Nursing Diagnoses, Interventions, and Activities As Described by a Nursing Minimum Data Set

Copyright B 2018 Wolters Kluwer Health, Inc. All rights reserved. Gianfranco Sanson, PhD Rosaria Alvaro, MSN Antonello Cocchieri, PhD Ercole Vellone, PhD John Welton, PhD Massimo Maurici, MD Maurizio Zega, PhD Fabio D’Agostino, PhD Nursing Diagnoses, Interventions, and Activities as Described by a Nursing Minimum Data Set

A Prospective Study in an Oncology Hospital Setting

KEY WORDS Background: Oncological diseases affect the biopsychosocial aspects of a Nursing activities person’s health, resulting in the need for complex multidisciplinary care. The quality Nursing diagnosis and outcomes of healthcare cannot be adequately assessed without considering the Nursing interventions contribution of nursing care, whose essential elements such as the nursing diagnoses Nursing Minimum Data Set (NDs), nursing interventions (NIs), and nursing activities (NAs) can be recorded in Oncology the Nursing Minimum Data Set (NMDS). There has been little research using the NMDS in oncology setting. Objective: The aim of this study was to describe the prevalence and distribution of NDs, NIs, and NAs and their relationship across patient age and medical diagnoses. Methods: This was a prospective observational study. Data were collected between July and December 2014 through an NMDS and the hospital discharge register in an Italian hospital oncology unit. Results: On average, for each of 435 enrolled patients, 5.7 NDs were identified on admission; the most frequent ND was risk for infection. During the hospital stay, 16.2 NIs per patient were planned, from which 25.2 NAs per day per patient were delivered. Only a third of NAs were based on a medical order, being the highest percentage delivered on nursing prescriptions. The number of NDs, NIs, and NAs was not related to patient age, but differed significantly among medical diagnoses. Conclusions: An NMDS can depict patient needs and nursing care delivered in oncology patients. Such data can effectively describe nursing contribution to patient

Author Affiliations: Department of Biomedicine and Prevention, University of The authors have no conflicts of interest to disclose. Rome Tor Vergata, Rome, Italy (Drs Sanson, Vellone, Maurici, D’Agostino and Correspondence: Fabio D’Agostino, PhD, Department of Biomedicine and Alvaro); University of Colorado College of Nursing, Aurora (Dr Welton); and Prevention, University of Rome Tor Vergata, Via Montpellier, 1, 00133 Rome, University Hospital Agostino Gemelli, Rome, Italy (Drs Cocchieri and Zega). Italy ([email protected]). This work was funded by the Italian Center of Excellence for Nursing Accepted for publication November 13, 2017. Scholarship, Rome, Italy. DOI: 10.1097/NCC.0000000000000581

Nursing Diagnoses, Interventions, and Activities Cancer NursingA, Vol. 00, No. 0, 2018 n 1

Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. care. Implications for Practice: The use of an NMDS raises the visibility of nursing care in the clinical records. Such data enable comparison and benchmarking with other healthcare professions and international data.

urses form the largest portion of health personnel, patient demographics (eg, age, sex), services (eg, patient’s health providing up to 80% of care in most of the world’s register number, date of admission and discharge, payment source), health services.1 Nurses are an essential part of any and nursing care (ND, nursing interventions [NIs], nursing out- N 20 health organization pursuing the goal of quality of care and comes). This data set can meet the information needs of mul- patient safety.2 For this reason, the quality of healthcare cannot tiple data users in the healthcare system, to make nursing visible be adequately determined without evaluating the effectiveness in the fields of research and healthcare effectiveness and policy of nursing in achieving patient outcomes.3 analysis.21 Regrettably, the use of the NMDS is still limited Oncological diseases represent major public health prob- worldwide.22 Only a few studies have described, mostly retro- lems,4 resulting in the need for complex treatments and high- spectively and on small groups of patients, NDs and NIs in intensity multidisciplinary care for extended periods. Such oncology hospital settings.5 No study based on electronic NMDS pathologies affect the physical integrity of cancer patients and has studied NDs, NIs, and NAs and their relationship across heavily impact their psychological health and life goals by alter- patient age and medical diagnoses in oncologic hospitalized ing their body image and impacting family, work, and incomes.5 patients. Thus, the medical diagnosis cannot exhaustively or exclusively describe the overall effects of oncological diseases on patients. Aims Using the nursing process, nurses do ensure that all aspects of patient care are identified, addressed, and documented in a The aims of this study were to (1) describe the prevalence of uniform and holistic way.6 This process gives nurses responsi- NANDA International (NANDA-I)23 NDs on admission, NIs, bility for assessing health problems and interpreting the acquired and NAs as described according to the Italian Nomenclature of information to make decisions about interventions.7 Also, the Nursing Care Performance24 in a population of hospitalized nursing process, using nursing diagnoses (NDs), complements oncology patients and (2) compare the number of NDs, NIs, a medical diagnosis to better describe care complexity.8 and NAs across (a) patient age and (b) medical diagnoses. Despite its importance, nursing practice is generally poorly represented in electronic health records,9 and this happens because nursing data are often incomplete and recorded without n Materials and Methods a standardized terminology.3,10,11 This makes it impossible to analyze the impact of nursing care on health processes and out- Study Design, Setting, and Population comes, and consequently the contribution of nurses remains invisible.12 A prospective observational study was performed at the Uni- Although many years have passed since standardized NDs versity Hospital ‘‘Agostino Gemelli’’ in Rome, Italy. The hos- have been included into oncology nursing practice to ensure pital has 1547 beds organized into 8 departments and 55 inpatient comprehensive and individualized care for the oncology pa- units. The study was carried out in a 21-bed oncology unit. All tient,13 their use is still largely neglected in daily clinical prac- consecutive adult patients admitted to this unit over a period of tice. A study conducted in outpatient cancer settings reported 6months (July 1 to December 31, 2014) were considered eligible that, in daily nursing practice, nurses collect a very large number for inclusion in the study. We excluded: (a) patients with length of nonstandardized data to document the care they provide (29 of stay (LOS) of less than 2days because they could be assimilated variables, on average). These data create multiple nonuniform to ‘‘day-hospital’’ admissions, but patients who died were in- ‘‘hidden nursing data sets.’’14 Collecting standardized nursing cluded even if their LOS was shorter than 2days; (b)patients data as part of hospital care in oncology may allow a better transferred to another inpatient unit before hospital discharge; (c) understanding of patients’ complexity and allow analysis of the patients for which nursing planning was not fully documented; impact of nursing on outcomes of care.15 The analysis of such (d) patients discharged with ‘‘V64.x’’ coding according to the data may permit a more detailed measure of nursing workload. ICD-9-CM taxonomy (International Classification of Diseases, Even if nursing workload is mostly measured in relation of spe- Ninth Revision, Clinical Modification,Italianversion)25 corre- cific tasks performed by nurses,16 the complexity of care re- sponding to patients admitted for specific procedures not quired by the patients as expressed by NDs can affect the carried out because of unforeseen circumstances. number of nursing activities (NAs) needed.17,18 To gather uniform nursing information, it is necessary to Data Collection identify a standardized set of data able to intercept the essential elements of clinical nursing practice across various settings and Data were collected from the Professional Assessment Instru- patient groups.19 The Nursing Minimum Data Set (NMDS) ment (PAI)10 and the hospital discharge register. The PAI is a identifies and operationalizes a powerful but limited set of data clinical nursing information system implemented at the study representing core components of nursing practice, including hospital in February 2013. This system allows collection of the

