Anxiety subtypes in rural ischaemic stroke survivors: A latent profile analysis

Abstract

Aim

To determine the potential profile classes of anxiety reported by ischaemic stroke survivors in rural China, and to explore the characteristics of patients having different types of post‐stroke anxiety.

Design

A cross‐sectional survey.

Methods

A cross‐sectional survey was conducted by using convenience sampling to collect data from 661 ischaemic stroke survivors in rural Anyang city, Henan Province, China, from July 2021 to September 2021. The parameters included in the study were the socio‐demographic characteristics, self‐rating anxiety scale (SAS), self‐rating depression scale (SDS) and the Barthel index of daily activity ability. Potential profile analysis was done to recognize subgroups of post‐stroke anxiety. The Chi‐square test was performed to explore the characteristics of individuals with different types of post‐stroke anxiety.

Results

The model fitting indices of stroke survivors supported three classes of anxiety models which were as follows: (a) Class 1, low‐level, stable group (65.3%, N = 431); (b) Class 2, moderate‐level, unstable group (17.9%, N = 118) and (c) Class 3, high‐level, stable group (16.9%, N = 112). The risk factors associated with post‐stroke anxiety were female patients, lower levels of education, living alone, lower monthly household income, other chronic diseases, impaired daily activity ability and depression.

Conclusions

This study identified three different subgroups of post‐ischaemic stroke anxiety and their characteristics in patients in rural China.

Impact

This study has significance in providing evidence for the development of targeted intervention measures to reduce negative emotions in different subgroups of post‐stroke anxiety patients.

Patient or Public Contribution

In this study, the researchers arranged the time of questionnaire collection with the village committee in advance, gathered the patients to the village committee for face‐to‐face questionnaire survey and collected the household data of the patients with mobility difficulties.

1. INTRODUCTION

Stroke is the most common cerebrovascular disease (Benjamin et al., 2017), it is the second leading cause of death in the world, the third leading cause of disability in the world (Li et al., 2020) and the leading cause of death in China (Wu et al., 2019). Previous studies have shown that China will have more than 30 million stroke patients by 2030. Moreover, 75% of stroke survivors have different degrees of disability (Chen et al., 2017). Stroke is divided into two types: haemorrhagic stroke and ischaemic stroke. Ischaemic stroke refers to localized brain tissue ischaemia, necrosis or softening caused by cerebral blood circulation disorder, ischaemia, and hypoxia. It shows an elevation in epidemiological indices, like the incidence, recurrence rate, disability rate, fatality rate and economic burden (Goldstein et al., 2011). The risk factors for stroke are smoking, drinking, hypertension, diabetes, low physical activity, etc. (Wang et al., 2017). Studies have shown that the incidence of stroke is higher in rural areas (298 cases per 1,00,000 person‐years) than in urban areas (204 cases per 1,00,000 person‐years) in China due to below‐average economic conditions and inadequate medical facilities. Hence, the rehabilitation of stroke survivors in rural areas is lower than in urban areas (Wu et al., 2019).

2. BACKGROUND

The physical disorders associated with stroke could cause mental health disorders, like anxiety and depression. These psychological disorders are more pronounced and harmful in stroke survivors as compared to patients with other physical disorders. The incidence of post‐stroke anxiety (PSA) ranges from 9.4% to 36.7% in the world (Almhdawi et al., 2021). Moreover, studies have shown that during the COVID‐19 pandemic, due to inconvenient and inadequate medical treatment, stroke patients have been experiencing an increase in anxiety symptoms (Ahmed et al., 2020). The factors influencing post‐stroke anxiety were being a female patient, non‐marital status, grip strength, sleep apnoea, short sleep duration, psychological disorder and functional ability. Post‐stroke anxiety (PSA) and post‐stroke depression (PSD) are common comorbidities of stroke. Hence, a significant positive correlation between PSD and PSA is present (Almhdawi et al., 2021; Sanner Beauchamp et al., 2020). Functional ability is defined as the ability of individuals to happily participate in social activities as per their wishes and preferences (Beard et al., 2016). Basic ADL is a commonly used criterion to evaluate functional ability, which includes basic self‐care activities, like bathing, dressing, eating, etc. A disturbance in these activities is closely associated with the anxiety of stroke survivors (Lee et al., 2019). Therefore, it is suggested that urban stroke survivors have a certain degree of anxiety, which is also affected by socio‐demographic, physical and psychological barriers. However, the level of anxiety among stroke survivors in rural China remains unknown.

