANDA-I) in Hematology Oncology: A Delphi Study Table 2. Baseline Characteristics Experts Relevance of NANDA-I Diagnoses Characteristics Gender Male 4 Female 24 Working experience in hematology oncology (years active) 3 5 years years 5 > 10 years 19 Work position Nurse 13 Nurse practitioner 3 Clinical nurse specialist 4 Nurse manager 4 Assistant nurse manager 3 Data manager 1 Setting Hospital 26 Outpatient clinic 2 Familiar with nursing diagnoses Never heard of nursing diagnoses 1 Familiar with nursing diagnoses 15 Active use of nursing diagnoses at work 12 Country Belgium 2 Denmark 5 Germany 2 Iceland 1 Italy 2 Spain 3 Switzerland 2 The Netherlands 9 United Kingdom 2 Number of experts (n = 28) members and 16 nurses were proposed by EBMT-NG members. Finally, 28 experts (93%) responded for both Delphi rounds. Demographic data are presented in Table 2. In this sample, 12 experts worked with nursing diagnoses on their hematology oncology ward. Only one expert was not familiar with nursing diagnoses. The expert panel considered 64 NANDA-I diagnoses of the NANDA-I classification and three health problems relevant to adult patients with a hematological malignancy. Table 3 summarizes the study results. Eleven nursing diagnoses and one health problem were assessed as relevant by all experts (the category absolute consensus in Table 3). An overview of all nursing diagnoses under study can be found in the Appendix. First Delphi round. A total of 42 of the proposed 62 diagnoses from the scientific literature (hematology oncology and oncology) attained high consensus (80 99%) in the first round. Absolute consensus (100%) was reached for 11 diagnoses: imbalanced nutrition: less than body requirements, diarrhea, fatigue, risk for bleeding, risk for infection, impaired oral mucous membrane, risk for impaired skin integrity, impaired skin integrity, hyperthermia, nausea, and acute pain. Fifteen diagnoses attained moderate consensus and were therefore examined again in the second Delphi round. The level of no consensus was considered for five diagnoses. In total, 19 diagnoses were added by five or more experts, and three health problems were added by at least two experts in the first round: pruritis, excessive sweating, and risk for or actual financial problems. Definitions of these health problems were formulated by two independent researchers (Table 1). Second Delphi round. The second round started with 15 moderate consensus diagnoses from the first Delphi round. Experts got the opportunity to reconsider or modify their responses with the overall response of the first Delphi round as background knowledge (see Appendix). In the second round, 20 nursing diagnoses and three health problems added by experts in the first round were also studied. From the 15 moderate consensus diagnoses (first round), 11 diagnoses were found to be less relevant than in the 82 International Journal of Nursing Terminologies and Classifications Volume 22, No. 2, April-June, 2011

7 Table 3. Relevant NANDA-I Nursing Diagnoses and Health Problems in Hematology Oncology Nursing after Two Delphi Rounds NANDA-I domains Code Absolute consensus (100%) Code High consensus (80 99%) Code Moderate consensus (65% 79%) Domain 1: Health promotion Domain 2: Nutrition Domain 3: Elimination and exchange Domain 4: Activity/rest Domain 5: Perception/ cognition Imbalanced nutrition: less than body requirements Risk for electrolyte imbalance Risk for deficient fluid volume Deficient fluid volume Risk for imbalanced fluid volume Excess fluid volume Risk for unstable blood glucose level Impaired swallowing Readiness for enhanced nutrition Readiness for enhanced self-health management Risk for impaired liver function Diarrhea Risk for constipation Impaired urinary elimination Constipation Impaired gas exchange Risk for dysfunctional gastrointestinal motility Dysfunctional gastrointestinal motility Bowel incontinence Fatigue Activity intolerance Dressing self-care deficit Risk for bleeding Disturbed sleep pattern Impaired physical mobility Risk for activity intolerance Risk for shock Bathing self-care deficit Feeding self-care deficit Impaired walking Toileting self-care deficit Deficient knowledge Acute confusion Risk for acute confusion Disturbed sensory perception (Specify: visual, auditory, kinesthetic, gustatory, tactile, or olfactory) International Journal of Nursing Terminologies and Classifications Volume 22, No. 2, April-June,