2 n Cancer NursingA, Vol. 00, No. 0, 2018 Sanson et al

Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. standard and uniform essential nursing data recommended by tient’s health code and medical record number) through a the NMDS (eg, NDs and NIs), patient’s sociodemographic probabilistic matching process30 completing the final study variables (eg, age, gender, job, level of education), and orga- data set. nizational information (eg, identification of responsible nurses, admission and discharge dates).26 The PAI includes a set of 44 Ethical Considerations NANDA-I diagnoses27 selected on the basis of a literature review.10 Nurses are assisted to identify NDs by a previously vali- The study was approved by the hospital ethics committee. All dated clinical decision support assessment tool28 integrated into data by which patients could be identified were anonymized, the PAI. Based on clinical findings collected during nursing as- and a progressive and unique numeric code was assigned to sessment, this tool suggests the possible NDs that should be each anonymized record to be included in the data set. confirmed by nurses. If nurses identify an ND that is not present in the set of 44 NDs, they have also the option to insert ‘‘other’’ Data Analysis NDs as free-text entry (in this study, we have labeled these NDs Statistical analysis was performed using the software IBM SPSS as ‘‘other’’). Statistics for Windows, version 21.0 (IBM Corp, Armonk, New Once an ND is generated by the PAI and confirmed by nurses, York). Nominal variables (eg, gender, ND, NI, and NA dis- nurses identify 1 or more appropriate NIsVdefined as any treat- tribution) were displayed as numbers and percentages. Contin- ment that a nurse performs to enhance patient outcomes6Vto be uous variables (eg, age, number of NDs on admission, NIs, and included in the patient’s nursing care plan. A set of 29 standard NAs) were displayed as mean, SD, median, and range. The NIs is included in the PAI. These NIs were developed from a differences between the means in 2 groups (eg, mean age be- review of the Italian Nomenclature of Nursing Care Perfor- tween genders) were analyzed by an unpaired Student t test, mance.24 Because the information in an NI is at a general level after considering whether the subgroups had equal variance that does not allow detailed analysis of the specific activities using Levene’s test. One-way analysis of variance and Tukey- associated with such interventions, each NI is associated with a Kramer post hoc tests were applied for all comparisons among set of related NAs (from 2 to 32 NAs for each NI, 340 NAs more than 2 groups (eg, average number of NDs/NIs/NAs overall), defined as specific actions and behaviors adopted by among medical diagnoses). Bivariate associations between the nurses to carry out the intervention in practice. For example, number of NDs, NIs, NAs, and patient age were investigated for the NI named biological samples collection, there are 14 NAs with Spearman’s nonparametric correlation coefficient (>); (eg, venous blood collection, capillary blood collection, sterile positive or negative correlation strengths were interpreted as urine specimen collection). Therefore, the total number of NAs follows: 0 to 0.29, little (if any); 0.30 to 0.49, low; 0.50 to 0.69, (n=340) is much higher than that of the NIs (n=29) in the PAI. moderate; 0.70 to 0.89, high; and 0.90 to 1, very high.31 For all Moreover, the same NA or NI may be associated with different tests, an ! level of P=.05 was set for statistical significance. NDs and vice versa. All NAs belonging to administration of therapies (eg, administration of drugs, oxygen, medications by inhalation, arti- n Results ficial nutrition, and blood or related products) and to diagnostic procedures (eg, arterial, venous, or capillary blood draw; non- Population General Data sterile or sterile collection of excretions; secretion; and other biological materials) were delivered under medical order, because in During the study period, 439 patients were admitted to the Italy only physicians can order such interventions. The re- oncology unit. Four patients were excluded (1 was discharged maining NAs were delivered under autonomous nursing pre- after only 1day, 2 were transferred to another unit, and 1 patient scription after patient’s clinical assessment according to Italian was discharged with a V64.3 diagnosis). The final study popu- law,29 based on the best nursing practice treatments and pro- lation comprised 435 patients; 207 (47.6%) were females, and cedures, evidence-based guidelines, and standards of practice. 228 (52.4%) were male; the mean age was 59.5 (SD, 13.5) years All the above data are routinely and real-time recorded in the (median, 62years; range, 20Y86years). The ages of the 2 genders PAI system. were comparable (females: 58.7 [SD, 13.8] years; males: 60.3 For the purposes of this study, the following variables were [SD, 13.2] years; t=j1.169; P=.243). considered: (a) NDs identified by the nurses within 24hours For the vast majority of patients (n=369 [84.9%]), the med- of the patient’s admission, (b) planned NIs (NIs for which ical diagnosis belonged to neoplastic diseases ICD-9-CM diag- provision was planned by nurses in care plans), (c) delivered nostic categories; in particular, 4 of these diagnostic categories NIs (NIs actually delivered to the patient during his/her stay), described 73.3% of the population (antineoplastic chemother- and (d) NAs (specific NAs belonging to each NI actually de- apy, n=163 [37.5%]; secondary malignant neoplasms of respi- livered to the patient during his/her stay). ratory and digestive apparatus, n=57 [13.1%]; malignant neoplasms Data obtained from hospital discharge register were: (a)main of digestive organs and peritoneum, n=56 [12.9%]; malignant admission medical diagnosis, coded according to the ICD-9-CM; neoplasms of respiratory and intrathoracic organs, n=43 [9.9%]). (b) LOS, in days; and (c) condition at discharge (deceased, trans- The average LOS was 7.8 (SD, 6.3) days (median, 5days; ferred to another unit, discharged). The data of the PAI and range, 1Y49days). A total of 110 patients (25.3%) had an LOS hospital discharge register were linked via 2 key variables (pa- exceeding the 75th percentile (9days) of the enrolled population

