ORIGINAL ARTICLE
Ineffective breathing pattern in children and adolescents with congenital heart disease: accuracy of defining characteristics Beatriz Amorim Beltr~ ao, Tracy Heather Herdman, Lıvia Maia Pascoal, Daniel Bruno Resende Chaves, Viviane Martins da Silva and Marcos Venıcios de Oliveira Lopes
Aims and objectives. The purpose of this study was to assess the prevalence of ineffective breathing pattern and the accuracy of its defining characteristics, among children and adolescents with congenital heart disease. Background. The NANDA International nursing diagnosis, ineffective breathing pattern, has been noted to have high prevalence in different clinical contexts and age groups. Despite that, nurses continue to report difficulties in confirming this diagnosis. The lack of data regarding the sensitivity, specificity and predictive values of the defining characteristics contribute to decreased certainty in diagnosing ineffective breathing pattern. Design. A diagnostic accuracy study. Methods. This study of diagnostic accuracy was conducted with 61 children and adolescents with congenital heart disease. Two nurses were trained by the primary investigator on use of defining characteristics in the diagnostic process for ineffective breathing pattern. Results. Ineffective breathing pattern was present in 262% of the children and adolescents sampled. When analysing the defining characteristics, alterations in depth of breathing, showed high values of sensitivity and specificity. In addition, orthopnoea, tachypnoea and use of accessory muscles to breathe, showed high values of specificity; dyspnoea showed high values of sensitivity. Finally, assumption of three-point position, bradypnoea and increased anterior–posterior diameter were not found to be statistically significant for this sample population. Conclusion. Five defining characteristics of ineffective breathing pattern presented measures of accuracy with statistically significant values in children with congenital heart disease. Relevance to clinical practice. The findings can help nurses during the diagnostic process, as they identify which defining characteristics can be used to confirm or rule out the probability of occurrence of the diagnosis.
Authors: Beatriz Amorim Beltr~ao, RN, Doctoral Student, Nursing Department, Federal University of Ceara, Fortaleza, Brazil; Tracy Heather Herdman, PhD, RN, FNI, CEO/Executive Director, NANDA International, Inc., Kaukauna, WI, USA; Lıvia Maia Pascoal, RN, Assistant Professor, Nursing Department, Federal University of Maranh~ao, Imperatriz, MA; Daniel Bruno Resende Chaves, RN, Doctoral Student, Nursing Department, Federal University of Ceara, Fortaleza; Viviane Martins da Silva, PhD, RN,
© 2015 John Wiley & Sons Ltd Journal of Clinical Nursing, 24, 2505–2513, doi: 10.1111/jocn.12838
What does this paper contribute to the wider global clinical community?
• Measures of accuracy for defin-
•
•
ing characteristics of ineffective breathing pattern in children with congenital heart disease. The measures of accuracy of nursing diagnoses allow the precocious identification and screening of children with high probability to develop ineffective breathing pattern. The effectiveness of the care plan can be improved by an early and accurate identification of ineffective breathing pattern in children with congenital heart disease.
Assistant Professor, Nursing Department, Federal University of Cear a, Fortaleza; Marcos Venıcios de Oliveira Lopes, PhD, RN, Associate Professor, Nursing Department, Federal University of Cear a, Fortaleza, Brazil Correspondence: Beatriz Amorim Beltr~ ao, Doctoral Student, Federal University of Cear a, Rua Monsenhor Cat~ ao, 1491 – Meireles, Fortaleza, CE 60170-010, Brazil. Telephone: +55 85 9958-1325. E-mail: [email protected]
2505
BA Beltr~ao et al.
