Knowledge, attitude and practice of poststroke depression among patients with poststroke depression and their family members in Heilongjiang Province, China: a cross-sectional study


Stroke is a major cause of long-term disability and the fourth leading cause of death globally, resulting in various physical, cognitive and emotional sequelae.1 Poststroke depression (PSD) is one of the most prevalent and burdensome neuropsychiatric conditions that can occur after stroke, characterised by symptoms such as sadness, hopelessness, irritability and loss of interest in previously enjoyable activities.2 3 Studies have reported the prevalence of PSD to range from 18% to 61%, with a pooled estimation of 27%.4 For example, the prevalence of PSD within 2 weeks after stroke was found to be 25.4% in China.5–7 PSD could increase the risk of mortality, exacerbate cognitive deficits, lead to greater long-term disability and lower quality of life for stroke survivors and increase the rates of suicidal ideation in individuals with PSD compared with those without depression after stroke.1 8–11 Family members of patients with PSD may also experience emotional, social and financial burdens, as well as a decline in overall mental and physical health.12

Undiagnosed PSD is common, partly because stroke survivors may have difficulty recognising their own depressive symptoms.13 Insufficient knowledge about PSD and its symptoms among patients and their family members can impede timely treatment and optimal recovery.14 After a stroke, survivors may face physical impairment, comorbid health problems and cognitive deficits, which can lead to negative self-perception, negative worldview and pessimism.15 Conversely, good knowledge of PSD and its symptoms may be associated with increased compliance with treatment recommendations and greater symptom relief. Moreover, family members who are knowledgeable about PSD are more likely to detect the condition early, facilitating prompt treatment and management. Nevertheless, it is worth noting that knowledge alone may not improve patients’ overall quality of life as it depends on their attitude towards lifestyle modification.16

Knowledge, attitude and practice (KAP) surveys are valuable tools for assessing the understanding and behaviour regarding public health problems. These surveys can identify obstacles that may hinder adherence to treatment plans, including medication adherence, access to healthcare services and lifestyle modifications.17 However, no KAP studies have been conducted in China to evaluate the scores of patients with PSD and their family members. The objective of this study is to gather KAP scores of patients with PSD and their family members in China. This lack of research impedes a thorough understanding of the needs of patients with PSD and their family members, limiting the effectiveness of treatment and personalised interventions that could enhance their quality of life and treatment outcomes.


Study design and participants

A web-based cross-sectional study was conducted in Heilongjiang Province between October 2022 and April 2023 to investigate the KAP of patients with PSD and their family members who voluntarily participated in the study. Subjects also had to meet basic inclusion and exclusion criteria. Inclusion criteria: (1) aged between 18 and 75 years; (2) patients who had a stroke diagnosed by CT or MRI with reference to diagnostic points of various major cerebrovascular diseases in China 201918; (3) patients who had a stroke diagnosed with PSD with reference to Health Commission issues Chinese version of International Classification of Diseases (Eleventh Revision),19 without restriction of disease stage; (4) score ≥7 on the 17-item version of the Hamilton Depression Inventory; (5) depressive symptoms occur after stroke and (6) patients or their families provided written informed consent. Exclusion criteria: (1) patients with Brief Mental State Assessment Scale ≤2120 and with moderate-to-severe cognitive impairment, hearing impairment or aphasia; (2) experiencing severe memory impairment and could not recall past events accurately and (3) history of mental illness.

The family members herein must be (1) fully aware of the patient’s past habits and personality; (2) knowledge of the patient’s illness and treatment and (3) accompanying the patient throughout the treatment process.


The self-administered questionnaire in this web-based cross-sectional study comprised four dimensions, which were based on relevant literature and guidelines.21–23 To ensure the validity and reliability of the questionnaire, feedback from three experts, including two neurology specialists and one psychology specialist, was incorporated. A pretest was conducted on a small number of participants (n=36), which were included in the data collecting process. The Cronbach’s α value of 0.825 from the pilot study indicated good internal consistency.

