Abstract

1. INTRODUCTION

Respiratory diseases are a major global public health problem due to their severity, frequency, trend and economic impact. An estimated 4million people die prematurely each year worldwide due to respiratory causes (GBD Chronic Respiratory Disease Collaborators, 2020). At the same time, the crude prevalence rate of respiratory diseases has increased by 40% since 1990, reaching 544.9million worldwide in 2017 (GBD Chronic Respiratory Disease Collaborators, 2020), especially due to the contribution of chronic respiratory diseases. In Spain, chronic obstructive pulmonary disease (COPD) is the fourth leading cause of death, with almost 30,000 deaths per year (Soriano et al., 2018). In addition, according to the EPISCAN II study (Soriano et al., 2021), COPD is also one of the most prevalent diseases (12% estimated prevalence). In the near future, accelerated population ageing and increased exposure to risk factors, including particulate pollution, will increase the global burden of chronic respiratory diseases and their consequences (Li et al., 2020); this calls for comprehensive and collaborative actions to combat this epidemic trend (Viegi et al., 2020).

Chronic respiratory diseases also put health systems under growing pressure. Specifically, chronic respiratory diseases are one of the main causes of unscheduled hospitalisations and hospital readmissions (HR) (Press et al., 2018; Ruan et al., 2023), with a strong negative impact on both the patient and the healthcare system (Suissa et al., 2012). Recent studies show that between 5% and 24% of COPD patients are readmitted within 30days of hospital discharge (pooled rate: 11%) (Alqahtani et al., 2020; Press et al., 2021; Ruan et al., 2023). A logical strategy to address the high rates of HR and the resulting healthcare costs is the identification of high‐risk respiratory patients. Advanced age, male sex, low educational level, comorbidity or the presence of dyspnoea and severe COPD at admission have been identified as risk factors for HR (Almagro et al., 2006; Gudmundsson et al., 2005; Price et al., 2006; Wong et al., 2008). Other healthcare‐related variables such as length of hospitalisation, patient's discharge destination (e.g., home, residence, etc.), number of previous hospitalisations or discharge follow‐up are also considered potential predictors of HR for respiratory disease (Baker et al., 2013; Shah et al., 2016). However, the picture is not yet completely clear, and it has been suggested that the inclusion of social and functional variables, in addition to exclusively clinical variables, could improve the ability to predict HR risk (Kansagara et al., 2011).

2. BACKGROUND

During hospital stay, patients receive both multispecialty medical interventions and comprehensive nursing care; however, only disease characteristics and medical treatment are routinely considered for evaluation of the health outcomes. The nursing process represents a problem‐solving structure based on patient needs; thereby, nursing is described as a profession based on establishing professional‐patient relationships (Müller‐Staub et al., 2015). The North American Nursing Diagnosis Association (NANDA) recommends the use of the nursing process to maintain scientific, professional and quality assurance. As part of the nursing process, nursing diagnoses (ND) represent a systematised clinical judgement made by a registered nurse that elicits actual or potential health problems requiring nursing care and which nurses are therefore competent to prevent and/or treat independently (Herdman et al., 2023). The problem can be real or potential and can affect a patient, a healthy person, their families and even the community.

Medical diagnoses and ND are not the same clinical judgements made by different healthcare workers. While medical diagnoses focus on diseases, ND focus on patient's responses to diseases, providing a comprehensive evaluation of the physical, emotional, psychological and social impact of diseases on functioning and well‐being. Moreover, in contrast with the stable nature of medical diagnoses, ND changes as needs and patient responses evolve. Furthermore, according to Chiffi and Zanotti (2015), ‘a medical diagnosis identifies a variation from a norm, while a nursing diagnosis should judge the existence of a potential for enhancing self‐care’. From a nursing perspective, care plans must be aligned with medical treatment to contribute to managing diseases, but above all, they should be at the service of patients' health‐related quality of life. Thereby, ND complements medical diagnoses for providing a comprehensive and effective patient care. Nevertheless, in spite of the recognition of the great potential of ND to predict health outcomes (Fennelly et al., 2021; Sanson et al., 2017), evidence of high certainty on the scientific and clinical utility of ND is still limited.