Key words: adolescent, child, clinical reasoning, ineffective breathing pattern, nursing diagnosis, respiration Accepted for publication: 14 Feruary 2015
Introduction Clinical reasoning represents the thinking processes by which nurses evaluate patient assessment data, diagnose human responses, attribute causes to these identified conditions and plan or refine nursing care to benefit the patients (Falan & Han 2013). Some authors indicated that clinical reasoning includes critical thinking and reflective problemsolving, whereas another definition indicates that it is a process of purposeful, self-regulatory judgement that gives reasoned consideration to evidence, contexts, conceptualisations, methods and criteria (Facione 1990, Facione & Facione 2008). The ability to make decisions about clinical events has been addressed by different terms within the health care literature, including decision-making, clinical reasoning, critical thinking and reflective reasoning (Falan & Han 2013). Clinical reasoning in nursing is a complex task that requires a high level of critical thinking, clinical decisionmaking ability, and a substantial knowledge base that includes nursing concepts, human responses and other information such as physiology/pathophysiology (Paans et al. 2012). Because it depends on the skills, experience and knowledge of nurses, the clinical reasoning process is uncertain, and will not determine with complete certainty the correct nursing diagnosis (Lunney 2003). However, it does support nurses in determining the probability of a patient having a particular nursing diagnosis (Levin 2012). To provide quality nursing care to patients, it is expected that nurses are able to assess clinical or patient events, to identify cues for the inference of nursing diagnoses. Accurate diagnosis is essential as the appropriate nursing diagnoses are needed to guide the interventions that should be implemented for the achievement of positive patient outcomes (Paans et al. 2012). Thus, the correct identification of nursing diagnoses should be based on clinical indicators that provide support for the diagnostic inference. Nurses should be able to identify defining characteristics that are sensitive and specific enough to determine a diagnosis. Therefore, the findings of studies that analyse the measures of accuracy may contribute to improving the accuracy of nursing diagnostic inference. The defining characteristics present relations of major or minor force with nursing diagnoses. Measures of accuracy
2506
such as sensitivity and specificity can be used to demonstrate these relationships and, consequently, assist nurses in their clinical reasoning process. Studies concerning the measures of accuracy may provide information to enable nurses to suspect or confirm the occurrence of certain nursing diagnosis. To analyse such measures, studies that address the sensitivity and specificity, based on pre-established reference standards, has been recommended (Lopes et al. 2012). Improvements in diagnosing may occur when accuracy studies assess the power of the defining characteristics (DCs) to correctly differentiate individuals with a nursing diagnosis from those without this diagnosis (Lopes et al. 2012). Nurses could use the measures of accuracy to analyse if a specific DC is more useful to confirm or to refute the occurrence of a nursing diagnosis (Levin 2012). Despite these advantages, only recently have studies addressed the accuracy of the DCs of the nursing diagnoses, and these are still scarce.
Background Regarding the nursing diagnoses presented by NANDA International, Inc. (NANDA-I), ineffective breathing pattern (00032 – IBP) has been found to occur with high prevalence in different clinical contexts and age groups (CarlsonCatalano et al. 1998, Silva et al. 2006b, Zeitoun et al. 2007). Despite this, nurses report difficulty in diagnosing IBP. Some of this difficulty is associated with the presence of DCs that are shared with other nursing diagnoses, such as ineffective airway clearance (00031 – IAC) and ineffective gas exchange (00032 – IGE). Authors who published about this nursing diagnosis report this fact as one of the main difficulties with diagnosing IBP (Andrade et al. 2012). Although several studies have been published on this nursing diagnosis, only a few included concerns about the measurements of accuracy. Of those that did, they referred to populations of children with asthma or acute respiratory infection (Silveira et al. 2008, Cavalcante et al. 2010, Pascoal et al. 2014). Thus, the lack of data regarding the sensitivity, specificity and predictive values of the DCs for IBP led to decreased certainty in diagnosing IBP. Another gap in the literature involves the accuracy of DCs of IBP among children with congenital heart disease
© 2015 John Wiley & Sons Ltd Journal of Clinical Nursing, 24, 2505–2513
Original article
(CHD). Due to cardiac malformation, children with CHD often have changes in lung structure and physiology (Markowitz & Fellows 1998). For these reasons, one might assume that the IBP diagnosis would be easily identified in this population; however, few studies have evaluated its occurrence in this specific population. Studies that do exist used samples of children younger than two years of age (Silva et al. 2004, 2006b). Therefore, there is a scarcity of literature regarding the prevalence of this diagnosis among older children and adolescents with CHD. In addition, some studies have shown that the IBP is a human response that precedes other more serious clinical conditions. For example, Pascoal et al. (2012) found that IBP occurred in the early period of hospitalisation of children with acute respiratory infection and proceeded to the occurrence of IAC. Furthermore, the survival time of the IBP was less than IAC and IGE in a study conducted among children with congenital heart disease (Silva et al. 2007). Thus, IBP is a warning sign to the worsening of the clinical condition of children and their early and accurate identification must be priority in the evaluation of children with congenital heart disease. Based on this information, this study aimed to assess the prevalence of the IBP nursing diagnosis and the accuracy of its DCs, among children and adolescents with CHD.
