The Role of Nurse Social Support in Hospital Readmission in Patients With Heart Failure: A Cross‐Sectional Study

ABSTRACT

Background and Aims

Hospitalizations and hospital readmissions account for 75%–80% of the economic burden related to heart failure (HF). This study aims to determine the relationship between perceived social support from nurses and hospital readmission rates in HF patients and to identify potential factors associated with hospital readmission, including sociodemographic and clinical variables, perceived social support, and country.

Method

We recruited 380 HF patients from two countries (Iceland 92 and Iran 288) from October 2018 to May 2020. We used the social support scale of the Family Functioning, Health, and Social Support (FAFHES) questionnaire for data collection. The hospital readmission rates were collected from patients' hospital records. We performed a general linear model (GLM) to explore the significant factors associated with readmission.

Results

A total of 125 (33%) patients were readmitted within 6 months. Among them, 17% were readmitted once, 15% were readmitted twice, and 1% were readmitted three times. The strongest factor associated with the hospital readmission rate was perceived social support. Other significant factors included unemployment, age, and country of residence.

Conclusion

This study underlines the important role of perceived social support in hospital readmission among HF patients. Our findings may be useful for better tailoring social support for HF patients to reduce hospital readmission. Moreover, it is recommended to investigate the effects of system‐related factors, such as the strategy of healthcare systems in managing HF, on hospital readmissions.

Summary

• What is already known about this topic?

  • ∘
    Hospital readmission is associated with increased mortality rates and poor quality of life in patients with heart failure (HF).
  • ∘
    There is very limited evidence on the association between social support provided by nurses and hospital readmission in patients with HF.
• What this paper adds?

  • ∘
    Higher levels of perceived social support were associated with a lower hospital readmission rate.
  • ∘
    More research is needed to better understand the mechanism of the role of social support on hospital readmission, particularly by identifying possible mediators.
• The implications of this paper:

  • ∘
    This study is the first to underline the significant role of perceived social support from nurses on hospital readmission in patients with HF.
  • ∘
    This study's findings may be useful to better tailor social support for patients with HF to reduce hospital readmission.

1. Introduction

Heart failure (HF) is associated with frequent hospitalization [1]. Although there have been advances in diagnosis and treatment, HF is still a big clinical and public health issue due to the high rate of morbidity, mortality, and economic burden on the healthcare system [2, 3, 4]. Hospitalizations and hospital readmissions have accounted for up to 75%–80% of the economic burden related to HF in some countries [2]. Hospital readmission is also associated with increased mortality rates and poor quality of life in patients with HF [5]. HF includes 2% of all admissions to the hospital and is the main cause of hospital admission in adults over 65 years [6, 7]. In the next 25 years, an increase is expected in hospital admissions for HF by 50% [6]. Moreover, hospital readmission in HF is still high, approximately 25% in 30 days and 30% in 60 and 90 days after discharge [6, 8]. The annual and 5‐year readmission rates after HF diagnosis are approximately 50% and 80%, respectively [6, 8, 9]. Hence, it is vital to identify the associated factors of hospital readmission in patients with HF to improve patient outcomes and reduce the economic burden on the healthcare system. Several risk factors for hospital readmission in patients with HF have been identified. For instance, sociodemographic and clinical characteristics (e.g., older age and severity of illness) [10], behavioral factors (e.g., noncompliance with medications and diet and inadequate self‐care) [11, 12], and psychosocial factors (e.g., social isolation) [4] are associated with a high rate of hospital readmission in HF.

Most studies of hospital readmission in HF have focused on the clinical and sociodemographic characteristics of the patients and very few studies have investigated the role of social support [1, 3, 8]. Some studies have shown that perceived social support from family, friends, or peers is correlated with a low rate of hospitalization in HF [13, 14]. On the other hand, social isolation is indicated to be correlated with a high hospital readmission rate in patients with HF [15]. Social support from healthcare professionals, particularly from nurses, has rarely been studied. One study indicated that patients with HF who perceived post‐discharge professional support from nurses in the form of home visits had a lower likelihood of hospital readmission [10]. Hence, it is important to identify psychosocial factors, such as social support, particularly social support provided by healthcare professionals, to design interventions to reduce hospital readmission in patients with HF.