Nursing Diagnoses, Interventions, and Activities Cancer NursingA, Vol. 00, No. 0, 2018 n 3

Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. LOS. The average LOS was significantly different (F=55.367, Table 1 & Frequency Distribution of Nursing PG.001) among the 4 most frequent medical diagnostic cate- Diagnoses on Admission (n=2460) in gories. Indeed, LOS was lower for 2 medical diagnoses (antineo- the Study Population (n=435) plastic chemotherapy, 4.1 [SD, 1.3], and secondary malignant neoplasm of respiratory and digestive apparatus, 6.3 [SD, 4.3] Nursing Diagnosis n % days) than for the other 2 medical diagnoses (malignant neo- Risk for infection 359 82.5 plasm of respiratory and intrathoracic organs, 10.7 [SD, 7.4], Risk for impaired skin integrity 183 42.1 and malignant neoplasm of digestive organs and peritoneum, Disturbed sleep pattern 167 38.4 10.9 [SD, 5.5] days). Four hundred nine patients (94%) were Imbalanced nutrition: less than body requirements 161 37.0 discharged from hospital, and 26 (6%) died during hospitalization. Risk for constipation 159 36.6 Anxiety 114 26.2 Activity intolerance 101 23.2 Prevalence of Nursing Diagnoses, Nursing Acute pain 100 23.0 Interventions, and Nursing Activities Impaired physical mobility; risk for activity 98 22.5 intolerance Overall, 2460 NDs were identified on admission (5.7 [SD, 3.7] Risk for falls 84 19.3 per patient; median, 5; range, 0-24). Risk for infection was the Bathing self-care deficit 82 18.9 most frequently assigned ND (Table 1). Constipation 77 17.7 A total of 7052 NIs were planned (16.2 [SD, 3.1] NI per Dressing self-care deficit 63 14.5 patient; median, 16; range, 8Y25). Five interventions (biological Ineffective peripheral tissue perfusion 61 14.0 samples collection, medications/blood products administration, vas- Imbalanced nutrition: more than body 58 13.3 cular devices, sleep and rest, welcoming/information)wereincluded requirements in care planning of more than 99% of patients (Table 2). Overall, Impaired walking 57 13.1 6405 of 7052 planned NIs were actually delivered (90.8%, 14.7 Risk for injury 52 12.0 [SD, 3.2] per patient; median, 14; range, 7Y25) through a vary- Toileting self-care deficit 45 10.3 ing number of NAs. Considering all enrolled patients for the Chronic pain 40 9.2 Impaired urinary elimination 39 9.0 whole hospital stay, nurses delivered a total of 91233 NAs Deficient fluid volume 38 8.7 (209.7 [SD, 214.6]) NAs per patient; median, 123; range, 30- Ineffective breathing pattern 36 8.3 1668), for an average of 25.2 (SD, 7.4) per day of hospitaliza- Diarrhea 32 7.4 tion (median, 24.2; range, 9.8Y71.0). Bowel incontinence 22 5.1 A total of 1184 NAs (1.3%) were recorded as ‘‘other activity’’ Fear; impaired skin integrity; impaired swallowing 20 4.6 because a specific description was not available in the set of the Feeding self-care deficit 14 3.2 PAI NAs. Among the 89309 specifically coded activities, more Risk for aspiration 12 2.8 than 50% were associated with NIs medications/blood products Fatigue 10 2.3 administration (21817 [24.4%]), comfort and pain management Perceived constipation 9 2.1 (15111 [16.9%]), and vascular devices (8987 [10.1%]). A large Disturbed body image 7 1.6 variability between the actual number of NAs delivered for the Ineffective coping 4 0.9 Reflex urinary incontinence; impaired memory 3 0.7 different NIs was documented (Table 3). Acute confusion; chronic confusion; 2 0.5 Overall, 31.6% of NAs (8.3 NAs per day per patient) were noncompliance; urge urinary incontinence; othera delivered under medical order. In detail, 23535 of such NAs Ineffective airway clearance; impaired social 1 0.2 (26.4%, 6.9 NAs per day per patient) were related to admin- interaction istration of therapies (eg, administration of drugs, oxygen, med- Functional urinary incontinence; stress urinary 0 0.0 ications by inhalation, artificial nutrition, and blood or related incontinence products), and a further 4746 NAs (5.3%, 1.4 NA per day per aNursing diagnosis not included in the set of 44 nursing diagnoses of the patient) were related to diagnostic procedures (eg, arterial, venous, Professional Assessment Instrument. or capillary blood draw; nonsterile or sterile collection of excretions; secretion; and other biological materials). No correlation was found between the number of delivered NIs and patient’s age (>=j0.033, P=.496), while it varied sig- Nursing Diagnoses, Interventions, and Activities nificantly among medical diagnoses (F=18.047, PG.001); post Across Patient Age and Medical Diagnoses hoc tests showed a significantly higher number of planned NIs for patients diagnosed as having a malignant neoplasm of respi- No correlation was found between the number of identified ratory and intrathoracic organs (16.9 [SD, 3.4] NIs) compared NDs and patient’s age (>=j0.008, P=.864). The number of with the other 3 categories (ranging from 13.5 to 14.9 NIs). NDs differed significantly among the medical diagnoses (F=8.151, No correlation was found between the number of NAs per PG.001); post hoc tests showed a significantly higher number of day and patient’s age (>=0.030, P=.533). The number of NDs for patients diagnosed with malignant neoplasm of respi- delivered NAs varied significantly among medical diagnoses ratory and intrathoracic organs (7.1 [SD, 4.3] NDs) compared (F=43.610, PG.001); post hoc tests showed a significantly higher with the other 3 categories (ranging from 4.4 to 5.1 NDs). number of NAs per day for patients diagnosed as having a