Method Design A study of diagnostic accuracy was conducted to determine the prevalence of DCs of the IBP diagnosis in children with CHD. This method also enabled the evaluation of the accuracy of these DCs in diagnosing IBP. As mentioned in the literature, studies of accuracy could be used to investigate the ability of the DCs to discriminate between patients with and without a specific nursing diagnosis (Lopes et al. 2012). For this study, the population of interest was children and adolescents, aged 5–17, receiving care for CHD. This investigation conformed to the principles outlined in the Declaration of Helsinki, and was approved by the local Ethics Committee (judgement number 795/10). Study objectives were explained to participants and the responsible parent, and all children and parents signed the Terms of Free and Informed Consent. The inclusion criteria were a confirmed medical diagnosis of CHD, ages between 5–17 years. Exclusion criteria included individuals who had comorbidities associated with CHD (e.g. genetic syndromes, neurological disorders) or physical or psychological conditions that impaired the © 2015 John Wiley & Sons Ltd Journal of Clinical Nursing, 24, 2505–2513
Accuracy of defining characteristics of IBP
assessment of the child or adolescent at rest (e.g. inability to sit for assessment, inability to converse or to understand directions). The sample was estimated using the following parameters: a confidence level of 95% (Za); conjectured sensitivity (Se) of 80%; length of confidence interval of 11% (d); and the estimated proportion of 86% of the children having the IBP diagnosis. The prevalence was estimated using the ratio of previous studies (Silva et al. 2006a,b), and the parameters were applied using the formula n = (Za)2 * Se(1 Se) /(d2 * P) (Zhou et al. 2011). The sample was estimated to require 61 children and adolescents.
Data collection The study was carried out in a public specialised teaching hospital, considered as a reference in cardiopulmonary diseases, which receives children for diagnose and specific treatment in cardiology and pneumology. Data were collected with parents/child who were inpatient at paediatric ward, or followed at an ambulatory treatment programme. The authors developed a data collection instrument and a research protocol for the data collection and evaluation. A two-part instrument, used in clinical data collection, included a patient/parent questionnaire and a physical assessment component. The research protocol identified conceptual and operational definitions for each DC. The research protocol was developed based on a review of literature and the NANDA-I DCs associated with the IBP diagnosis. Researches on electronic databases and textbooks were carried out to the development of the research protocol. Operational and conceptual definitions were then identified for each DC. The operational definitions were used to describe what should be measured and how the measurement can be performed; while the conceptual definitions were used to provide a connotative meaning established through the analysis of a certain concept. The conceptual and operational definitions were provided to three doctorally prepared nurse experts, with health assessment expertise and experience in research on nursing diagnosis. Those experts are members of NANDA-I and participants in national and international conferences on nursing diagnoses. The three experts were consulted to provide feedback on appropriateness of operational and conceptual definitions. Changes were recommended regarding specificity of a few of the definitions, and these changes were made. This instrument was then used to evaluate the data collected in the second instrument.
2507
BA Beltr~ao et al.
The data collection instrument contained the DCs of IBP and additional pulmonary assessment criteria identified in the literature. A consecutive sample of parents and children who were at the paediatric ward or attending the ambulatory treatment programme was used for this study. Parents/ child were contacted and informed about the study and invited to participate. An interview with the child/parent was conducted to answer those questions regarding their observations of characteristics of the disease that occurred at home, and also to obtain sociodemographic data. Once data were collected, the primary investigator used the operational and conceptual definitions to indicate if the DCs were present in each child. The presence or absence of these DCs was then entered into an Excel spreadsheet, for all assessed DCs: alterations in depth of breathing; altered chest excursion; assumption of three-point position; bradypnoea; dyspnoea; increased anterior–posterior diameter; nasal flaring; orthopnoea; pursed-lip breathing; tachypnoea; and use of accessory muscles to breathe. Five of the NANDA-I DCs were not used in this study because they required the use of pulmonary function testing, which was not performed for this research. This spreadsheet was then provided to the nurses involved in the diagnostic inference component of the research project. Two nurses who were completing their master’s degrees in nursing, with a focus on nursing diagnosis and previous publications in this area, were identified to participate in the diagnostic inference process. A training session was provided, lasting eight hours, which included a discussion of the diagnostic process, the IBP diagnosis, a review of all identified research studies conducted on DCs related to IBP from 1973– 2010, and a presentation of the developed research protocol. After completing the training session, the competency of the nurses in diagnostic inference of IBP was evaluated using a protocol that was previously developed, and which has been described in detail in an earlier publication (Lopes et al. 2012). Upon successful evaluation, the nurses were asked to diagnose the presence or absence of IBP. It is important to clarify that the diagnosis of IBP required absolute agreement between the two nurses. In those cases in which there was divergence of opinion, the two nurses were required to reach a consensus. Due to the chronic nature of CHD and IBP, it was necessary to evaluate patients in inpatient and outpatient settings. In addition, studies of accuracy require an understanding of the full spectrum of manifestations that occur within multiple contexts in which the diagnosis of interest occurs. Therefore, data collection occurred in an inpatient paediatric ward and an ambulatory treatment programme of a tertiary teaching hospital.