The concept of social support is complex, and there is no consistency in the definition of social support in the literature [16, 17]. A broad definition of social support is an interactive process of exchanging information and/or resources between two or more persons, which leads to emotional effects or positive behaviors [16, 18, 19]. In the most common approach in health sciences, social support is classified into social integration or isolation, social networks, and functional content. Social integration or isolation relates to the quantity of social relationships or frequency of interaction. Social networks represent the structure of social relationships and can be either a dyadic relationship between two persons or a network of more than two persons. Functional content relates to the quality of social relationships and includes emotional support (e.g., empathy, physical affection, caring, or listening), instrumental support (e.g., tangible assistance or financial help), informational support (e.g., giving advice on how to cope with a stressful situation), and appraisal support (e.g., giving feedback) [13, 16, 20]. Several studies have indicated the positive effect of social support on health‐related outcomes in patients with HF [13, 14, 21, 22]. This study adopts the family system perspective, which views the family as an interconnected unit that interacts with larger systems, such as healthcare. Families rely on internal and external resources to maintain balance, especially during stress, such as managing chronic illnesses like HF, which can destabilize family functioning. When internal resources are insufficient, external support, including social support from nurses, becomes crucial to restore stability and enhance family functioning.

There is very limited evidence on the association between perceived social support from nurses and hospital readmission in patients with HF. There is also a lack of evidence on the role of social support in hospital readmission in relation to other factors, such as clinical and sociodemographic characteristics, particularly in different clinical settings and countries. Therefore, we aimed to determine the association between perceived social support and hospital readmission rates in patients with HF and to identify potential factors associated with hospital readmission, including sociodemographic and clinical variables, perceived social support, and country.

2. Methods

2.1. Study Design, Settings, and Participants

We conducted a multicenter cross‐sectional study at a HF clinic in Iran and a heart outpatient clinic in Iceland. Consecutive patients diagnosed with HF were enrolled. We included patients who were in any stage of the New York Heart Association Classification (NYHA) from I to IV and had a left ventricular ejection fraction (LVEF) ≤ 40%. We excluded patients in the terminal stage of other diseases.

2.2. Instrument

Social support was measured using the social support scale from the Family Functioning, Health, and Social Support (FAFHES) questionnaire [23]. The social support scale measures the perceived social support from nurses by patients and is based on an interactive relationship between nurse and patient to promote health outcomes by improving the planning and implementation of patient care [16]. The scale contains 20 items with three dimensions: affect (emotional support), affirmation (informational support), and concrete aid (instrumental support) [23]. The scale has shown good reliability in previous studies measured by Cronbach's α coefficients ranging from 0.92 to 0.95 [23, 24, 25]. In the present study, Cronbach's α coefficients for the Icelandic and Persian [26] versions of the social support scale were 0.97 and 0.94, respectively. The questionnaire also contained patients' self‐reported demographic characteristics (e.g., sex, age, marital status, work status, and education). The patients' clinical characteristics (e.g., NYHA classification, duration of HF, and LVEF) and the hospital readmission rates during the past 6 months were collected from the hospital records.

2.3. Data Collection

The procedure of data collection has been described previously [27]. In Iran, data were collected from October 2018 to December 2019. A project research nurse enrolled patients in the study after they gave informed consent to participate in the study. Then, the research nurse distributed questionnaires during clinical visits in the HF patient clinic, collected the completed questionnaires during the same clinical visits, and checked for missing data. In the case of missing data, patients were asked to complete the missing data at the same time. So, the final sample did not contain any missing data.

Data were collected from October 2019 to May 2020 in Iceland. A project research nurse prepared a list of HF patients who were admitted to a heart outpatient clinic and met the inclusion criteria. Subsequently, the research nurse contacted patients by telephone and asked them if they were interested in participating in the study. We then mailed questionnaires to the postal addresses of the patients who agreed to participate. Patients returned the completed questionnaires via prepaid mail to the research team in Iceland. The final dataset had missing data of 3.5%, which were replaced using an expectation‐maximization algorithm.