4 n Cancer NursingA, Vol. 00, No. 0, 2018 Sanson et al

Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. Table 2 & Frequency Distribution of Planned The most frequently documented ND was risk for infection, with a prevalence of 83% among enrolled patients. This Nursing Interventions (n=7052) in 5 the Study Population (n=435) finding is consistent with a literature review that showed the following NANDA-I NDs in adult hospitalized oncology pop- Planned Nursing Intervention n % ulations as the more frequent (from 20% to 100%): anxiety, Biological samples collection 435 100.0 knowledge deficit, constipation, bathing/hygiene self-care deficit, Medications/blood products administration 435 100.0 disturbed body image, acute/chronic pain, fear, disturbed sleep Vascular devices 434 99.8 pattern, risk for infection,anddeficient fluid volume.Inour Sleep and rest 433 99.5 study, although with different prevalence rates compared with Welcoming/information 432 99.3 the review noted above, all NDs were present except for knowl- Vital signs measurement/monitoring 427 98.2 edge deficit. The reason for this may be related to the absence Feeding/hydration 419 96.3 of this diagnosis in the PAI system. Consequently, even if de- Comfort and pain management 415 95.4 tected, this ND could not have been selected. However, nurses Safety/risk reduction 399 91.7 had the possibility to insert ‘‘other’’ NDs as free-text entry in Surgical care/diagnostics 388 89.2 Bowel elimination 319 73.3 the PAI, but only 2 ‘‘other’’ NDs were identified by nurses. It Dependency levels assessment 298 68.5 is important to emphasize that the use of a knowledge deficit 32,33 Hygiene/care of the tissues 274 63.0 diagnosis has been criticized over time, and NANDA-I 23 Urinary elimination 270 62.1 does not attribute any level of evidence to this ND. Amore Collaboration with other professions 258 59.3 rigorous comparison of ND epidemiology in these populations Continuity of care 246 56.6 should therefore start from a preliminary sharing of an iden- Mobility and correct posture 234 53.8 tical data collection modality.34 Even with this limitation, the Thermoregulation 216 49.7 identification of NDs may have an important impact both from Cardiovascular assessment/monitoring 171 39.3 clinical and organizational point of view; for example, the level Neurocognitive assessment/monitoring 135 31.0 of nursing complexity described by the number of NDs strongly Communication: guidance/support 127 29.2 influences the time spent by nurses with the patient and there- Breathing 119 27.4 34 Dressing and undressing 69 15.9 fore the nursing workload. Worship and personal values 30 6.9 In our population, none of the problems described by an Health education and learning 25 5.7 ND could in any way be deduced from the medical diagnosis. 8 End-of-life care 23 5.3 This is similar to the finding of Welton and Halloran. This Adaptation (coping) assistance 15 3.4 finding emphasizes how the nursing diagnostic process is essen- Orientation in using health system 4 0.9 tial to a complete definition of the overall impact of oncologic disease. This finding highlights the complexity of nursing care delivered to oncology patients. Care that is delivered only con- malignant neoplasm of the respiratory and intrathoracic organs sidering medical diagnoses would miss important interventions (27.6 [SD, 8.3] NAs) compared with the other 3 categories that only nurses are likely to give. (malignant neoplasms of digestive organs and peritoneum, 24.3 The analysis of PAI NMDS allowed us to uniquely associate [SD, 6.8] NAs per day; secondary malignant neoplasm of respi- each NI with a coherent group of delivered NAs. The analyses ratory and digestive apparatus, 24.1 [SD, 6.8] NAs per day; showed that the most frequently delivered NAs were related to antineoplastic chemotherapy, 22.8 [SD, 5.3] NAs per day). the administration of medications. Interestingly, a little less than a third of nursing care was delivered on medical order (eg, ad- n Discussion ministration of therapies and interventions related to diagnostic procedures). This strongly suggests that nurses prescribed and The aims of this study were to describe the prevalence of NDs, delivered up to two-thirds of all NAs. These activities were mainly NIs, and NAs in a population of hospitalized oncology patients focused on monitoring the patient’s conditions (eg, monitoring and to investigate the relationships between these data and nutritional/hydration status, cognitive condition, pain, vital signs), patient age and medical diagnoses. A mean of 6 NDs per patient supporting for nutrition, mobilization and elimination, manag- was identified on admission. More than 90% of the 16 NIs ing vascular and respiratory devices, and supporting for wellness planned per patient actually generated 1 or more NAs, for an and relational/psychological issues (eg, information, sleep and average of 25 delivered NAs per patient per hospitalization day. rest, comfort, emotional support in end-of-life care). These re- Neither the number of NDs nor the number of NIs/NAs was sults, obtained with a standardized collection of clinical data, related to patient age. The number of NDs, NIs, and NAs dif- seem to indicate that in the hospital setting the nurse effectively fered significantly among the most frequent medical diagnoses, implements and documents (nursing) diagnostic reasoning and being higher for patients diagnosed as having malignant neo- that the choice of delivering NAs is mainly the result of an plasm of respiratory and intrathoracic organs. To the best of our independent nursing decision. knowledge, this is the first study prospectively documenting the Choi and colleagues35 analyzed the nursing documentation prevalence of NDs on admission, NIs, and NAs in a large of 44 patients with terminal cancer and showed that the fre- population of hospitalized oncology patients. quency of documented NIs was lower than the frequency perceived