2508
Analysis The statistical analysis included data regarding the prevalence of IBP diagnosis and its DCs, as well as information such as sociodemographic data and conditions related to CHD. Statistical analysis was performed using the statistical package R, version 2.15.1 (R Core Team 2014). The measures of accuracy (sensitivity, specificity and predictive values) of the DCs were calculated based on the inferences made by nurses, after the consensus on the occurrence of IBP. Table 1 provides the statistical measures used for determining the accuracy of DCs. To verify the legitimacy of a test, to check if a DC was a legitimate indicator for nursing diagnosis, the chi-square test or Fisher tests were applied.
Results Sixty-one children and adolescents were evaluated in the study; ages ranged from 5–17 years of age, with a mean of 1015 years (311). The majority of the sample was male (541%). The most prevalent CHD were ventricular septal defect (262%), atrial septal defect (246%), tricuspid regurgitation (230%), pulmonary insufficiency (131%) and tricuspid insufficiency (115%). Many of these defects occurred together; therefore, most subjects had more than one cardiac malformation (MEAN = 157). Regarding CHD correction, most of the individuals evaluated had not undergone surgical correction (557%); those who had were still affected by residual heart disease, a fact that was essential for inclusion in this research. The IBP diagnosis was present in 262% of the children and adolescents sampled. Among the DCs, the most frequent were dyspnoea (459%) and alterations in depth of breathing (361%) (Table 2). The analysis of agreement between the nurses who performed the diagnostic inference in the study, regarding the occurrence of IBP, demonstrated a Cohen’s Kappa value of 087 (p < 0001). Table 3 provides the measures of accuracy of DCs. According to the nurses, the characteristic alterations in depth of breathing showed, at the same time, high values of sensitivity (9444) and specificity (8511) for IBP. Moreover, it showed high values of area under the ROC curve (0897) and efficiency (8769). The diagnostic odds ratio also indicated that, in the presence of this characteristic, there was a greater chance of the IBP diagnosis occurring (DOR = 7799; likelihood ratios with statistically significant values). Three DCs demonstrated high values of specificity and efficiency: orthopnoea (8889; 8525 respectively); tachypnoea (9556; 8689 respectively); and use of accessory mus© 2015 John Wiley & Sons Ltd Journal of Clinical Nursing, 24, 2505–2513
© 2015 John Wiley & Sons Ltd Journal of Clinical Nursing, 24, 2505–2513 Value >800%
Value >800%
Cut-off point used in this study
Value >800%
(+) Probability of IBP being present in patients with a specific DC
Predictive value
Value >800%
( ) Probability of IBP being absent in patients without a specific DC
DC, defining characteristic; IBP, ineffective breathing pattern; CI, confidence interval.
Probability of the absence of a DC in patients without IBP
Probability of a DC being present in patient with IBP
Specificity
Definition
Sensitivity
Value > 1 and CI 95% (does not include value of 100)
(+) Probability of the presence of a DC in patients with IBP, divided by the probability of this DC in patients without IBP
Likelihood ratio
Value < 1 and CI 95% (does not include value of 100)
( ) Probability of the absence of a DC in patients with IBP, divided by the probability of this DC in patients without IBP
Table 1 Statistical measures used for determining the accuracy of defining characteristics of the nursing diagnosis, ineffective breathing pattern
Measure of the effectiveness of a diagnostic test; defined as the ratio of the odds of the test being of the test being positive if the subject has IBP, relative to the odds of the test being positive if the subject does not have IBP Value > 1 and CI 95% (does not include value of 100)
Diagnostic odds ratio
p-value < 005
Probability of the agreement between the DC and IBP
Efficiency
Original article Accuracy of defining characteristics of IBP
2509
2510
Se, sensitivity; Sp, specificity; PV, predictive value; LR, likelihood ratio; DOR, diagnostic odds ratio; CI 95%, confidence interval of 95%; ROC, area under the ROC curve. *Fisher’s exact test. † Chi-square test.