In most previous studies, the severity of anxiety in stroke survivors was evaluated based on relevant assessment tools and divided into different grades according to the total score. However, such approaches could not accurately distinguish the different response characteristics of each item, which made it easier to ignore a few characteristic groups. Latent profile analysis (LPA) is an individual‐centred statistical analysis method that explains the relationship between explicit indicators through potential category variables to evaluate the relationship between explicit variables and maintain the local independence of explicit indicators (Chong et al., 2021). Potential profile analysis (LPA) can study the heterogeneity of different categories of anxiety in stroke survivors through the specific characteristics of different items. Meanwhile, previous studies indicated that PSA is associated with socio‐demography, PSD and basic ADL, bringing a series of physical and psychological damage to stroke survivors (Almhdawi et al., 2021; Lee et al., 2019; Sanner Beauchamp et al., 2020). This indicates that it is necessary to clarify the potential anxiety classes of rural stroke survivors and whether different social demographic characteristics have predictive effects on anxiety classes to reduce the incidence of PSA and improve the quality of life of stroke survivors, providing a basis for the development of targeted interventions to alleviate PSA in the future.

Hence, in the present study, anxious subgroups among stroke survivors were evaluated by LPA, and the association of anxious subgroups with socio‐demographic, primary ADL status and PSD level was detected to explain the heterogeneity of PSA.

3. THE STUDY

3.1. Aim

The goals of this study are as follows: (a) to determine the latent classes of post‐stroke anxiety in rural ischaemic stroke survivors and (b) to investigate whether socio‐demographic factors, ADL ability and depression levels are associated with a different latent class of post‐stroke anxiety.

3.2. Design

A cross‐sectional survey.

3.3. Participants

In the present study, 661 ischaemic stroke survivors from some rural areas of Anyang city, Henan Province, China, were included by convenience sampling method between July 2021 and August 2021. The inclusion criteria for participants are as follows: (a) age ≥ 18 years; (b) patients who met the diagnostic criteria of cerebrovascular disease and were confirmed as patients of ischaemic stroke by craniocerebral CT examination or magnetic resonance imaging (MRI); (c) the illness of the participants was in convalescence and (d) the patients were able to communicate and gave consent for participation. The exclusion criteria for participants are (a) patients with mental disorders and (b) patients having an additional serious physical illness.

3.4. Instruments/validity and reliability

3.4.1. Socio‐demographic characteristics

The socio‐demographic data of this study included sex, age, marital status, education level, living arrangement, household monthly income (¥), chronic disease, familial history of stroke, smoking, drinking, rehabilitation training and types of medicines consumed.

3.4.2. Self‐Rating anxiety scale

The self‐rating anxiety scale (SAS) was developed by Zung (Zung, 1971) to assess the symptoms of adult anxiety and could be used to evaluate changes in the anxiety status during treatment. It consisted of 20 items, among which 15 items were scored positively and five items were scored negatively. These items used a Likert 4 scale that ranges from ‘no or sometimes’ to ‘most or all of the time’ and was scored from 1 to 4. The total score was each item added up and multiplied by 1.25. A higher score indicated a higher level of anxiety. An anxiety score below 50 was considered normal, an anxiety score between 50 and 60 was considered mild anxiety, an anxiety score between 61 and 70 was considered moderate anxiety and an anxiety score above 70 was considered severe anxiety. The Cronbach's α was 0.835 in this study.

3.4.3. Self‐rating depression scale

Self‐rating depression scale (SDS) was adapted by Zung (Zung, 1971) from the self‐rating anxiety scale, which was widely used to assess patients' depression state and severity of depression. The scale consisted of 20 items, among which 10 items were scored positively and 10 items were scored negatively. Items used a 4‐point Likert scale that ranged from ‘No or little time’ to ‘Most or all of the time’ and were scored from 1 to 4. The 20 items were added to the rough score and multiplied by 1.25 to get the standard score. According to the Chinese norm, the cut‐off value of an SDS standard score was 53, and SDS standard score ≥ 53 was considered depression, with higher the score, higher the level of depression. A score between 53 and 62 was classified as mild depression, a score between 63 and 72 was classified as moderate depression and a score greater than 72 was classified as major depression. The Cronbach's alpha coefficient of the SDS was 0.917 in this study.