Nursing Diagnoses, Interventions, and Activities Cancer NursingA, Vol. 00, No. 0, 2018 n 5

Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. Table 3 & Frequency Distribution of the Most Frequently Provided Nursing Activities Grouped According to the Respective Nursing Interventions Delivered Nursing Interventions (Total No. of Activities) Delivered Nursing Activities n (%) Medications/blood products administration Enteral medication administration 8645 (39.6) (n=21817) Parenteral medication administration 7358 (33.7) Intravenous infusion management 5405 (24.8) Inhalation medication administration 231 (1,1) Blood or blood products administration 133 (0.6) Topical medication administration 45 (0.2) Comfort and pain management (n=15111) Provide comfort measures 6216 (41.1) Provide privacy and ensure confidentiality 5193 (34.4) Comprehensive pain assessment 2398 (15.9) Monitoring change in pain and effects of pain management 1304 (8.6) Vascular devices (n=8987) Central venous catheter management 5933 (66.0) Peripheral venous catheter management 2414 (26.9) Peripheral venous catheter positioning/removal 489 (5.4) PICC insertion/removal 117 (1.3) Other activities 34 (0.4) Feeding/hydration (n=5954) Assessment and monitoring of nutritional/hydration conditions 3415 (57.4) Education about fasting prescription 1061 (17.8) Monitoring nausea and vomiting 542 (9.1) Preparation/administration of parenteral nutrition 537 (9.0) PEG management 127 (2.1) Nasogastric or nasojejunal tube management 87 (1.5) Assist in feeding 82 (1.4) Assessment of swallowing reflex and management of dysphagia 72 (1.2) Preparation/administration of enteral nutrition 25 (0.4) Nasogastric or nasojejunal tube insertion 6 (0.1) Sleep and rest (n=5327) Assessment of sleep/rest length and quality 3024 (56.8) Preparation of patient/environment for sleep/rest 2099 (39.4) Promote adequate rest 105 (2.0) Facilitate sleep or relieve insomnia 67 (1.3) Organize a program of activities to foster sleep/wake rhythms 32 (0.6) Vital signs measurement/monitoring (n=5253) Vital signsa measurement 4902 (93.3) Activating a close monitoring plan 351 (6.7) Urinary elimination (n=5137) Urinary catheter management 1739 (33.9) Urine output measurement/monitoring 1475 (28.7) Urostomy management 552 (10.7) Elimination management through absorbent incontinence 439 (8.5) productsb External (condom) urinary catheter management 273 (5.3) Assist in toileting or using a commode chair 199 (3.9) Assist in using female/men urinal 187 (3.6) External (condom) urinary catheter positioning/removal 143 (2.8) Internal (Foley) urinary catheter positioning/removal 114 (2.2) Urethral/vesical instillation/irrigation 16 (0.3) Biological samples collection (n=4839) Capillary blood draw 2307 (47.7) Blood draw for blood chemistry testsc 1871 (38.7) Urine specimen collection for chemistry tests 314 (6.5) Venous blood draw for blood culture 164 (3.4) Urine specimen collection for culture test 93 (1.9) 24-hour urine specimen collection 62 (1.3) Other specimen collectionsd 28 (0.6) Welcoming/information (n=3060) Information about nursing care or diagnostic tests 2717 (88.8) Reception and orientation of the patient 343 (11.2) (continues)