2.4. Ethical Considerations

The study followed the principles of the Declaration of Helsinki. The study received ethical approval in Iceland from the National Bioethics Committee in Reykjavik (No: VSN‐19‐078) and in Iran from the Isfahan University of Medical Sciences in Isfahan (No: 297009). Informed consent was obtained from all participants before their involvement in the study.

2.5. Data Analysis

Data were analyzed in IBM SPSS Statistics (version 24), with a significance level of two‐sided p value < 0.05. One‐way analysis of variance (ANOVA) and χ ² test were conducted to compare continuous and categorical variables, respectively. To investigate the relationship of social support with readmission and to identify the potential factors associated with readmission, we first examined the bivariate association of social support, sociodemographic, and clinical variables with 6‐month hospital readmission in the total sample. We then performed a univariate general linear model (GLM) of the variables that had a significant bivariate association with readmission to explore the significant factors associated with readmission. The statistical power analysis for linear multiple regression was performed in the G*Power 3.1.9.7 program. With α = 0.05, and a sample size of 380, there was 95% power to detect a small effect (f ² = 0.05).

3. Results

3.1. Patients' Characteristics

In this study, we analyzed data from 380 patients with HF from Iran (288) and Iceland (92). Table 1 shows the summary and comparison of patients' characteristics between the two countries. Most patients were male (77%), married (80%), and retired (51%); had a high school education (47%); and were in NYHA classes I and II (59%). Patients had a mean age of 63 years, mean duration of HF of 22 months, and mean LVEF of 30%.

Table 1.

Patients' characteristics.

Characteristics	Total sample (n = 380)	Iran (n = 288)	Iceland (n = 92)	p value
Sex				0.572b
Male	293 (77.1)	224 (77.8)	69 (75)
Female	87 (22.9)	64 (22.2)	23 (25)
Marital status				0.04b
Married	306 (80.5)	242 (84)	64 (69.6)
Unmarried	74 (19.5)	46 (16)	28 (30.4)
Education				0.057b
Elementary school	159 (41.8)	128 (44.4)	31 (33.7)
High school	178 (46.8)	133 (46.2)	45 (48.9)
Academic	43 (11.3)	27 (9.4)	16 (17.4)
Work status				< 0.001b
Employed	83 (21.8)	62 (21.5)	21 (22.8)
Unemployed	104 (27.4)	103 (35.8)	1 (1.1)
Retired	193 (50.8)	123 (42.7)	70 (76.1)
NYHA classification				0.181b
NYHA I and II	223 (58.7)	163 (56.6)	60 (65.2)
NYHA III and IV	157 (41.3)	125 (43.4)	32 (34.8)
Age (years)	63 ± 14.4	60.7 ± 14	70.2 ± 12.9	< 0.001c
Duration of HF (months)	22.3 ± 21.4	18.6 ± 17.2	33.9 ± 28.3	< 0.001c
	(14 [6–25])a	(12 [6–24])a	(26 [13.5–42])a
LVEF (%)	29.6 ± 7.8	29.5 ± 8	29.7 ± 7.5	0.849c
Social support	84.9 ± 18.9	86.3 ± 17.8	80.8 ± 21.6	0.017c
Previous 6‐month readmission				< 0.001b
No	255 (67.1)	179 (62.2)	76 (82.6)
Yes	125 (32.9)	109 (37.8)	16 (17.4)
1 time	63 (16.6)	51 (17.7)	12 (13)
2 times	57 (15)	53 (18.4)	4 (4.4)
3 times	5 (1.3)	5 (1.7)	0.0

Note: Data are presented as mean ± standard deviation or number (%). Abbreviations: NYHA = New York Heart Association Classification, LVEF = left ventricle ejection fraction.

^a Median and interquartile range (25%–75%).

^b The results of χ ² test.

^c The results of independent t‐test.