3.4.4. The Barthel index of ADL

The scale was designed by Florence Mahoney and Dorothy Barthel of the United States and applied in clinical practice for measuring the ability to perform daily living activities of stroke patients (Collin et al., 1988). It has good reliability and validity and is widely recognized internationally (Wang et al., 2022). The scale included 10 items: eating, bathing, grooming, dressing, controlled stool, controlled urination, going to the toilet, bed and chair transfer, flat walking for 45 meters and moving up and down the stairs. The score was based on four functional levels of 0, 5, 10 and 15, with an overall score of 100. A higher score demonstrated a better ability to take care of daily life. A score of 0 to 20 indicated a very serious functional deficiency requiring complete dependence for life; a score of 25 to 45 points indicated a severe functional impairment that requires a lot of help in life; a score of 50 to 70 indicated moderate functional impairment and required some help in life; a score of 75 to 95 demonstrated an ability to perform basic self‐care and a score of 100 indicated complete self‐sufficiency in daily life. The Cronbach's alpha coefficient of the ADL was 0.885 in this study.

3.5. Ethics

The participants were surveyed face to face by nursing graduate students who had received uniform training. Informed consent was obtained during the survey and participants' privacy was strictly protected. The researchers collected and numbered questionnaires immediately after the participants completed them. The double input principle was used to ensure the accuracy of the questionnaire input. This study was conducted according to the principles outlined in the Declaration of Helsinki (World Medical Association, 2013).

3.6. Data analysis

Data were analysed using SPSS Version 26.0. Social demography was described by SPSS. Continuous variables were represented by mean ± standard deviation, like age, SAS and SDS total. Frequency and percentage were expressed using categorical variables, including sex, living arrangement, chronic disease and education level.

The LPA was conducted using the Mplus version 7.4 for statistical analysis. The initial model assumed to be a category was analysed and the number of sections in the potential category model was gradually increased until the model fitting data reached the optimal requirements. The main model fitting indices in LPA were used to determine the optimal model including Akaike information criterion (AIC), Bayesian information criterion (BIC), sample size adjusted BIC (SABIC), entropy index, Lo–Mendell–Rubin adjusted likelihood ratio test (LMRT) and Bootstrapped likelihood ratio test (BLRT). The AIC, BIC, SABIC. AIC, BIC and aBIC were the signal evaluation indices in potential profile analysis with lower values indicating better model fit. Entropy was used as an indicator of classification accuracy and higher values indicate better classification accuracy. The value ranges from 0 to 1, and ≥0.8 indicated that the accuracy of model classification was at least 90%. The LMRT and BLRT represented the difference in model fitting, and p < 0.05 indicated the model fit was optimal (Muthen & Muthen, 2012). In addition, the default full information maximum likelihood (FIML) in the Mplus program is used to process low‐level missing data.

4. RESULTS

4.1. Basic characteristics and two scales

In Anyang, China, 661 rural ischaemic stroke patients completed the questionnaire survey of which 55.4% were men. The participants were in the age range of 40–99 years, with the mean (±SD) age of the patients being 66.00 (±10.57) years. A total number of 457 (69.1%) patients lived with their spouses, 396 (59.9%) patients had received education till the elementary level or below and 560 (84.7%) patients had additional chronic diseases. The household monthly income (¥) of patients was usually less than 2000 RMB (n = 518, 78.4%), 452 (68.4%) patients had a family history of stroke and 488 (73.8%) had acquired no rehabilitation training. Most patients took two to five types of medicine (n = 531, 80.3%). Additionally, the mean (±SD) values of anxiety and depression were 38.52 ± 9.74 and 42.37 ± 11.86 respectively. Other socio‐demographic characteristics of the sample are shown in Table 1.

TABLE 1.