8 Nursing Diagnoses (NANDA-I) in Hematology Oncology: A Delphi Study Table 3. Continued NANDA-I domains Code Absolute consensus (100%) Code High consensus (80 99%) Code Moderate consensus (65% 79%) Domain 6: Self-perception Domain 7: Role relationships Hopelessness Risk for loneliness Disturbed body image Situational low self-esteem Powerlessness Interrupted family processes Ineffective role performance Impaired social interaction Domain 8: Sexuality Sexual dysfunction Domain 9: Coping/ stress tolerance Domain 10: Life principles Domain 11: Safety/ protection Domain 12: Comfort Domain 13: Growth/ development Risk for infection Impaired oral mucous Membrane Risk for impaired skin integrity Hyperthermia Impaired skin integrity Anxiety Ineffective coping Death anxiety Grieving Fear Disabled family coping Impaired individual resilience Noncompliance Risk for vascular trauma Impaired tissue integrity Risk for imbalanced body temperature Nausea Chronic pain Acute pain Social isolation Health problems Pruritis Excessive sweating Risk for or actual financial problems 84 International Journal of Nursing Terminologies and Classifications Volume 22, No. 2, April-June, 2011

9 first round. High consensus levels were given to 12 diagnoses and two health problems of the studied items in the second round. Pruritis was the only health problem with an absolute consensus level. A moderate consensus level was given to 14 diagnoses/health problems and 10 reached no consensus level. As 12 diagnoses/health problems with a moderate consensus level were inventoried after the second round and no new main topics were added, the researchers decided that saturation of data was reached. NANDA-I Domains NANDA-I diagnoses are classified into 13 domains. Many relevant diagnoses are found in domain 2 (nutrition), domain 3 (elimination and exchange), domain 4 (activity/rest), domain 9 (coping/stress tolerance), and domain 11 (safety/protection). In domain 11, many diagnoses with an absolute relevance are found. Diagnoses from domain 13 (growth/development) were not identified as relevant to hematology oncology nursing in adults. Discussion This is the first broad study identifying NANDA-I diagnoses relevant to hematology oncology nursing care in Europe. The purpose of this study was to answer the following question: Which nursing diagnoses formulated by the NANDA-I classification and health problems (not named in nursing diagnoses) are assessed as relevant by experts in hematology oncology for the adult patient with a hematological malignancy in Europe? The results suggest that 64 nursing diagnoses of the NANDA-I classification and three further health problems are relevant to hematology oncology nursing. Comparing Results Frequently occurring nursing diagnoses in hematology oncology were identified in a systematic literature review. Compared with the literature the following nursing diagnoses were not found in the present study: effective control of the therapeutic regimen: individual, ineffective management of therapeutic regimen: family, impaired adjustment, risk for injury, skin problems, change in parenting, and anticipatory nausea (Courtens & Huijer Abu-Saad, 1998; De Jesus & de Carvalho, 1997; De Souza & de Gorini, 2006; Magalhăes et al., 2005). Some of these diagnoses are not formulated by NANDA-I diagnoses and, for that reason, experts could have missed these diagnoses. On the other hand, experts did not list these diagnoses as important health problem in patients with a hematological malignancy and, therefore, it can be concluded that these nursing diagnoses are not relevant to hematology oncology nursing. Limitations and Strengths of the Study The experts in the present study were not selected randomly, so representativeness cannot be