3.2. Readmission and Perceived Social Support

The mean hospital readmission during the previous 6 months was 0.51 ± 0.8 (ranging from 1 to 3) in the total sample. That is, 125 patients (33%) were readmitted during the previous 6 months. Of whom, 63 (17%) were readmitted once, 57 (15%) were readmitted twice, and 5 (1%) were readmitted three times. χ ² test indicated a significantly higher number of readmissions in Iranian patients (38%) than in Icelandic (17%) patients (p < 0.001, Table 1). In addition, the independent t‐test indicated a significantly higher mean score of readmissions in Iranian patients (0.6 ± 0.7) than in Icelandic (0.21 ± 0.5) patients (p < 0.001, Table 2).

Table 2.

Bivariate association between variables and the mean score of readmissions within 6 months in total sample (n = 380).

Variables	Readmission M ± SD or β	p value
Country		< 0.0001a
Iran	0.6 ± 0.7
Iceland	0.21 ± 0.5
Sex		0.883a
Male	0.51 ± 0.8
Female	0.49 ± 0.7
Marital status		0.361a
Married	0.49 ± 0.7
Unmarried	0.58 ± 0.8
Education		0.695b
Elementary school	0.47 ± 0.7
High school	0.54 ± 0.8
Academic	0.51 ± 0.7
Work status		0.012b
Employed	0.45 ± 0.7
Unemployed	0.7 ± 0.9
Retired	0.42 ± 0.7
NYHA classification		0.457a
NYHA I and II	0.48 ± 0.8
NYHA III and IV	0.54 ± 0.7
Duration of HF (months)	β = −0.067	0.191c
LVEF (%)	β = –0.081	0.116c
Age (years)	β = –0.187	< 0.0001c
Social support	β = –0.284	< 0.0001c

Abbreviations: M = mean, SD = standard deviation, β = standardized regression coefficient.

^a The results of independent t‐test.

^b The results of one‐way ANOVA.

^c The results of linear regression.

The total scores for perceived social support varied significantly between the two countries, with a significantly higher level of perceived social support in Iranian patients than in Icelandic patients (p = 0.017, Table 1). The bivariate association of the mean scores of perceived social support with the mean scores of readmission rates in the total sample indicated a significantly negative association of perceived social support with readmission (β = –0.284, p < 0.001, Table 2). That is, a one‐score increase in perceived social support decreased the mean score of readmissions by 0.284.

3.3. GLM of Factors Associated With Readmission

Variables that showed a significant bivariate association with readmission were included in the GLM to assess the significant associating factors of readmission (Table 2). Four variables were entered into the model: perceived social support, country, work status, and age. The result of the GLM indicated that four variables, including country (p = 0.005), perceived social support (p < 0.001), work status (p = 0.026), and age (p < 0.001) were significantly associated with the mean score of readmissions. The mean score of readmissions in Iranian patients compared to Icelandic patients was increased by 0.274 points (p = 0.005; Table 3). A higher level of perceived social support was a negative factor associated with readmission (β = –0.014, p < 0.001). Moreover, being unemployed increased the readmission score by a mean of 0.21 (p = 0.026; Table 3). The overall goodness of fit for the GLM was R ² = 0.184, indicating that the variables included in the model explained the variations in readmission scores by 18.4%. The perceived social support accounted for the highest variation in the readmission scores with 11.9% (η_p ² = 0.119; Table 3).

Table 3.

Parameters' estimates for readmission in a general linear model (n = 380).

Variable		β	SE	t	p value	95% CI		Partial η 2
Country	Iran	0.274	0.097	2.838	0.005	0.084	0.464	0.021
	Iceland—reference	#
Work status	Employed	–0.064	0.101	–0.634	0.527	–0.262	0.132	0.019
	Unemployed	0.210	0.094	2.236	0.026	0.025	0.395
	Retired—reference	#
Social support		–0.014	0.002	–7.097	< 0.001	–0.018	–0.01	0.119
Age		–0.011	0.003	–3.973	< 0.001	–0.017	–0.006	0.04

Note: R ² = 0.184.