The final questionnaire was composed of four parts and provided in online supplemental material: (1) demographic characteristics; (2) knowledge dimension, which consisted of 10 questions with a possible score range of 0–10, where 1 point was awarded for each correct answer and zero for incorrect or unclear answers; (3) attitude dimension, which consisted of eight questions rated on a 5-point Likert Scale, with a score range of 8–40 and (4) practice dimension, which included 10 questions also rated on a 5-point Likert Scale, with a score range of 10–50. In the knowledge dimension, several items were inaccurately expressed. Respondents received 1 point for indicating ‘incorrect’, while responses of ‘correct’ or ‘uncertain’ scored no points. In the attitude and practice dimensions, responses labelled with ‘P’ received positive ratings on a scale from ‘Strongly agree/Very conforming’ (5 points) to ‘Strongly disagree/Very non-conforming’ (1 point). Conversely, responses labelled with ‘N’ indicated negatively framed questions, with scores reversed from ‘Strongly agree/Very conforming’ (1 point) to ‘Strongly disagree/Very non-conforming’ (5 points). Participants with scores below 70% in each dimension were classified as having an insufficient score, those with scores between 70% and 80% were categorised as having a moderate level, and those with scores above 80% were deemed to have a high score.

Supplemental material

To safeguard the participants’ privacy, the study team employed various efforts. The online questionnaire platform used a secure data transmission channel and data backup to securely store all collected data anonymously. The study team also ensured that the sample size was sufficient to produce meaningful results. Based on a sample size calculation with a CI of 95%, a margin of error of 5% and an estimated response rate of 50%, the study required a minimum of 385 participants. To account for a 20% potential dropout rate or incomplete responses, the study team aimed to recruit at least 480 participants. During the data collection period, regular checks were conducted to monitor progress and ensure data quality. Any incomplete or inaccurate questionnaires were excluded from the analysis. To ensure the completeness and accuracy of the questionnaire data, the online questionnaires were created using the Wen Juan Xing platform ( Participants accessed and completed the questionnaire by scanning a quick response code via WeChat. Quality assurance measures were implemented to prevent incomplete responses or data errors. First, only one submission per internet protocol address was allowed, and all items in the questionnaire were mandatory. Following data collection, the research team reviewed the completeness, internal coherence and reasonableness of all questionnaires. The questionnaires were distributed to eligible patients or their family members during recruitment, and a research assistant was available to provide assistance throughout the process to ensure that they fully understood the questionnaire. The Second Affiliated Hospital of Heilongjiang University of Traditional Chinese Medicine was the sole site for data collection.

Statistical analysis

Stata V.17.0 (Stata Corporation, College Station, Texas, USA) was used for the analysis. The continuous data were expressed as mean±SD, while categorical information was expressed as n (%). The KAP scores among respondents with different demographic characteristics were compared using t-test and one-way analysis of variance. Pearson correlation analysis was conducted to evaluate the correlations among KAP scores. Furthermore, multivariate logistic regression analysis was conducted with attitude and practice scores as dependent variables to analyse the relationships between sociodemographic information and KAP. The results were categorised using the 70% of the distribution of KAP scores. Univariate variables with p<0.05 were included in the multivariate regression model. Structural equation modelling (SEM) was used to validate the following hypotheses: (1) the knowledge has a direct impact on their attitude and practice and (2) the attitude directly impacts the practice. The model fitting was evaluated with model chi-square χ2 (CMIN)/df, root mean square error of approximation (RMSEA), Incremental Fit Index (IFI), Tucker-Lewis Index (TLI) and Comparative Fit Index (CFI). A two-sided p<0.05 was considered statistically significant.

Patient and public involvement statement

For a comprehensive exploration of KAP regarding PSD among patients, the research team incorporated a patient representative. The involvement of this representative extended to providing insights into the study’s design and conception. The patient representative was selected from the Second Hospital affiliated with Heilongjiang University of Traditional Chinese Medicine.


Sociodemographic information

A total of 682 questionnaires were collected from patients or their family members who volunteered to participate in the study. Among them, 190 questionnaires were excluded due to incomplete KAP data, and 3 were excluded due to incorrect age information, leaving 489 valid questionnaires (80.30%) for analysis. Among the respondents, 53.09% were patients. The participants had a mean age of 54.68±13.80 years, and the majority were unmarried (72.39%), male (60.12%) and lived in urban areas (66.26%). Participants with junior college education and above accounted for 41.19% of the sample, and 127 (25.97%) participants were employed. Nearly half of the respondents had basic medical insurance for urban employees (49.69%) (table 1).

Table 1

Participants’ sociodemographic information and KAP scores

Knowledge, attitude and practice

The participants achieved an average knowledge score of 6.36±2.66 (possible range: 0–10). There was no significant difference in knowledge scores observed between patients who had a stroke and family members (figure 1A). The knowledge dimension had a correct response rate ranging from 5.52% to 80.37%. A majority of participants (80.37%) recognised that the best treatment effect could be achieved by systematically using a variety of therapies such as pharmacotherapy, psychotherapy and rehabilitation training (item 5, table 2). However, only 5.52% of participants were aware that combining psychotherapy and medication should not be used in cases with mild symptoms and without cognitive disorders and communication impairment (item 8, table 2).