assured. The composition of the sample was only partially controlled by the researchers because of self-selection in network sampling. The composition of the sample was only partially controlled by the researchers. The Delphi panel was formed by experts from nine West European countries. However, the countries were not equally represented; Denmark and the Netherlands provided 50% of the experts and several European countries were not represented. This imbalance will possibly make the process of generalizing the results in applying to all nursing in hematology oncology in Europe rather imbalanced (target population). Besides this, 28 experts might not be enough to represent the complete range of European hematology oncology nursing. Several nursing functions were present in the study, and many of the experts (n = 19) have had working experience in hematology oncology nursing for 10 years or more, so the results of this study are based on proven experience. Twenty-seven out of 28 experts have knowledge of the components of International Journal of Nursing Terminologies and Classifications Volume 22, No. 2, April-June,

10 Nursing Diagnoses (NANDA-I) in Hematology Oncology: A Delphi Study a nursing diagnosis and the use of diagnoses in nursing practice. Twelve out of 28 experts work actively with nursing diagnoses on their ward. This was very important for this study, as we aimed for well-advised, practice-based judgments on the relevance of nursing diagnoses and the health problems studied. It was recognized as a limitation to the study results that one expert was not familiar with nursing diagnoses at all. One of the inclusion criteria was that experts had to understand English. Clear interpretations of labels and definitions of the nursing diagnoses were very important in making relevant judgments. However, many experts live in countries where English is not the first language. This could possibly hinder the interpretation and affect the reliability of the results. Specified inclusion criteria for experts and a predefined method of analysis strengthened the Delphi approach (Hasson et al., 2000). Using an electronic survey eased completion of the Delphi exercise. It was time- and cost-effective, and provided easy access for the international experts. Data were analyzed by two independent researchers, which increased the level of reliability (Keeney et al., 2006). Operating within tight consensus levels, the researchers allowed clear identification of the most relevant NANDA-I diagnoses. The higher the percentage of concensus of diagnoses/health problems by experts, the more relevant a NANDA-I diagnosis is to hematology oncology nursing. The no consensus level does not necessarily mean that a diagnosis does not appear at all, but that it is probably not very specific to hematology oncology nursing. The relevance of NANDA-I diagnoses to labels and definitions to limit the time investment of experts in judging relevance of diagnoses was examined. It was seen that judging relevance only by label and definition did not give enough information to experts to make a well-advised choice. Adding underlying characteristics and related or risk factors might have given more detailed information about the diagnoses studied. Because of the iterative, multistage process of the Delphi technique, the status of the group s collective opinion was fed back after both questionnaires. Reliability was achieved by giving experts the opportunity to see their responses in relation to the group response and to give them the opportunity to change their views to bring experts toward group consensus (Hasson et al., 2000; Keeney et al., 2006). Diagnoses from the first Delphi round with a moderate consensus were examined again. Some reached a high consensus level in the second round, others a level of no consensus, thanks to group responses. The high response rate of 93% also increased the validity of results (Hasson et al., 2000). Conclusion This study has identified 64 relevant nursing diagnoses, formulated by the NANDA-I classification This classification describes nursing diagnoses