Ineffective breathing pattern in children and adolescents with congenital heart disease: accuracy of defining characteristics Beatriz Amorim Beltr~ ao, Tracy Heather Herdman, Lıvia Maia Pascoal, Daniel Bruno Resende Chaves, Viviane Martins da Silva and Marcos Venıcios de Oliveira Lopes
Aims and objectives. The purpose of this study was to assess the prevalence of ineffective breathing pattern and the accuracy of its defining characteristics, among children and adolescents with congenital heart disease. Background. The NANDA International nursing diagnosis, ineffective breathing pattern, has been noted to have high prevalence in different clinical contexts and age groups. Despite that, nurses continue to report difficulties in confirming this diagnosis. The lack of data regarding the sensitivity, specificity and predictive values of the defining characteristics contribute to decreased certainty in diagnosing ineffective breathing pattern. Design. A diagnostic accuracy study. Methods. This study of diagnostic accuracy was conducted with 61 children and adolescents with congenital heart disease. Two nurses were trained by the primary investigator on use of defining characteristics in the diagnostic process for ineffective breathing pattern. Results. Ineffective breathing pattern was present in 262% of the children and adolescents sampled. When analysing the defining characteristics, alterations in depth of breathing, showed high values of sensitivity and specificity. In addition, orthopnoea, tachypnoea and use of accessory muscles to breathe, showed high values of specificity; dyspnoea showed high values of sensitivity. Finally, assumption of three-point position, bradypnoea and increased anterior–posterior diameter were not found to be statistically significant for this sample population. Conclusion. Five defining characteristics of ineffective breathing pattern presented measures of accuracy with statistically significant values in children with congenital heart disease. Relevance to clinical practice. The findings can help nurses during the diagnostic process, as they identify which defining characteristics can be used to confirm or rule out the probability of occurrence of the diagnosis.
Authors: Beatriz Amorim Beltr~ao, RN, Doctoral Student, Nursing Department, Federal University of Ceara, Fortaleza, Brazil; Tracy Heather Herdman, PhD, RN, FNI, CEO/Executive Director, NANDA International, Inc., Kaukauna, WI, USA; Lıvia Maia Pascoal, RN, Assistant Professor, Nursing Department, Federal University of Maranh~ao, Imperatriz, MA; Daniel Bruno Resende Chaves, RN, Doctoral Student, Nursing Department, Federal University of Ceara, Fortaleza; Viviane Martins da Silva, PhD, RN,
© 2015 John Wiley & Sons Ltd Journal of Clinical Nursing, 24, 2505–2513, doi: 10.1111/jocn.12838
What does this paper contribute to the wider global clinical community?
• Measures of accuracy for defin-
•
•
ing characteristics of ineffective breathing pattern in children with congenital heart disease. The measures of accuracy of nursing diagnoses allow the precocious identification and screening of children with high probability to develop ineffective breathing pattern. The effectiveness of the care plan can be improved by an early and accurate identification of ineffective breathing pattern in children with congenital heart disease.
Assistant Professor, Nursing Department, Federal University of Cear a, Fortaleza; Marcos Venıcios de Oliveira Lopes, PhD, RN, Associate Professor, Nursing Department, Federal University of Cear a, Fortaleza, Brazil Correspondence: Beatriz Amorim Beltr~ ao, Doctoral Student, Federal University of Cear a, Rua Monsenhor Cat~ ao, 1491 – Meireles, Fortaleza, CE 60170-010, Brazil. Telephone: +55 85 9958-1325. E-mail: [email protected]
2505
BA Beltr~ao et al.