Descriptive characteristics of the sample (N = 661)

Variables	Groups	Mean ± SD/N (%)
Sex	Male	366 (55.4)
	Female	295 (44.6)
Age	40~	21 (3.2)
	50~	121 (18.3)
	60~	215 (32.5)
	70~	304 (46.0)
Spouse	Have	550 (83.2)
	No	111 (16.8)
Education	Elementary school or less	396 (59.9)
	Middle school	177 (26.8)
	High school or Vocational Training	88 (13.3)
Living arrangement	Alone	86 (13.0)
	Living with spouse	457 (69.1)
	Living with others	118 (17.9)
Household monthly income (¥)	<2000 RMB	518 (78.4)
	2001 ~ 3000 RMB	72 (10.9)
	3001 ~ 5000 RMB	50 (7.6)
	>5000 RMB	21 (3.2)
Chronic disease	Have	560 (84.7)
	No	101 (15.3)
Family history of stroke	Have	452 (68.4)
	No	209 (31.6)
Smoking	Have	265 (40.1)
	No	396 (59.9)
Drinking	Have	210 (31.8)
	No	451 (68.2)
Rehabilitation training	Yes	173 (26.2)
	No	488 (73.8)
Types of medicine	1	70 (10.6)
	2 ~ 5	531 (80.3)
	6	60 (9.1)
ADL	No dependence	316 (47.8)
	Mild dependence	284 (43.0)
	Moderate dependence	32 (4.8)
	Heavy dependence	29 (4.4)
Anxiety	Total	38.52 ± 9.74
	No anxiety	579 (87.6)
Variables	Groups	Mean ± SD/N (%)
	Mild anxiety	64 (9.7)
	Moderate anxiety	14 (2.1)
	Severe anxiety	4 (0.6)
Depression	Total	42.37 ± 11.86
	No depression	518 (78.4)
	Mild depression	102 (15.4)
	Moderate depression	37 (5.6)
	Severe depression	4 (0.6)

4.2. Latent profile analysis of anxiety

The fitting index of anxiety LCPA in rural ischaemic stroke patients was shown in Table 2. The fitting index reveals that model 3 is the best because AIC, BIC and aBIC have smaller values and the highest entropy. The LMR (p < 0.05) and BLRT (p < 0.001) of model 3 were statistically significant. Additionally, the probability distribution of model 3 was more reasonable as compared to models 4 and 5 which were 0.653 (n = 431), 0.179 (n = 118) and 0.169 (n = 112) respectively. Anxiety in rural ischaemic stroke patients was divided into three latent classes, namely: (a) class 1, low‐level, stable group; (b) class 2, moderate‐level, unstable group; and (c) class 3, high‐level, stable group.

TABLE 2.

Model fit indices of LCPA for post‐stroke anxiety

	LL	FP	AIC	BIC	aBIC	LMR	BLRT	Entropy	类别概率
C = 1	−15711.441	40	31502.883	31682.633	31555.632	—	—	—	—
C = 2	−14820.530	61	29763.060	30037.179	29843.502	0.0004	<0.001	0.932	0.783, 0.217
C = 3	−14426.954	82	29017.907	29386.395	29126.042	0.0058	<0.001	0.967	0.653, 0.179, 0.169
C = 4	−13843.643	103	27893.287	28356.143	28029.115	0.0121	<0.001	0.925	0.039, 0.261, 0.609，0.091
C = 5	−13568.984	124	27385.967	27943.193	27549.489	0.2435	<0.001	0.937	0.539, 0.105, 0.091, 0.225, 0.039

Figure 1 demonstrated the scores of the three latent classes of anxiety in rural stroke survivors on 20 items. The scores under class 1 were generally low and stable. Hence, it was called the low‐level stable group. The scores under class 2 had a medium level, but items 5, 13 and 17 scored higher than the others and were, therefore, called a moderate‐level, unstable group The sentence, “The scores under …” has been slightly modified for clarity. Kindly check if the intended meaning of the sentence is retained.". The scores under class 3 were the highest, but stable for each item. Hence, class 3 was labelled a high‐level, stable group.

FIGURE 1.

The distribution of three different potential classes with anxiety levels.

4.3. Subgroup analysis

Table 3 reveals that the individual characteristics of the three potential classes of anxiety are also different, including the socio‐demographic variables. Other disease‐related information, like gender, education level, living arrangement, family monthly income (¥), chronic diseases, ADL level and depression level, had statistical significance for the categories of three different latent anxiety classes. Female patients with a low level of education, living with a spouse, household monthly income <2000 (¥), stroke survivors with other chronic diseases, impaired ADL function and depressed mood were more likely to be classified into class 3 compared with class 1 or class 2.

TABLE 3.