as being relevant to adult patients with a hematological malignancy in almost all disease- and treatment-related problems (98%), and is therefore very relevant to hematology oncology nursing. To get a complete overview of relevant nursing diagnoses in hematology oncology nursing, the health problems described here should be studied and translated into NANDA-I nursing diagnoses. The use of identified relevant diagnoses and health problems will give hematology oncology nursing evidence-based care. It will enable nurses to describe their methods of care to adult patients with hematological malignancy in a common nursing language. By using NANDA-I diagnoses, nurses can name, control, finance, teach, and research hematology oncology nursing. Relevant diagnoses can also be used in the development of educational programs and materials for students in hematology oncology nursing. Experts in hematology oncology nursing have judged the relevance of nursing diagnoses. Future research must examine the usage of these findings in adults with a hematological malignancy. 86 International Journal of Nursing Terminologies and Classifications Volume 22, No. 2, April-June, 2011

11 Recommendations and Implications for Practice The authors recommend nurses to use these identified relevant nursing diagnoses from the NANDA-I classification and relevant health problems to describing nursing care in hematology oncology. The relevance of these diagnoses should be studied again, with the addition of signs and symptoms of a diagnosis in a larger sample, with a more equal representation of countries. Besides that, it is very important that all experts have based knowledge of nursing diagnoses. In addition, signs and symptoms of relevant diagnoses can be examined specific to the illness- and treatment-related problems of hematology-oncology patients. The relevance of these diagnoses can be investigated in pediatric hematology oncology nursing to give a complete overview of relevant nursing diagnoses in hematology oncology nursing. The identified health problems, pruritis, excessive sweating, and risk for or actual financial instability, are identified as relevant to hematology oncology nursing. In order to develop nursing diagnoses for these health problems, research is necessary to describe characteristics and related or risk factors. To describe the whole nursing process (diagnosing, intervening, and measuring outcomes) nursing interventions (nursing intervention classification) and nursing outcomes (nursing outcomes classification) can be linked to these diagnoses. This allows nurses to build a body of knowledge based on actual patient care and science. The information can be used to assist nurses in making better clinical decisions. The methodology used in this study will be of interest to researchers who are interested in validating nursing diagnoses in other patient populations. Electronic Delphi rounds on the Internet provide the possibility of performing studies internationally by using the expertise of international experts. Author contact: h.speksnijder@erasmusmc.nl, with a copy to the Editor: jane.flanagan@bc.edu References Antall, G. F. (1989). Nursing diagnoses identified in the oncology patient at discharge. In R. M. Carroll-Johnson (Ed.), Classification of nursing diagnoses: Proceedings of the eighth conference (pp ). Philadelphia, PA: Lippincott. Chang, B. L., Vredevoe, D., & Hirsch, M. (1995). Allergy as a risk factor for nursing care problems in the elderly cancer patient. Cancer Nursing, 18(2), Retrieved from Colson, K., Doss, D. S., Swift, R., Tariman, J., & Thomas, T. E. (2004). Bortezomib, a newly approved proteasome inhibitor for the treatment of multiple myeloma: Nursing implications. Clinical Journal of Oncology Nursing, 8(5), doi: / 04.CJON Courtens, A. M., & Huijer Abu-Saad, H. (1998). Nursing diagnoses in patients with leukemia. Nursing Diagnosis, 9(2), Retrieved from De Jesus, C. A. C., & de Carvalho, E. C. (1997). Diagnósticos de enfermagem em clientes com alteraçoes hematológicas: Uso da taxonomia i da nanda. Revista Latino-Americana de