Key words: adolescent, child, clinical reasoning, ineffective breathing pattern, nursing diagnosis, respiration Accepted for publication: 14 Feruary 2015
Introduction Clinical reasoning represents the thinking processes by which nurses evaluate patient assessment data, diagnose human responses, attribute causes to these identified conditions and plan or refine nursing care to benefit the patients (Falan & Han 2013). Some authors indicated that clinical reasoning includes critical thinking and reflective problemsolving, whereas another definition indicates that it is a process of purposeful, self-regulatory judgement that gives reasoned consideration to evidence, contexts, conceptualisations, methods and criteria (Facione 1990, Facione & Facione 2008). The ability to make decisions about clinical events has been addressed by different terms within the health care literature, including decision-making, clinical reasoning, critical thinking and reflective reasoning (Falan & Han 2013). Clinical reasoning in nursing is a complex task that requires a high level of critical thinking, clinical decisionmaking ability, and a substantial knowledge base that includes nursing concepts, human responses and other information such as physiology/pathophysiology (Paans et al. 2012). Because it depends on the skills, experience and knowledge of nurses, the clinical reasoning process is uncertain, and will not determine with complete certainty the correct nursing diagnosis (Lunney 2003). However, it does support nurses in determining the probability of a patient having a particular nursing diagnosis (Levin 2012). To provide quality nursing care to patients, it is expected that nurses are able to assess clinical or patient events, to identify cues for the inference of nursing diagnoses. Accurate diagnosis is essential as the appropriate nursing diagnoses are needed to guide the interventions that should be implemented for the achievement of positive patient outcomes (Paans et al. 2012). Thus, the correct identification of nursing diagnoses should be based on clinical indicators that provide support for the diagnostic inference. Nurses should be able to identify defining characteristics that are sensitive and specific enough to determine a diagnosis. Therefore, the findings of studies that analyse the measures of accuracy may contribute to improving the accuracy of nursing diagnostic inference. The defining characteristics present relations of major or minor force with nursing diagnoses. Measures of accuracy
2506
such as sensitivity and specificity can be used to demonstrate these relationships and, consequently, assist nurses in their clinical reasoning process. Studies concerning the measures of accuracy may provide information to enable nurses to suspect or confirm the occurrence of certain nursing diagnosis. To analyse such measures, studies that address the sensitivity and specificity, based on pre-established reference standards, has been recommended (Lopes et al. 2012). Improvements in diagnosing may occur when accuracy studies assess the power of the defining characteristics (DCs) to correctly differentiate individuals with a nursing diagnosis from those without this diagnosis (Lopes et al. 2012). Nurses could use the measures of accuracy to analyse if a specific DC is more useful to confirm or to refute the occurrence of a nursing diagnosis (Levin 2012). Despite these advantages, only recently have studies addressed the accuracy of the DCs of the nursing diagnoses, and these are still scarce.
Background Regarding the nursing diagnoses presented by NANDA International, Inc. (NANDA-I), ineffective breathing pattern (00032 – IBP) has been found to occur with high prevalence in different clinical contexts and age groups (CarlsonCatalano et al. 1998, Silva et al. 2006b, Zeitoun et al. 2007). Despite this, nurses report difficulty in diagnosing IBP. Some of this difficulty is associated with the presence of DCs that are shared with other nursing diagnoses, such as ineffective airway clearance (00031 – IAC) and ineffective gas exchange (00032 – IGE). Authors who published about this nursing diagnosis report this fact as one of the main difficulties with diagnosing IBP (Andrade et al. 2012). Although several studies have been published on this nursing diagnosis, only a few included concerns about the measurements of accuracy. Of those that did, they referred to populations of children with asthma or acute respiratory infection (Silveira et al. 2008, Cavalcante et al. 2010, Pascoal et al. 2014). Thus, the lack of data regarding the sensitivity, specificity and predictive values of the DCs for IBP led to decreased certainty in diagnosing IBP. Another gap in the literature involves the accuracy of DCs of IBP among children with congenital heart disease
© 2015 John Wiley & Sons Ltd Journal of Clinical Nursing, 24, 2505–2513
Original article
(CHD). Due to cardiac malformation, children with CHD often have changes in lung structure and physiology (Markowitz & Fellows 1998). For these reasons, one might assume that the IBP diagnosis would be easily identified in this population; however, few studies have evaluated its occurrence in this specific population. Studies that do exist used samples of children younger than two years of age (Silva et al. 2004, 2006b). Therefore, there is a scarcity of literature regarding the prevalence of this diagnosis among older children and adolescents with CHD. In addition, some studies have shown that the IBP is a human response that precedes other more serious clinical conditions. For example, Pascoal et al. (2012) found that IBP occurred in the early period of hospitalisation of children with acute respiratory infection and proceeded to the occurrence of IAC. Furthermore, the survival time of the IBP was less than IAC and IGE in a study conducted among children with congenital heart disease (Silva et al. 2007). Thus, IBP is a warning sign to the worsening of the clinical condition of children and their early and accurate identification must be priority in the evaluation of children with congenital heart disease. Based on this information, this study aimed to assess the prevalence of the IBP nursing diagnosis and the accuracy of its DCs, among children and adolescents with CHD.