Bertocchi et al. (2023) conducted a systematic review and meta‐analysis aimed at assessing the impact of NDs on several health care outcomes. According to their findings, NDs have positive effects both on patients' (knowledge, self‐efficacy, quality of life and mortality) and organisational outcomes (length of stay and HR rate), suggesting that the combination of nursing records with medical diagnoses can contribute to a better prediction of a patient's evolution during hospitalisation and post‐discharge. Regarding the impact of ND on HR, evidence came from three randomised controlled trials conducted in China (Liu et al., 2020; Wong & Yeung, 2015; Zhang et al., 2018). They found that a nurse‐led transitional care programme based on the assessment‐intervention‐evaluation Omaha System framework achieved relevant reductions in the HR rate in patients with arthritis, stroke and coronary artery disease (pooled OR=0.46; 95% CI=0.09–0.83). These findings should encourage institutions to implement standardised nursing terminologies in clinical practice (Zhang et al., 2021). Nevertheless, in those clinical settings where NDs are already fully implemented, it would be interesting to ascertain if any of the NDs can also predict HR.

In summary, considering the high rate of HR due to respiratory pathology, the importance of identifying patients at risk of HR, and the potential use of ND to predict patient outcomes, the main aim of the present study was to identify the ND that were associated with HR after hospitalisation for respiratory disease in a tertiary care hospital in Spain. The secondary aim was to study the demographic, care and clinical variables that were associated with a greater probability of HR.

3. METHODS

4. RESULTS

A total of 10,625 hospital admission records for respiratory pathology were identified between January 1, 2016 and March 31, 2019. Subsequently, patients who died during admission, readmissions beyond 30days and episodes of two or more readmissions were eliminated. Of the resulting 5699 episodes, 466 were eliminated because the main diagnosis was not a respiratory disease, despite having it among the secondary diagnoses. Finally, only 4275 episodes had complete data in the electronic health records, corresponding to 3871 patients.

Of the 3781 subjects admitted to the HUCA for respiratory causes during the study period, 499 (13.1%) patients had an episode of HR in the first 30days after hospital discharge. Table 3 shows the baseline characteristics of the study sample according to their status as control (non‐HR) or case (HR). Moreover, statistical tests for differences between control cases are also presented. Compared with the controls, the subjects who had a HR were mostly men, were older, were admitted to the hospital on a scheduled basis, were admitted to the internal medicine service and had longer hospital stays. In addition, they had a greater number of comorbidities. Specifically, among subjects who were readmitted, there was a higher frequency of ND related to infectious and parasitic diseases; neoplasms; blood, blood‐forming and immune systems; endocrine, nutritional and metabolic; nervous system; circulatory system; skin and subcutaneous tissue; musculoskeletal and connective tissue; genitourinary system; congenital malformations and deformities; trauma and poisoning; factors influencing health status and contact with health services.

TABLE 3

Baseline sample characteristics (n=3781).

	Controls (n=3282)	Cases (n=499)	Statistical test	p‐value
Sex, n (%)
Men	1728 (52.7)	315 (63.1)	19,1 a	<0.001
Women	1554 (47.3)	184 (36.9)
Age, mean (sd)	69,9 (17,5)	71.9 (13.8)	−3.68 b	<0.001
Type of admission, n (%)
Emergencies	3153 (96.1)	465 (93.2)	11.4 a	0.010
External consultations	25 (0.8)	5 (1.0)
Programmed	93 (2.8)	28 (5.6)
Other centre	11 (0.3)	1 (0.2)
Admission service, n (%)
Pneumology	2195 (66.9)	290 (58.1)	14.9 a	<0.001
Internal Medicine	1086 (33.1)	208 (41.9)
Duration of admission, mean (sd)	9,45 (8,6)	10.8 (7.5)	−3.32 b	<0.001
Type of discharge, n (%)
Home	3133 (95.5)	475 (95.2)	1.11 a	0.770
Voluntary	15 (0.5)	1 (0.2)
Social‐health centre	82 (2.5)	15 (3.0)
Another hospital	52 (1.6)	8 (1.6)
No. of diseases, mean (sd)	5,73 (4,5)	6.94 (3.9)	−7.11 b	<0.001
Comorbidity, n (%)
Infectious and parasitic	523 (15.9)	117 (23.4)	17.4 a	<0.001
Neoplasms	355 (10.8)	72 (14.4)	5.68 a	0.018
Blood, blood‐forming and immune systems	422 (12.9)	81 (16.2)	4.28 a	0.039
Endocrine, nutritional and metabolic	1066 (32.5)	193 (38.7)	7.49 a	0.006
Mental and behavioural	767 (23.4)	111 (22.2)	0.31 a	0.579
Nervous system	373 (11.4)	84 (16.8)	12.2 a	<0.001
Eye and its adnexa	17 (0.5)	6 (1.2)	3.36 a	0.067
Ear and mastoid process	29 (0.9)	1 (0.2)	2.57 a	0.109
Circulatory system	1605 (48.9)	269 (53.9)	4.34 a	0.037
Digestive system	353 (10.8)	62 (12.4)	1.24 a	0.266
Skin and subcutaneous tissue	80 (2.4)	24 (4.8)	9.11 a	0.003
Musculoskeletal and connective tissue	196 (6.0)	52 (10.4)	13.9 a	<0.001
Genitourinary system	700 (21.3)	140 (28.1)	11.3 a	0.001
Pregnancy, childbirth and puerperium	6 (0.2)	–		–
Congenital malformations and deformities	20 (0.6)	9 (1.8)	8.12 a	0.004
Symptoms, signs and abnormal findings	585 (17.8)	91 (18.2)	0.05 a	0.823
Injuries and poisoning	312 (9.5)	64 (12.8)	5.33 a	0.021
External causes	110 (3.4)	16 (3.2)	0.03 a	0.866
Factors influencing health status	1144 (34.9)	223 (44.7)	18.1 a	<0.001