6 n Cancer NursingA, Vol. 00, No. 0, 2018 Sanson et al

Abbreviations: PEG, percutaneous endoscopic gastrostomy; PICC, peripherally inserted central catheter. aTemperature, arterial blood pressure, pulse, and respiration. bAdult diapers. cVenous (n=1868), arterial (n=3). dSputum (n=19), stool (n=9). by nurses as real in clinical practice. Their finding highlights the purpose of vigilance is the early recognition of patient deteriora- risk of underrecording NIs and stresses the need to document tion to keep him/her safe, so far studies were unable to define what NIs in a standardized and timely manner. It should be noted nurses identify or deliver constitutes surveillance. Although a lower that nurses also carry out activities that are completely unrelated nurse-to-patient ratio and the availability of experienced nurses to patient care in order to obviate organizational weaknesses (eg, were described as relevant factors to ensure safer vigilance, research administrative and clerical tasks). These go unrecorded, and thus showed that it is not the patients’ number but the presence of we could not take them into account. If such activities had been critically ill cancer patients assigned to nurses that impacts on considered, the nurses’ actual workload would have been even nurses’ abilities to be vigilant.42 Unfortunately, the recommenda- greater. These extraneous non-nursing activities increase the nurses’ tion that nursing surveillance/vigilance be included in standard- workload and move them away from bedside nursing care. This ized nursing terminologies43,44 has not yet been transposed, so is one of the strongest predictors of the depersonalization it cannot be included as an NMDS element. dimension inducing burnout syndrome in nurses.36 Another important result of this study concerns the rela- The precise identification of delivered activities facilitates the tionships of the nursing NMDS variables with patients’ age measurement of time spent in their selection and execution, and medical diagnosis. The analysis of planning profiles showed which is a key variable in the quantification, qualification, and that the number of NDs identified on admission was not as- distribution of nursing workloads and can contribute to the sociated with patient age. In addition, it becomes clear that there discussion on the adequacy of nursing resources in different was a significant difference in the number of identified NDs and settings of care.37,38 Measuring nursing workload by tracking the NIs and NAs delivered according to the main medical diag- only the amount of time spent in NAs has been criticized nosis. Patients with malignancy of the respiratory and intrathoracic because it considers only 2 basic dimensions (direct and indirect organs had greater nursing complexity (a higher number of NDs, care) of nursing work.39 Understanding that the complexity of NIs, and NAs). In the study population, patients with this medical nursing in actual situations may influence nursing workload (and diagnosis have the highest impact on the nursing workload. It affect patient outcomes) requires an understanding of many other is also interesting to note that patients belonging to the medical factors.40 These include the nursing complexity of the patient diagnostic category of antineoplastic chemotherapy have the least (determined through the ND), the interventions delivered by a impact on nursing resource use, requiring on average up to 5 NAs nurse, the severity of the patient’s medical condition, and the per day less than other medical diagnostic categories. In addition characteristics of the nursing skill mix (eg, registered nurses and to the obvious clinical implications, the availability of all this nurses assistants) and work environment.18,41 information can be useful in organizations, for example, in distri- The analysis of NMDS can permit examination of most of buting nursing resources appropriately across different shifts. It the above dimensions of nursing work. Nurses’ interactions with would be interesting to see if there is any association between patients can be difficult to define from a diagnostic point of this differential distribution of NAs and hospital costs identified view and hard to quantify in terms of workload. A paradigmatic by the diagnosis-related groups (DRGs). example is patient vigilance (or surveillance), previously described Consistent with the nursing process, each ND should cor- as a critical role for oncology nurses.42 Even if it is clear that the respond with the activation of a personalized care plan with

Nursing Diagnoses, Interventions, and Activities Cancer NursingA, Vol. 00, No. 0, 2018 n 7

Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. the provision of a variable number of NIs and NAs. The large n variability between the actual numbers of NAs delivered for Conclusions the different NIs documented by the present study highlight the importance of documenting not only the planned and The use in daily clinical practice of the PAI nursing information provided NIs but also each related NA because the activities system enabled us to trace a profile of patient needs and related are not interventions.6 Documenting only NIs keeps the in- NAs in a population of oncology inpatients. The ND, which formation at a too general level and does not allow detailed allows documenting and recognizes the value of the intellectual analysis of the specific activities associated with such interven- aspects of nursing, emerged as the crucial determinant for the tions. To better investigate the association between the nurs- identification of specific and personalized interventions to be ing problems identified by the NDs and the amount of NAs implemented. The highest percentage of NAs was delivered on provided, multivariate analysis should be performed taking autonomous nursing prescriptions. into account, for example, the severity of the specific ND, the The use of a structured NMDS, comprising a detailed set complexity of the delivered NA, and other significant variables. of specific NAs, is able to provide the basic information nec- In hospitals, nurses are the only healthcare professionals to essary to describe the nursing contribution to the process of provide care with an uninterrupted continuity of physical pres- patient care. This creates a basis for a new vision of the nurs- ence. For 24hours a day, the nurse is responsible for the correct ing profession by enabling systematic comparison and bench- and safe provision of treatments, for documenting changes in marking with other healthcare professions and international the patient’s condition, and for making critical decisions about data. The extension of the analysis to other populations, both treatments to be activated. For example, although a patient may at hospital and community levels, will open the way to greater be in hospital for surgery, after leaving the operating room nurs- knowledge and enhancement of the professional role of the ing care may be required for days, weeks, or months, depending nurse, highlighting the nursing contribution in pursuing citizens’ on the patient’s recovery path. Patient outcomes depend not only health outcomes. on the proper performance of the surgery, but also on the quality Additional analyses are needed to determine to what extent of nursing care in the postoperative period. This inevitably affects the NAs are dependent on the medical and nursing needs of recovery time and overall outcomes of treatment (eg, LOS, qual- cancer patients. Further studies should evaluate the predictive ity of life, mortality), including the expected medical outcome.45 power of the nursing (NDs, NIs) and medical care (eg, DRGs, Despite the heavy workload and responsibility of nursing, there comorbidity) elements for outcomes such as LOS, mortality, is growing evidence for linking NDs to patient outcomes in and institutionalization at discharge in similar populations. Y different settings and populations.46 48 The impact of nursing on positive or negative outcomes for hospitalized cancer pa- References tients is still essentially neglected. The services provided by a hospital and the related costs and outcomes are documented 1. Wilson A, Whitaker N, Whitford D. Rising to the challenge of health care exclusively through data based on medical diagnoses and pro- reform with entrepreneurial and intrapreneurial nursing initiatives. Online cedures, hospital discharge registers, and the DRGs.8 Unfor- JIssuesNurs. 2012;17(2):5. 2. Nurse N. Charting nursing’s future. Am J Nurs. 2016;116(9):61Y62. tunately, in the present study we could not explore the relationship 3. Maas ML, Delaney C. Nursing process outcome linkage research: issues, among NDs, NIs, NAs, and key outcomes such as hospital LOS current status, and health policy implications. Med Care. 2004;42(2 suppl): and mortality, which would be appropriate to test by per- II40YII48. forming multivariate analyses including important confounders 4. Eurostat. Causes of death statistics. http://ec.europa.eu/eurostat/statistics- such as social and economic conditions, degrees of clinical explained/index.php/Causes_of_death_statistics. Accessed December 1, 2016. 5. Jomar RT, de Souza Bispo VR. The most common nursing diagnosis among severity, number of comorbidities, or medical procedures. The adults/seniors hospitalised with cancer: integrative review. Ecancermedicalscience. incomplete availability of such data did not allow us to perform 2014;8:462. such analyses. 6. Bulechek GM, Butcher HK, McCloskey Dochterman JM, Wagner C. Nursing Interventions Classification (NIC). 6th ed. St. Louis, MO: Elsevier; 2013. 7. American Nurses Association. What nurses do. http://www.nursingworld. Limitations org/EspeciallyForYou/What-is-Nursing/Tools-You-Need/RNsAPNs.html. Accessed January 19, 2017. The findings of this study should be considered in the light of 8. Welton JM, Halloran EJ. Nursing diagnoses, diagnosis-related group, several limitations. The decision to adopt an observational study and hospital outcomes. J Nurs Adm. 2005;35(12):541Y549. design seemed correct, being the most appropriate to provide 9. Westra BL, Latimer GE, Matney SA, et al. A national action plan for information to describe the outcomes achieved in daily clinical sharable and comparable nursing data to support practice and translational 49 research for transforming health care. J Am Med Inform Assoc. 2015;22(3): practice. However, this choice resulted in the enrolment of a 600Y607. population in which confounding factors could not be con- 10. D’Agostino F, Vellone E, Tontini F, Zega M, Alvaro R. Development of trolled a priori. a computerized system using standard nursing language for creation of a Furthermore, we analyzed only standardized clinical data, nursing minimum data set. Prof Inferm. 2012;65(2):103Y109. namely, medical diagnoses, NDs, NIs, and NAs. Other data 11. Wang N, Yu P, Hailey D. The quality of paper-based versus electronic nursing care plan in Australian aged care homes: a documentation audit commonly recorded in free text were not analyzed; these data study. Int J Med Inform. 2015;84(8):561Y569. often contain a large percentage of items containing poten- 12. Welton JM, Dismuke CE. Testing an inpatient nursing intensity billing tially important information.50 model. Policy Polit Nurs Pract. 2008;9(2):103Y111.