Method Design A study of diagnostic accuracy was conducted to determine the prevalence of DCs of the IBP diagnosis in children with CHD. This method also enabled the evaluation of the accuracy of these DCs in diagnosing IBP. As mentioned in the literature, studies of accuracy could be used to investigate the ability of the DCs to discriminate between patients with and without a specific nursing diagnosis (Lopes et al. 2012). For this study, the population of interest was children and adolescents, aged 5–17, receiving care for CHD. This investigation conformed to the principles outlined in the Declaration of Helsinki, and was approved by the local Ethics Committee (judgement number 795/10). Study objectives were explained to participants and the responsible parent, and all children and parents signed the Terms of Free and Informed Consent. The inclusion criteria were a confirmed medical diagnosis of CHD, ages between 5–17 years. Exclusion criteria included individuals who had comorbidities associated with CHD (e.g. genetic syndromes, neurological disorders) or physical or psychological conditions that impaired the © 2015 John Wiley & Sons Ltd Journal of Clinical Nursing, 24, 2505–2513
Accuracy of defining characteristics of IBP
assessment of the child or adolescent at rest (e.g. inability to sit for assessment, inability to converse or to understand directions). The sample was estimated using the following parameters: a confidence level of 95% (Za); conjectured sensitivity (Se) of 80%; length of confidence interval of 11% (d); and the estimated proportion of 86% of the children having the IBP diagnosis. The prevalence was estimated using the ratio of previous studies (Silva et al. 2006a,b), and the parameters were applied using the formula n = (Za)2 * Se(1 Se) /(d2 * P) (Zhou et al. 2011). The sample was estimated to require 61 children and adolescents.
Data collection The study was carried out in a public specialised teaching hospital, considered as a reference in cardiopulmonary diseases, which receives children for diagnose and specific treatment in cardiology and pneumology. Data were collected with parents/child who were inpatient at paediatric ward, or followed at an ambulatory treatment programme. The authors developed a data collection instrument and a research protocol for the data collection and evaluation. A two-part instrument, used in clinical data collection, included a patient/parent questionnaire and a physical assessment component. The research protocol identified conceptual and operational definitions for each DC. The research protocol was developed based on a review of literature and the NANDA-I DCs associated with the IBP diagnosis. Researches on electronic databases and textbooks were carried out to the development of the research protocol. Operational and conceptual definitions were then identified for each DC. The operational definitions were used to describe what should be measured and how the measurement can be performed; while the conceptual definitions were used to provide a connotative meaning established through the analysis of a certain concept. The conceptual and operational definitions were provided to three doctorally prepared nurse experts, with health assessment expertise and experience in research on nursing diagnosis. Those experts are members of NANDA-I and participants in national and international conferences on nursing diagnoses. The three experts were consulted to provide feedback on appropriateness of operational and conceptual definitions. Changes were recommended regarding specificity of a few of the definitions, and these changes were made. This instrument was then used to evaluate the data collected in the second instrument.
2507
BA Beltr~ao et al.
The data collection instrument contained the DCs of IBP and additional pulmonary assessment criteria identified in the literature. A consecutive sample of parents and children who were at the paediatric ward or attending the ambulatory treatment programme was used for this study. Parents/ child were contacted and informed about the study and invited to participate. An interview with the child/parent was conducted to answer those questions regarding their observations of characteristics of the disease that occurred at home, and also to obtain sociodemographic data. Once data were collected, the primary investigator used the operational and conceptual definitions to indicate if the DCs were present in each child. The presence or absence of these DCs was then entered into an Excel spreadsheet, for all assessed DCs: alterations in depth of breathing; altered chest excursion; assumption of three-point position; bradypnoea; dyspnoea; increased anterior–posterior diameter; nasal flaring; orthopnoea; pursed-lip breathing; tachypnoea; and use of accessory muscles to breathe. Five of the NANDA-I DCs were not used in this study because they required the use of pulmonary function testing, which was not performed for this research. This spreadsheet was then provided to the nurses involved in the diagnostic inference component of the research project. Two nurses who were completing their master’s degrees in nursing, with a focus on nursing diagnosis and previous publications in this area, were identified to participate in the diagnostic inference process. A training session was provided, lasting eight hours, which included a discussion of the diagnostic process, the IBP diagnosis, a review of all identified research studies conducted on DCs related to IBP from 1973– 2010, and a presentation of the developed research protocol. After completing the training session, the competency of the nurses in diagnostic inference of IBP was evaluated using a protocol that was previously developed, and which has been described in detail in an earlier publication (Lopes et al. 2012). Upon successful evaluation, the nurses were asked to diagnose the presence or absence of IBP. It is important to clarify that the diagnosis of IBP required absolute agreement between the two nurses. In those cases in which there was divergence of opinion, the two nurses were required to reach a consensus. Due to the chronic nature of CHD and IBP, it was necessary to evaluate patients in inpatient and outpatient settings. In addition, studies of accuracy require an understanding of the full spectrum of manifestations that occur within multiple contexts in which the diagnosis of interest occurs. Therefore, data collection occurred in an inpatient paediatric ward and an ambulatory treatment programme of a tertiary teaching hospital.