Abbreviations: β = unstandardized regression coefficients, SE = standard error, CI = confidence interval, NYHA = New York Heart Association Classification.

^# This parameter is set to zero.

4. Discussion

In this international study, approximately one‐third of patients with HF were readmitted in the previous 6 months. Hospital readmission rates in previous studies vary significantly. The 6‐month readmission rates of 23.6% in China [5] and 27% in Japan [10] have been reported in patients with HF, which are close to our readmission rate. In contrast, a 12.2% readmission rate within 6 months was reported in Nigerian patients with HF [28] A high rate of 61.8% hospital readmission was reported within 6 months in Iranian patients with HF [29]. One reason for differences in hospital readmission rates may be attributed to the heterogeneity of samples' sociodemographic and clinical characteristics across studies. For instance, in the Nigerian study [28], patients were younger, and most of them had nonischemic HF. Previous studies have reported that older age and HF with ischemic etiology are associated with a higher rate of hospital readmission [1, 30]. In the Iranian study [29], patients were in NYHA classes II and III. A higher NYHA class is associated with a higher rate of hospital readmission [1].

In our study, demographic variables (age and work status), perceived social support from nurses, and country had significant bivariate associations with readmission. In addition, these variables were significantly associated with hospital readmission in the GLM. In line with our results, a study has reported that patients with HF who have no occupation were 2.6 times more likely to be readmitted. This might be related to the social isolation that these patients perceived as a result of decreased social activities [10]. In our study, older age was associated with lower hospital readmission rates. Similarly, a previous study investigating 30‐day all‐cause HF readmission reported that patients at a younger age are more likely to be readmitted [31]. In contrast, another study reported that older age was an independent factor associated with higher 30‐day readmission rates [1]. However, the role of age in relation to other factors, such as country and social support in hospital readmission, particularly in the long term, was not investigated. As a result, conclusions about the role of age on hospital readmission could not be drawn.

Country was a significant factor in this study. That is, Iranian patients had a significantly higher rate of hospital readmission compared to Icelandic patients. This underlines the importance of system‐related factors in hospital readmission in HF, such as access to follow‐up care and appropriate transitions of care [8]. In Iceland, patients with HF have access to an outpatient clinic where they can receive follow‐up care preventing unnecessary hospitalization. However, in Iran, the HF outpatient clinic is not a part of routine care, and patients must be referred to the hospital for follow‐up and observation.

Furthermore, perceived social support in a bivariate association and after adjustment for demographic/clinical variables and country emerged as the most influential factor associated with hospital readmission. Patients who perceived a high level of social support from nurses had lower readmission rates. There is limited evidence on the association of perceived social support, particularly from nurses/healthcare professionals, with hospital readmission among patients with HF. One study indicated a significant bivariate association between professional support (mostly from nurses) and readmission in HF, but professional support was not a statistically significant predictor when adjusted for clinical variables in the multiple logistic regression model [10]. Some studies that investigated perceived social support from family and friends indicated an increased risk of hospital readmission in patients with HF who perceived a low level of social support [4, 32]. Most studies have investigated social isolation with hospital readmission in HF. They found that a high level of perceived social isolation was associated with an increased risk of hospital readmission, which supports our findings. For instance, a meta‐analysis that included 13 studies reported that patients with perceived social isolation had 1.55 times increased risk of hospital readmission compared with those who did not perceive social isolation [4].