Effects of anxiety categories on individual characteristics based on potential profile analysis (n = 661)

Variables/groups	1(n = 431)	2(n = 118)	3(n = 112)	χ2	p‐value
Sex				7.568	0.023
Male	251 (68.6)	66 (18.0)	48 (13.4)
Female	180 (61.0)	52 (17.6)	63 (21.4)
Age				5.021	0.541
40~	17 (81.0)	2 (9.5)	2 (9.5)
50~	76 (62.8)	20 (16.5)	25 (20.7)
60~	139 (64.7)	44 (20.5)	32 (14.9)
70~	199 (65.5)	52 (17.1)	53 (17.4)
Spouse				1.633	0.442
Have	364 (66.2)	94 (17.1)	92 (16.7)
No	67 (60.4)	24 (21.6)	20 (18.0)
Education				14.570	0.006
Elementary school or less	251 (63.4)	61 (15.4)	84 (21.2)
Middle school	121 (68.4)	37 (20.9)	19 (10.7)
Variables/groups	1(n = 431)	2(n = 118)	3(n = 112)	χ2	p‐value
High school or Vocational Training	59 (67.0)	20 (22.7)	9 (10.2)
Living arrangement				10.699	0.030
Alone	48 (55.8)	26 (30.2)	12 (14.0)
Living with spouse	304 (66.5)	75 (16.4)	78 (17.1)
Living with others	79 (66.9)	17 (14.4)	22 (18.6)
Household monthly income (¥)				12.707	0.048
<2000 RMB	325 (62.7)	93 (18.0)	100 (19.3)
2001 ~ 3000 RMB	56 (77.8)	9 (12.5)	7 (12.2)
3001 ~ 5000 RMB	35 (70.0)	11 (22.0)	4 (8.0)
>5000 RMB	15 (71.4)	5 (23.8)	1 (4.8)
Chronic disease				12.219	0.002
Have	355 (63.4)	98 (17.5)	107 (19.1)
No	76 (75.2)	20 (19.8)	5 (5.0)
Family history of stroke				5.248	0.073
Have	149 (71.3)	29 (13.9)	31 (14.8)
No	282 (62.4)	89 (19.7)	81 (17.9)
Smoking				2.351	0.309
Have	176 (66.4)	51 (19.2)	38 (14.3)
No	255 (64.4)	67 (16.9)	74 (18.7)
Drinking				4.657	0.097
Have	135 (64.3)	46 (21.9)	29 (13.8)
No	296 (65.6)	72 (16.0)	83 (18.4)
Rehabilitation training				1.037	0.595
Yes	116 (67.1)	32 (18.5)	25 (14.5)
No	315 (64.5)	86 (17.6)	87 (17.8)
Types of medicine				7.615	0.107
1	51 (72.9)	14 (20.0)	5 (7.1)
2~5	345 (65.0)	94 (17.7)	92 (17.3)
6	35 (58.3)	10 (16.7)	15 (25.0)
ADL				35.524	<0.001
No dependence	221 (69.9)	62 (19.6)	33 (10.4)
Mild dependence	182 (64.1)	48 (16.9)	54 (19.0)
Moderate dependence	15 (46.9)	4 (12.5)	13 (40.6)
Heavy dependence	13 (44.8)	4 (13.8)	12 (41.4)
Depression				180.409	<0.001
no depression	386 (74.5)	92 (17.8)	40 (7.7)
Mild depression	39 (38.2)	22 (21.6)	41 (40.2)
Moderate depression	6 (16.2)	4 (10.8)	27 (73.0)
Severe depression	0 (0.0)	0 (0.0)	4 (100.0)

5. DISCUSSION

This is the first potential profile analysis of PSA among rural ischaemic stroke patients that recognized various subgroups of PSA and their influencing factors. Potential profile analysis is a new, scientific statistical analysis method that can classify post‐stroke anxiety into different potential categories based on patients' self‐report. This could further help healthcare providers to develop targeted therapeutic interventions as per the characteristics of different subgroups of post‐stroke anxiety to reduce it among specific subgroups.