Enfermagem, 5(4), Retrieved from script=sci_serial&pid= &ing=pt&nrm=iso&rep= De Souza, L. M., & de Gorini, M. I. P. C. (2006). Diagnósticos de enfermagem em adultos com leucemia mielóide aguda. Revista Gaúcha de Enfermagem, 27(3), Retrieved from =en&nrm=iso Dougherty, L. (2007). Using nursing diagnoses in prevention and management of chemotherapy-induced alopecia in the cancer patient. International Journal of Nursing Terminologies and Classifications, 18(3 4), doi: /j X x Green, B., Jones, M., Hughes, D., & Willimas, A. (1999). Applying the Delphi technique in a study of gps information requirements. Health & Social Care in the Community, 7(3), doi: / j x Hasson, F., Keeney, S., & McKenna, H. (2000). Research guidelines for the Delphi survey technique. Journal of Advanced Nursing, 32(4), doi: /j t x Hoyt, S. K. (1997). Validating nursing with NANDA, NIC, and NOC. Journal of Emergency Nursing, 23, Retrieved from Keeney, S., Hasson, F., & McKenna, H. P. (2006). Consulting the oracle: Ten lessons from using the Delphi technique in nursing research. Journal of Advanced Practice, 53(2), doi: / j x Kluin-Nelemans, J. C., de Brouwer, M. F., & Roodbol, P. F. (2006). Hematologie. Houten: Bohn Stafleu van Loghum. Lee, B. (2006). Developing a cancer nursing information system: Determining core nursing diagnoses for the six most common cancers in Korea. Studies in Health Technology and Informatics, 122, doi: /j.ijnurstu International Journal of Nursing Terminologies and Classifications Volume 22, No. 2, April-June,

12 Nursing Diagnoses (NANDA-I) in Hematology Oncology: A Delphi Study Lopes, R. A. M., Macedo, D. D., & de Moraes Lopes, M. H. B. (1997). Diagnósticos de enfermagem mais frequentes em uma unidade de internaçăo de oncologia. Revista Latino-Americana de Enfermagem, 5(4), Retrieved from scielo.php?script=sci_serial&pid= &ing=pt&nrm=iso &rep= Löwenberg, B. (2008). Handboek hematologie. Tijdstroom: Utrecht. MacAvoy, S., & Moritz, D. (1992). Nursing diagnoses in an oncology population. Cancer Nursing, 15(4), Retrieved from Magalhăes, A. M. M., Matzenbacher, B. C. M., & Pacheco, C. R. M. (2005). Diagnósticos de enfermagem de paciente submetido à transplante de medula óssea alogênico: Estudo de caso. Revista Gaúcha de Enfermagem, 26(1), Retrieved from =en&nrm=iso Miaskowski, C. A., & Garofallou, G. (1986). A study of nursing diagnoses in an oncologic patient population. In M. Hurley (Ed.), Classification of nursing diagnoses: Proceedings of the sixth conference (pp ). St. Louis, MO: Mosby. Montoya, L. (2007). Managing hematologic toxicities in the oncology patient. Journal of Infusion Nursing, 30(3), doi: / 01.NAN c3 Müller-Staub, M. (2004). Pflegeklassifikationen im vergleich. RIN- TERNET, 6(04), Retrieved from info/detail.asp?id=465 Müller-Staub, M., Lavin, M. A., Needham, I., & van Achterberg, T. (2006). Nursing diagnoses, interventions and outcomes application and impact on nursing practice: Systematic review. Journal of Advanced Nursing, 56(5), doi: /j x NANDA International. (2009). Nursing diagnoses: Definitions and classification West Sussex, UK: Wiley-Blackwell. Ogasawara, C., Hasegawa, T., Kume, Y., Takahashi, I., Katayama, Y., Furuhashi, Y.,... (2005). Nursing diagnoses and interventions of Japanese patients with end-stage breast cancer admitted for different care purposes. International Journal of Nursing Terminologies and Classifications, 16(3 4), doi: /j X x Pesut, D. J., & Herman, J. (1999). Clinical reasoning: The art & science of critical & creative thinking. Albany, NY: Delmar. Polit, D. F., & Beck, C. T. (2008). Nursing research: Generating and assessing evidence for nursing practice. New York: Lippincott Williams & Wilkins. Woodtli, M. A., & van Ort, S. (1991). Nursing diagnoses and functional health patterns in patients receiving external radiation therapy: Cancer of head and neck. Nursing Diagnosis, 2(4), Retrieved from = Woodtli, M. A., & van Ort, S. (1993). Nursing diagnoses and functional health patterns in patients receiving external radiation therapy: Cancer of the digestive organs. Nursing Diagnosis, 4(1), Retrieved from