The exact mechanisms underlying the association between social support and hospital readmission in HF are unclear. Several studies have reported better self‐care behaviors and self‐management skills in patients with HF who have a good social network and perceive a high level of social support [4, 18]. Two possible pathways have been postulated in previous research regarding the mechanisms underlying the association between social support and health‐related outcomes. One possible pathway is behavioral processes that depict social support's contribution to health‐promoting behavior, such as medication and diet adherence, exercise, and not smoking [13, 33]. This can be achieved by providing health‐related informational or tangible support [33]. For instance, the health‐promoting effect of social support has been addressed previously in patients with HF that perceived social support from nurses has an indirect effect (via family functioning) on family health [22]. Family health plays a significant role in HF hospital readmission [26]. Therefore, perceived social support may indirectly, through family health or family functioning, influence hospital readmission. Another possible pathway is psychological processes that are related to emotions and moods that social support, especially emotional support, may contribute to the buffering effect of social support, which reduces the detrimental effect of stress and gives a feeling of control over overwhelming situations [33]. These possible mechanisms may explain the influence of perceived social support from nurses on self‐care behaviors and self‐management skills in patients with HF, which may ultimately reduce hospital readmission. This mediation relationship has been addressed previously that a high level of perceived social support was indirectly associated, via increasing medication adherence, with reduced hospital readmission in HF [14]. Future research is warranted to better understand the mechanism of the role of social support on hospital readmission, particularly by identifying other possible mediators.

Our findings on the importance of perceived social support from nurses have some implications for clinical practice. Patients with HF face social limitations, emotional distress, and physical restrictions in their daily lives [34], and effective support from nurses and healthcare professionals can help them adjust to these challenges and limitations properly [13], which may indirectly prevent unnecessary hospitalizations [35]. For instance, nurses and healthcare professionals can provide emotional support when providing care to patients by active listening, expressing empathy, and giving them time to express their emotions and problems [20]. Nurses and healthcare professionals can also educate patients' family members about the importance of their emotional support and contributions to the patient's care, by teaching them basic skills such as active listening and expression of empathy and by providing health‐related information to families to facilitate adjustment to this new situation alongside patients. Nurses and healthcare professionals may also provide informational support using technology‐based applications or online support as alternatives to face‐to‐face support in clinical settings.

4.1. Limitations

The interpretation of our results should be considered along with some limitations. A patient is likely to see several nurses during hospitalization. Therefore, the measurement of perceived social support might be related to interactions with different nurses, of whom some might be considered supportive and some less supportive. Moreover, the data of 15 (16.3%) patients in Iceland were collected during the first lockdown of the COVID‐19 pandemic, which might reflect a selection bias and have impacted the results. Additionally, this study assessed the overall role of social support without differentiating between various types of social support and their specific impacts on readmission rates. Future research is warranted to explore the influence of distinct types of social support, with particular attention to the functional content of social support on readmission rates. Finally, the readmission rates in our study involved all‐cause readmissions, including cardiac/noncardiac and planned/unplanned readmissions. Detailed data on the causes of readmission were not available for further analyses, but all‐cause readmission can be considered to better represent readmission causes and the complexity of HF management.

5. Conclusion

This international study underlines the important role of perceived social support in hospital readmission in patients with HF. Future research is warranted to better understand the role of social support in hospital readmission, particularly by identifying other possible mediators. Moreover, system‐related factors such as the strategy of healthcare systems in managing HF strongly impact hospital readmissions. Our results may be useful for better tailoring social support for patients with HF to reduce hospital readmission.

Author Contributions

Mahdi Shamali: conceptualization, methodology, data curation, formal analysis, funding acquisition, writing – original draft, writing – review and editing, project administration, investigation. Birte Østergaard: conceptualization, methodology, writing – review and editing, supervision. Erla Kolbrún Svavarsdóttir: conceptualization, methodology, writing – review and editing, supervision, resources. Mohsen Shahriari: conceptualization, methodology, resources, writing – review and editing. Hanne Konradsen: writing – review and editing, conceptualization, methodology, supervision.

Ethics Statement

The study followed the principles of the Declaration of Helsinki. The study received ethical approval in Iceland from the National Bioethics Committee in Reykjavik (No: VSN‐19‐078) and in Iran from the Isfahan University of Medical Sciences in Isfahan (No: 297009). Informed consent was obtained from all participants before their involvement in the study.

Conflicts of Interest

The authors declare no conflicts of interest.

1. Transparency Statement

The lead author Mahdi Shamali affirms that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.

Data Availability Statement

The data that support the findings of this study are available on request from the first author. The data are not publicly available due to privacy or ethical restrictions.