This study divided post‐stroke anxiety among stroke patients in rural China into three different potential profiles that are as follows: (a) class 1, low‐level, stable group; (b) class 2, moderate‐level, unstable group; and (c) class 3, high‐level, stable group. The number of people in class 1 was the largest among the three categories, accounting for 65.3%. Classes 2 and 3 accounted for 17.9% and 16.9% of the three classes respectively. This result shows that the PSA scores of class 1 were generally low and stable, indicating that this group had a lower level of PSA. The PSA scores of class 2 were generally of moderate level, but three reverse items that scored higher were as follows: ‘I feel everything is fine and nothing bad happens to me’, ‘I breathe in and out easily’ and ‘My hands and feet are often dry and warm’. These results suggested that PSA in stroke survivors of class 2 had affected the physical status of patients, causing breathing difficulties and other conditions. Other studies have confirmed these results, showing that breathing activity is strongly associated with anxiety, and breath control could effectively help patients in reducing negative emotions (Jerath et al., 2015). The stroke survivors of class 3 generally had higher scores of PSA, but as shown in Figure 1, the scores in these patients were relatively stable in each item. This suggested that although class 3 had a highly anxious mood, the impact of anxiety on physical function was lesser than class 2. These results determined that healthcare providers should be more attentive to classes 2 and 3. The results also suggested creating targeted interventions as per the characteristics of different subgroups, thus providing a practical reference for future research on anxiety among rural ischaemic stroke survivors.

This study revealed that 12.3% of rural ischaemic stroke patients had post‐stroke anxiety, which was consistent with the global incidence of post‐stroke anxiety (9.4% ~ 36.7%) (Almhdawi et al., 2021). The results demonstrated that post‐stroke anxiety is prevalent in ischaemic stroke patients in rural China. Ischaemic stroke survivors with the following factors, like female patients, low education level, living with others, low monthly income, other chronic diseases, ADL disorder and depression, had higher PSA. Being a female patient is a risk factor for PSA, and a higher level of post‐stroke anxiety is included in classes 2 and 3. Previous studies have revealed that being a woman is an important predictor of PSD due to several differences in brain tissue, biology, hormones and function between men and women. Hence, this further indicated that women might be at higher risk of developing PSD (Alonso et al., 2004; Shaywitz et al., 1995). There is a positive correlation between anxiety and depression, indicating that the higher the level of depression, the higher the level of anxiety.

Stroke survivors with low education levels had higher anxiety. However, stroke survivors who had either passed high school or had received vocational training experienced the lowest levels of anxiety in class 3. This indicated that the anxiety level of stroke survivors could be changed by improving their educational level. Patel's (Patel et al., 2018) study demonstrated that illiteracy could hinder the recovery of stroke survivors and cause mood disorders. Lin FH's (Lin et al., 2019) study pointed out that acquiring more disease information could decrease the negative emotions of stroke survivors. Additionally, it was suggested that stroke survivors with higher education levels could easily understand disease information and were more proactive. Therefore, highly educated stroke survivors could deal with their negative emotions in a better way and overcome them. Hence, these findings suggested that healthcare providers should improve the disease‐associated awareness of stroke survivors, provide them with more information and support, help them in preventing disease recurrence and taking emergency measures, reduce the fear associated with the disease and relieve anxiety.

Stroke survivors who lived with their spouses and other family members had lower levels of PSA. However, stroke survivors who lived alone had a higher proportion in class 2 (medium‐level, unstable group). These results suggested that solitary stroke survivors had both anxiety and more physical adverse reactions. Previous studies have revealed that feeling lonely and living alone were associated with higher mortality and morbidity due to stroke (Hakulinen et al., 2018; Tillmann et al., 2017) in addition to increased ADL damage in stroke survivors. This might be a result of the poor physical function of stroke survivors who lived alone and did not receive help and support from caregivers, thus making them feel lonely, resulting in higher levels of PSA and adverse physical reactions.

Stroke survivors who had a lower monthly income were more likely to be assigned to classes 2 and 3 as compared to class 1. This denoted that the anxiety level of stroke survivors with lower monthly income is higher. In previous studies, it was established that the most common causes of anxiety symptoms were associated with material resources and serious illness (Paprocka‐Borowicz et al., 2021). The possible reason behind this could be that economically disadvantaged stroke survivors received fewer medical resources and could not do rehabilitation exercises effectively on time which further hindered their recovery from the disease and aggravated their anxiety. Meanwhile, a lower income also reduced the participation of these patients in social activities and reduced their general well‐being to a certain extent. The primary source of income for Chinese farmers is farm work, and their economic income is generally lower than urban residents. Therefore, PSA is more common among rural stroke survivors.