Open in a separate window

Abbreviation: sd, standard deviation.

^a Chi‐square value (categorical variables).

^b T‐value (continuous variables).

Table 4 depicts the association between ND at hospital discharge and the risk of suffering a HR during the subsequent 30days. According to logistic regression analysis, the presence of the ND ‘knowledge deficit’ (OR: 1.61; 95%CI: 1.13–2.31), ‘impaired skin integrity and risk of ulcer infection’ (OR: 1.45; 95%CI: 1.06–1.97) and ‘activity intolerance associated with fatigue’ (OR: 1.56; 95%CI: 1.21–2.01) were associated with an increased risk of suffering an episode of HR at 30days after hospital discharge, even after adjusting the data analyses for sociodemographic variables, care and comorbidity.

5. DISCUSSION

In this hospital‐based case‐control study conducted on a sample of 3781 subjects admitted to a tertiary level hospital for pathology, over 1 in 10 patients underwent a HR during the 30days following discharge. In other studies, the rates of HR were highly variable, and it is not easy to find an explanation for these differences beyond the inherent differences in the health system or hospital where the studies were performed. A study by Jiang et al. (2018), using data from a sample of 268,084 adults ≥18years old dSiagnosed with COPD between 2009 and 2014, belonging to the Healthcare Cost and Utilization Project Florida State Inpatient Database, observed a HR rate of 7.9%. Also, Li et al. (2022) found an overall HR rate of 15.8% in 2012–17 in Beijing (China) among a sample of patients ≥40years with a primary diagnosis of COPD. Goto et al. (2017) retrospectively analysed data from 845,465 patients hospitalised for COPD between 2006 and 2012 in eight US states and found a HR rate of 19.7%.

Our study also identified several sociodemographic and clinical variables that were associated with HR. In line with our findings, age and male sex have been consistently associated with a higher likelihood of HR (Chow et al., 2023; Lau et al., 2017). Both characteristics are indicative of a worse prognosis in respiratory patients. Furthermore, according to Li et al. (2022), differences by sex in the rate of HR may be increasing. Compared to men, women place greater emphasis on self‐care and use a wider variety of self‐care behaviours (Caruso et al., 2020; Grzywacz et al., 2012). In contrast, men tend to delegate self‐care (Grzywacz et al., 2012). Therefore, men may be more sensitive to the lack of family and social support, which leads to poorer continuity of care at home after hospital discharge, especially when this is early. Regarding variables related to health care, a higher proportion of those undergoing HR were admitted to the internal medicine unit and had longer hospital stays. In Spain, many patients with respiratory diseases are admitted to Internal Medicine, and they are usually older and have more comorbidities, both of which are risk factors for HR (Recio Iglesias et al., 2020). Length of hospital stay has been identified as an independent risk factor for HR in numerous studies (Alqahtani et al., 2021; Ruan et al., 2023). Nonetheless, Rinne et al. (2017) suggest that the length of hospital stay should be interpreted with caution since its association with the probability of HR could be confounded by the patient's baseline clinical situation, understood as the presence of a greater number of prior diseases. In our study, the difference in days of stay between cases and controls was not as large as the differences found in the comorbidities of both groups. Numerous studies have found that the presence of cardiovascular disease, diabetes, cancer or kidney disease is associated with an increased risk of HR in patients with COPD (Chakrabarti et al., 2021; Quintana et al., 2022).