8 n Cancer NursingA, Vol. 00, No. 0, 2018 Sanson et al

Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. 13. Herberth L, Gosnell DJ. Nursing diagnosis for oncology nursing prac- 33. Dennison PD, Keeling AW. Clinical support for eliminating the nursing tice. Cancer Nurs. 1987;10(1):41Y51. diagnosis of knowledge deficit. Image J Nurs Sch. 1989;21(3):142Y144. 14. Palese A, Zanini A, Carlevaris E, Morandin A, Carpanelli I, Dante A. 34. Sanson G, Vellone E, Kangasniemi M, Alvaro R, D’Agostino F. Impact of Hidden outpatient oncology clinical nursing minimum data set: findings nursing diagnoses on patient and organisational outcomes: a systematic from an Italian multi-method study. Eur J Oncol Nurs. 2013;17(4):423Y428. literature review [published online ahead of print January 2, 2017]. JClin 15. Milani A, Mauri S, Gandini S, Magon G. Oncology Nursing Minimum Nurs. 2017. Data Set (ONMDS): can we hypothesize a set of prevalent Nursing Sensitive 35. Choi JY, Jang KS. Comparison of cancer nursing interventions recorded Outcomes (NSO) in cancer patients? Ecancermedicalscience. 2013;7:345. in nursing notes with nursing interventions perceived by nurses of an on- 16. Carayon P, Gurses AP. Nursing workload and patient safetyVa human cology unitVpatients with terminal cancer. Taehan Kanho Hakhoe Chi. 2005; factors engineering perspective. In: Hughes RG, ed. Patient Safety and 35(3):441Y450. Quality: An Evidence-Based Handbook for Nurses. Rockville: Rockville 36. Canadas-de la Fuente GA, Vargas C, San Luis C, Garcia I, Canadas GR, Agency for Healthcare Research and Quality; 2008. de la Fuente EI. Risk factors and prevalence of burnout syndrome in the 17. Halloran EJ. Nursing workload, medical diagnosis related groups, and nursing profession. Int J Nurs Stud. 2015;52(1):240Y249. nursing diagnoses. Res Nurs Health. 1985;8(4):421Y433. 37. Possari JF, Gaidzinski RR, Fugulin FM, Lima AF, Kurcgant P. Stan- 18. O’Brien-Pallas L, Irvine D, Peereboom E, Murray M. Measuring nursing dardization of activities in an oncology surgical center according to nursing workload: understanding the variability. Nurs Econ. 1997;15(4):171Y182. intervention classification. Rev Esc Enferm USP. 2013;47(3):600Y606. 19. Spigolon DN, Moro CM. Nursing minimum data set based on ehr 38. Possari JF, Gaidzinski RR, Lima AF, Fugulin FM, Herdman TH. Use of the archetypes approach. Paper presented at the NI 2012: Proceedings of the nursing intervention classification for identifying the workload of a nursing 11th International Congress on Nursing Informatics, June 23-27, 2012, team in a surgical center. Rev Lat Am Enfermagem. 2015;23(5):781Y788. Montreal, Canada. 39. Baumann A, O’Brien-Pallas L, Armstrong-Stassen M, et al. Commitment 20. Werley HH, Ryan P, Zorn CR, Devine E. Why the Nursing Minimum and Care: The Benefits of a Healthy Workplace for Nurses, Their Patients Data Set (NMDS)? In: McCloskey JC, Grace HK, eds. Current Issues in and the System. A Policy Synthesis. Ontario, Canada: Canadian Health Nursing. St Louis, MO: Mosby-Year Book; 1994:113Y122. Services Research Foundation; 2001. 21. Werley HH, Devine EC, Zorn CR, Ryan P, Westra BL. The Nursing 40. Carayon P, Alvarado CJ. Systems Engineering Initiative for Patient Safety. Minimum Data Set: abstraction tool for standardized, comparable, essential Workload and patient safety among critical care nurses. Crit Care Nurs data. Am J Public Health. 1991;81(4):421Y426. Clin North Am. 2007;19(2):121Y129. 22. Laport N, Sermeus W, Vanden Boer G, Van Herck P. Adjusting for nursing 41. de Cordova PB, Lucero RJ, Hyun S, Quinlan P, Price K, Stone PW. care case mix in hospital reimbursement: a review of international practice. Using the nursing interventions classification as a potential measure of Policy Polit Nurs Pract. 2008;9(2):94Y102. nurse workload. J Nurs Care Qual. 2010;25(1):39Y45. 23. Herdman H. Nursing Diagnoses: Definitions and Classification. 2015Y2017. 42. Kooken WC, Haase JE. A big word for something we do all the time: oncology Oxford: Wiley-Blackwell; 2014. nurses lived experience of vigilance. Cancer Nurs. 2014;37(6):E15YE24. 24. National Federation of the Nursing Colleges. Nomenclatore tariffario delle 43. Meyer G, Lavin MA. Vigilance: the essence of nursing. Online J Issues prestazioni di assistenza infermieristica. http://www.ipasvi.it/norme-e-codici/ Nurs. 2005;10(3):8. tariffario.htm. Accessed December 10, 2016. 44. Meyer GA, Lavin MA, Perry AG. Is it time for a new category of nursing 25. Centers for Disease Control and Prevention. International Classification diagnosis? Int J Nurs Terminol Classif. 2007;18(2):45Y50. of Diseases, 9th RevisionVItalian Version. http://www.salute.gov.it/imgs/ 45. Rosa B. What nurses really contribute to health care: part 1 of an edu- C_17_pubblicazioni_2251_allegato.pdf. Accessed November 12, 2016. cation do-over. October, 1, 2017. http://www.huffingtonpost.com/billy- 26. Werley HH, Lang NM. Identification of the Nursing Minimum Data Set. rosa/what-nurses-really-contribute-to-health-care_b_9550926.html. New York: Springer Pub Co; 1988. Accessed January 10, 2017. 27. Herdman H. Nursing Diagnoses: Definitions and Classification. 2012Y2014. 46. Castellan C, Sluga S, Spina E, Sanson G. Nursing diagnoses, outcomes and Oxford: Wiley-Blackwell; 2012. interventions as measures of patient complexity and nursing care require- 28. Zega M, D’Agostino F, Bowles KH, et al. Development and validation ment in intensive care unit. JAdvNurs. 2016;72(6):1273Y1286. of a computerized assessment form to support nursing diagnosis. Int J 47. Morales-Asencio JM, Morilla-Herrera JC, Martin-Santos FJ, et al. The Nurs Knowl. 2014;25(1):22Y29. association between nursing diagnoses, resource utilisation and patient and 29. Italian Health Ministry. Regulation concerning the identification of the caregiver outcomes in a nurse-led home care service: longitudinal study. Int J nurse’s professional profile. http://www.gazzettaufficiale.it/eli/id/1995/01/ Nurs Stud. 2009;46(2):189Y196. 09/095G0001/sg;jsessionid=SHLl8ChHijzbRJ8XmSPO9Q__.ntc-as2- 48. O’Brien-Pallas L, Li XM, Wang S, Meyer RM, Thomson D. Evaluation guri2b. Accessed August 20, 2017. of a patient care delivery model: system outcomes in acute cardiac care. 30. Bradley CJ, Penberthy L, Devers KJ, Holden DJ. Health services research Can J Nurs Res. 2010;42(4):98Y120. and data linkages: issues, methods, and directions for the future. Health 49. Papanikolaou PN, Christidi GD, Ioannidis JP. Comparison of evidence Serv Res. 2010;45(5 pt 2):1468Y1488. on harms of medical interventions in randomized and nonrandomized 31. Hinkle DE, Wiersma W, Jurs SG. Applied Statistics for the Behavioral studies. CMAJ. 2006;174(5):635Y641. Sciences. 5th ed. Boston, MA: Houghton Mifflin; 2003. 50. Berger ML, Curtis MD, Smith G, Harnett J, Abernethy AP. Oppor- 32. Conley VM. Beyond knowledge deficit to a proposal for information- tunities and challenges in leveraging electronic health record data in on- seeking behaviors. Nurs Diagn. 1998;9(4):129Y135. cology. Future Oncol. 2016;12(10):1261Y1274.

Nursing Diagnoses, Interventions, and Activities Cancer NursingA, Vol. 00, No. 0, 2018 n 9