2508
Analysis The statistical analysis included data regarding the prevalence of IBP diagnosis and its DCs, as well as information such as sociodemographic data and conditions related to CHD. Statistical analysis was performed using the statistical package R, version 2.15.1 (R Core Team 2014). The measures of accuracy (sensitivity, specificity and predictive values) of the DCs were calculated based on the inferences made by nurses, after the consensus on the occurrence of IBP. Table 1 provides the statistical measures used for determining the accuracy of DCs. To verify the legitimacy of a test, to check if a DC was a legitimate indicator for nursing diagnosis, the chi-square test or Fisher tests were applied.
Results Sixty-one children and adolescents were evaluated in the study; ages ranged from 5–17 years of age, with a mean of 1015 years (311). The majority of the sample was male (541%). The most prevalent CHD were ventricular septal defect (262%), atrial septal defect (246%), tricuspid regurgitation (230%), pulmonary insufficiency (131%) and tricuspid insufficiency (115%). Many of these defects occurred together; therefore, most subjects had more than one cardiac malformation (MEAN = 157). Regarding CHD correction, most of the individuals evaluated had not undergone surgical correction (557%); those who had were still affected by residual heart disease, a fact that was essential for inclusion in this research. The IBP diagnosis was present in 262% of the children and adolescents sampled. Among the DCs, the most frequent were dyspnoea (459%) and alterations in depth of breathing (361%) (Table 2). The analysis of agreement between the nurses who performed the diagnostic inference in the study, regarding the occurrence of IBP, demonstrated a Cohen’s Kappa value of 087 (p < 0001). Table 3 provides the measures of accuracy of DCs. According to the nurses, the characteristic alterations in depth of breathing showed, at the same time, high values of sensitivity (9444) and specificity (8511) for IBP. Moreover, it showed high values of area under the ROC curve (0897) and efficiency (8769). The diagnostic odds ratio also indicated that, in the presence of this characteristic, there was a greater chance of the IBP diagnosis occurring (DOR = 7799; likelihood ratios with statistically significant values). Three DCs demonstrated high values of specificity and efficiency: orthopnoea (8889; 8525 respectively); tachypnoea (9556; 8689 respectively); and use of accessory mus© 2015 John Wiley & Sons Ltd Journal of Clinical Nursing, 24, 2505–2513
© 2015 John Wiley & Sons Ltd Journal of Clinical Nursing, 24, 2505–2513 Value >800%
Value >800%
Cut-off point used in this study
Value >800%
(+) Probability of IBP being present in patients with a specific DC
Predictive value
Value >800%
( ) Probability of IBP being absent in patients without a specific DC
DC, defining characteristic; IBP, ineffective breathing pattern; CI, confidence interval.
Probability of the absence of a DC in patients without IBP
Probability of a DC being present in patient with IBP
Specificity
Definition
Sensitivity
Value > 1 and CI 95% (does not include value of 100)
(+) Probability of the presence of a DC in patients with IBP, divided by the probability of this DC in patients without IBP
Likelihood ratio
Value < 1 and CI 95% (does not include value of 100)
( ) Probability of the absence of a DC in patients with IBP, divided by the probability of this DC in patients without IBP
Table 1 Statistical measures used for determining the accuracy of defining characteristics of the nursing diagnosis, ineffective breathing pattern
Measure of the effectiveness of a diagnostic test; defined as the ratio of the odds of the test being of the test being positive if the subject has IBP, relative to the odds of the test being positive if the subject does not have IBP Value > 1 and CI 95% (does not include value of 100)
Diagnostic odds ratio
p-value < 005
Probability of the agreement between the DC and IBP
Efficiency
Original article Accuracy of defining characteristics of IBP
2509
2510
Se, sensitivity; Sp, specificity; PV, predictive value; LR, likelihood ratio; DOR, diagnostic odds ratio; CI 95%, confidence interval of 95%; ROC, area under the ROC curve. *Fisher’s exact test. † Chi-square test.