The present study established that stroke survivors suffering from other chronic ailments, like diabetes mellitus, hypertension and heart disease, were more likely to be grouped into classes 2 and 3. Previous studies have shown that apart from stroke, other self‐reported physical ailments had a significant correlation with post‐stroke anxiety levels (Almhdawi et al., 2021), which is consistent with the results of this study. A study by Ćwirlej (Ćwirlej‐Sozańska et al., 2019) researchers revealed that multi‐morbidity was the main risk factor for ADL disability. Stroke is a serious cerebrovascular disease, which has caused a notable deterioration in the physical health of patients. Additionally, multi‐morbidity would increase the physical burden of stroke survivors. Particularly, rural stroke survivors had a weak knowledge about medication, disease awareness and lower compliance. This further aggravated the adverse outcome of the disease and caused a high level of PSA.

This study demonstrated that the detection rate of PSD in rural China was 21.3%, which is slightly lower than the global PSD prevalence range (22%–40%) (Hackett & Pickles, 2014; Shi et al., 2015). The reason for the detection of PSD might be due to the characteristics of Chinese people, like chatting and doing entertainment activities together with many people. Moreover, houses in rural China are relatively close to each other, hence, they have closer contact with neighbours. This could reduce PSD to a certain extent. Additionally, the present study also established that there was a significant positive correlation between PSA and PSD. These findings were consistent with other literature (Barker‐Collo, 2007 Reference 'Barker et al., 2017', year 2017 has been changed to 2007 to match the reference list. Please check for correctness."). In other words, stroke survivors with severe PSD are more likely to be assigned to classes 2 and 3 of PSA. Lee EH's study showed that the level of PSD in stroke patients was associated with cognitive impairment. However, the association between PSA and cognitive ability was limited, and it disappeared in the presence of PSD. This further indicated that the association between PSA and cognitive ability co‐occurred with depression, and also established that PSD and PSA are highly correlated. These results suggested that healthcare workers should be more attentive to depression while taking targeted intervention measures for PSA in stroke survivors, alleviate comorbid mood disorders, and further improve the cognitive ability of stroke survivors.

5.1. Limitations

There are several limitations to this study. Firstly, the cross‐sectional study could only reflect the current potential profile of anxiety in stroke survivors and could not observe the changes in PSA subgroups in longer stroke survivors. Therefore, further longitudinal studies would be conducted to explore the transformation of anxiety subtypes in stroke survivors and give a theoretical basis for the development of targeted interventions. Secondly, the sample size of this study was only collected from one central region in China, which might lead to deviations due to the differences in economy, culture and medical advancement in different regions. Future studies would include survivors of ischaemic stroke from different regions to conduct a large sample survey.

6. CONCLUSION

In this study, the anxiety status of ischaemic stroke survivors in rural China was evaluated by obtaining three different potential anxiety classes (Class 1, low‐level, stable group; Class 2, moderate‐level, unstable group; and Class 3, high‐level, stable group) and identifying the factors influencing PSA such as female patients, low education level, living with others, low monthly income, other chronic diseases, ADL disorder and depression. This approach could enable healthcare providers to identify subgroups of patients with PSA and also give a basis for further development of targeted interventions to reduce PSA. This study is of great clinical significance in promoting physical and mental well‐being among stroke survivors.

AUTHOR CONTRIBUTIONS

All authors have agreed on the final version and meet at least one of the following criteria (recommended by the ICMJE [http://www.icmje.org/recommendations/]):

1. Substantial contributions to conception and design, acquisition of data or analysis and interpretation of data;

2. Drafting the article or revising it critically for important intellectual content.

FUNDING STATEMENT

This work was supported by the Special Funding for the Construction of Innovative Provinces in Hunan (Grant No. 2019SK2141), Philosophy and Social Science Planning Fund of Henan Province (No.2020BSH013), Xinxiang Soft Science Research Project (No. RKX2021008) and the China Oceanwide Holding Group Project Fund (Contract As per style sheet Ethical statement is required. Please provide" No. H201910150780001).

CONFLICT OF INTEREST STATEMENT

No conflict of interest has been declared by the authors.

ETHICS STATEMENT

This study has been approved by the Ethical Review Committee of Xinxiang Medical College (XYLL‐2020146) with the informed consent of the participants.

Zhang, H. , Ma, J. , Sun, Y. , Xiao, L D. , Yan, F. , & Tang, S. (2023). Anxiety subtypes in rural ischaemic stroke survivors: A latent profile analysis. Nursing Open, 10, 4083–4092. 10.1002/nop2.1668

DATA AVAILABILITY STATEMENT

The data sets used and/or analyzed during the current study is available from the corresponding author on reasonable request.