Patient complexity is the result of different needs, not only linked to their medical condition but also to other functional or psychological aspects (Schaink et al., 2012). Consequently, the use of ND is extremely useful because it provides a global assessment of the patient's responses to certain health conditions or vital processes, complementary to clinical diagnoses (Pascoal et al., 2022). Our results have shown that the ND groups ‘knowledge deficit’, activity intolerance due to fatigue', and ‘impaired skin integrity and risk for ulcer infection’ were associated with an increased likelihood of HR. To date, this is the first study to explore the association between nursing diagnoses and the risk of HR in patients with respiratory pathology, and therefore comparison with other research is not straightforward. Nevertheless, our results are clinically consistent, as the identified ND draws the profile of a respiratory patient with severe disease, bedridden and a low educational level. In general, loss of independence and frailty syndrome are independent risk factors for HR (Alqahtani et al., 2021; Fernández‐García et al., 2020). Moreover, a panel of experts recruited by da Silva et al. agreed that a main defining characteristic of the ND ‘knowledge deficit’ was an increase in HR, as they were judged to be closely related (da Silva et al., 2023). Similarly, according to a study examining calls to an advice line of COPD patients within 30days post‐discharge, doubts about medications and aftercare instructions were the most common issues (Stella et al., 2014), which clearly pointed to the underlying effect of poor knowledge.

A few previous studies successfully used ND to study healthcare‐related variables, including mortality (Sanson et al., 2019). D'Agostino et al. (2019) determined using a purely quantitative approach that the number of ND is a sensitive indicator for predicting the length of hospital stay. However, the usefulness of describing each ND has been less studied. A study conducted on 300 nursing records of patients ≥70years old admitted to orthopaedic wards (Paans et al., 2016) observed that the ND ‘impaired tissue perfusion’, ‘pressure ulcer’ and ‘fluid volume deficit’ were associated with a longer length of hospital stay. In short, the results of our study underline the importance of the standardised use of nursing language, as it can contribute to explaining patient complexity and predicting health events in a complementary way to other sociodemographic and clinical variables (Welton & Halloran, 2005).

6. CONCLUSIONS

In summary, 13.1% of patients hospitalised for respiratory pathology in a tertiary public hospital in Spain suffered an episode of HR within 30days of discharge. Patients who were re‐admitted to the hospital were frequently men of advanced age, accessing the hospital on a scheduled basis, treated in the internal medicine unit, with long hospital stays and the presence of comorbidities. Knowledge deficit, activity intolerance due to fatigue and impaired skin integrity/risk for ulcer infection were independently associated with an increased likelihood of HR. Consideration of ND may be useful for predicting HR. Nonetheless, the usefulness of ND for predicting other relevant health events warrants further study.

AUTHOR CONTRIBUTIONS

All authors should have made substantial contributions to all of the following: (1) the conception and design of the study, or acquisition of data, or analysis and interpretation of data; (2) drafting the article or revising it critically for important intellectual content; (3) final approval of the version to be submitted. PS‐G, AS‐E and AL conceived and designed the study. LA‐F, IP‐R and FJJ‐D conducted the statistical analyses. PS‐G, AS‐E and AL drafted the manuscript. AL is a guarantor.

FUNDING INFORMATION

This research has been funded by a grant from the Official College of Nursing of Asturias (Spain). The funding agency had no role in the study design, data analysis, interpretation of results, writing of the report and in the decision to submit the article for publication.

CONFLICT OF INTEREST STATEMENT

None.

ETHICS STATEMENT

The study was approved by the Hospital Board of Directors and the Clinical Research Ethics Committee of Asturias, Spain (ref. 123/19).

Supporting information

ACKNOWLEDGEMENTS

Notes

Suárez‐González P., Suárez‐Elosegui A., Arias‐Fernández L., Pérez‐Regueiro I., Jimeno‐Demuth F. J., & Lana A. (2024). Nursing diagnoses and hospital readmission of patients with respiratory diseases: Findings from a case–control study. Nursing Open, 11, e2182. 10.1002/nop2.2182 [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]

REFERENCES