ref= Young, W. H., & Hogben, D. (1978). An experimental study of the Delphi technique. Education Research and Perspectives Journal, 5, Retrieved from research/journal Appendix Relevance NANDA-I Nursing Diagnoses and Health Problems in Hematology Oncology Nursing; All Nursing Diagnoses Studied Relevance 1st Relevance 2nd Nursing diagnoses Delphi round Delphi round NANDA-I classification N %(n = 28) N %(n = 28) Domain 1: Health promotion Ineffective health maintenance 75* Impaired home maintenance 71.4* Readiness for enhanced nutrition 78.6* Readiness for enhanced self-health management 75* Domain 2: Nutrition Imbalanced nutrition: less than body requirements Risk of electrolyte imbalance Risk of deficient fluid volume Excess fluid volume International Journal of Nursing Terminologies and Classifications Volume 22, No. 2, April-June, 2011

13 Appendix Continued Relevance 1st Relevance 2nd Nursing diagnoses Delphi round Delphi round NANDA-I classification N %(n = 28) N %(n = 28) Risk of imbalanced fluid volume Deficient fluid volume Risk for unstable blood glucose level Impaired swallowing Risk of impaired liver function 75* Imbalanced nutrition: more than body requirements Domain 3: Elimination and exchange Diarrhea Risk of constipation Constipation Risk of dysfunctional gastrointestinal motility Bowel incontinence 71.4* Dysfunctional gastrointestinal motility Impaired urinary elimination 71.4* 71.4* Impaired gas exchange 67.9* Overflow urinary incontinence Domain 4: Activity/Rest Fatigue Risk for bleeding Activity intolerance Disturbed sleep pattern Risk for activity intolerance Bathing self-care deficit Risk of shock Feeding self-care deficit Dressing self-care deficit 78.6* 71.4* Impaired physical mobility 78.6* Impaired walking 75* 78.6* Impaired spontaneous ventilation 75* Toileting self-care deficit 75* Ineffective breathing pattern Impaired spontaneous ventilation Risk of ineffective renal perfusion Decreased cardiac output Domain 5: Perception/ Cognition Deficient knowledge Risk for acute confusion Acute confusion 78.6* 75* International Journal of Nursing Terminologies and Classifications Volume 22, No. 2, April-June,

14 Nursing Diagnoses (NANDA-I) in Hematology Oncology: A Delphi Study Appendix Continued Relevance 1st Relevance 2nd Nursing diagnoses Delphi round Delphi round NANDA-I classification N %(n = 28) N %(n = 28) Disturbed sensory perception (Specify: visual, 75* 71.4* auditory, kinesthetic, gustatory, tactile, or olfactory) Impaired memory Impaired verbal communication Domain 6: Self-perception Hopelessness Disturbed body image Situational low self-esteem Powerlessness Risk for loneliness 67.9* Domain 7: Role relationships Interrupted family processes Ineffective role performance Impaired social interaction 82.1 Domain 8: Sexuality Sexual dysfunction 96.4 Domain 9: Coping/stress tolerance Anxiety Ineffective coping Death anxiety Grieving Fear Disabled family coping Impaired individual resilience Ineffective denial 75* Complicated grieving 71.4* Domain 10: Life principles Noncompliance Decisional conflict Domain 11: Safety/ protection Risk for infection Impaired oral mucous membrane Risk of impaired skin integrity Hyperthermia Impaired skin integrity Risk of vascular trauma Impaired tissue integrity Risk for imbalanced body temperature International Journal of Nursing Terminologies and Classifications Volume 22, No. 2, April-June, 2011

15 Appendix Continued Relevance 1st Relevance 2nd Nursing diagnoses Delphi round Delphi round NANDA-I classification N %(n = 28) N %(n = 28) Risk of injury 78.6* Ineffective airway clearance 67.9* Ineffective protection 67.9* Risk for aspiration Domain 12: Comfort Nausea Acute pain Chronic pain Social isolation 85.7 Domain 13: Growth/ development Health problems Pruritis 100 Excessive sweating 89.3 Risk of or actual financial problems 67.9* Absolute consensus among the experts: 100%; High consensus among the experts: 80 99%; Moderate consensus among the experts: 65% 79%; *No consensus among the experts: <65%. International Journal of Nursing Terminologies and Classifications Volume 22, No. 2, April-June,