Figure 1
Figure 1

Distribution of knowledge (A), attitude (B), and practice (C) scores of patients who had a stroke and family members. The vertical axis represents the KAP scores and the horizontal axis represents the population categories (respectively, patients who had a stroke and family members). The distribution of different populations in each score is shown by the width of the horizontal axis. KAP, knowledge, attitude and practice.

Table 2

Distribution of knowledge dimension

The participants exhibited an average attitude score of 29.07±5.18 (possible range: 8–40). Notably, there were no significant differences in attitude scores observed between patients and family members (figure 1B). The proportion of participants who responded ‘Strongly agree’ or ‘Agree’ towards positive statements in the attitude dimension ranged from 87.52% to 88.34%, whereas 55.56% of them expressed ‘Strongly disagree’ or ‘Disagree’ towards negative statements. A considerable proportion of the population, as high as 88.34%, agreed that it was important to not only focus on somatic symptoms and physical recovery but also mental and cognitive aspects (item 4, table 3). Moreover, over half of the participants (55.56%) did not agree that PSD should be taken lightly (item 2, table 3). Regarding the perception of stroke and PSD, almost half of the respondents (41.51%) always or often felt extremely worried and thought that the patient had caused trouble for others (item 5.1, table 3).

Table 3

Distribution of attitude dimension

The mean practice score for all participants was 37.50±8.49 (possible range: 10–50). No significant differences in practice were observed between patients and family members (figure 1C). As shown in table 4, the majority of participants (84.26%) expressed a willingness for patients to actively cooperate with treatment (item 1), and 74.03% of them acknowledged the importance of maintaining treatment for at least 4–6 months, even after the depressive symptoms have resolved (item 2). Furthermore, the top three therapies to which patients were most willing to undergo were acupuncture treatment (68.92%), physical exercise (66.47%) and Chinese medicinal preparations (62.37%).

Table 4

Distribution of practice section answers

Pearson correlations of KAP

This study revealed knowledge was significantly associated with attitude (r=0.103, p=0.023) and practice (r=0.369, p<0.001).

Multivariate logistic regression analysis

Multivariate logistic regression analysis showed that retirement (OR=0.29, 95% CI 0.11 to 0.77, p=0.012) and monthly income less than ¥2000 (OR=0.46, 95% CI 0.27 to 0.79, p=0.005) were independently associated with adequate knowledge. Knowledge (OR=2.12, 95% CI 1.44 to 3.14, p<0.001) was independently associated with positive attitude. Knowledge (OR=3.85, 95% CI 2.53 to 5.86, p<0.001) and attitude (OR=1.62, 95% CI 1.06 to 2.47, p=0.024) were independently associated with proactive practice (table 5).

Table 5

Multivariate logistic regression analysis

The SEM demonstrated good model fit (CMIN/df=4.466, RMSEA=0.084, IFI=0.831; TLI=0.814; CFI=0.831) (online supplemental table 2). The SEM results confirmed that knowledge was positively associated with attitude (path coefficient=0.503, p=0.030) and practice (path coefficient=0.554, p<0.001), which was consistent with the main findings from Pearson correlation analysis. Moreover, attitude was positively associated with practice (path coefficient=0.080, p<0.001). In addition, significant associations of KAP scores with individual items in each dimension of KAP were widely observed, including CK2–7, CK9–10, CA1–3, CA5.1, CA5.3–CA5.4 and CP2–CP3.8 (all p<0.05) (table 6, figure 2).

Figure 2
Figure 2

Structural equation model showing the associations between demographic characteristics and KAP. All variables are observed variables. Direction of causality is indicated by single-headed arrows, and double-headed arrow indicates a correlation among variables. The standardised path coefficients are presented alongside the arrows. KAP, knowledge, attitude and practice.

Table 6

The estimates of SEM


The study revealed that patients with PSD and their family members had inadequate knowledge, positive attitude and moderate practice towards PSD. Targeted educational interventions could be customised for those with lower monthly income or those who were retired to enhance their knowledge and attitude towards PSD.

The finding in the knowledge dimension indicated that a majority of participants recognised the benefits of combining multiple therapies for treating PSD was promising. Multidisciplinary and multimodal approaches to treating PSD have gained increasing recognition in recent years, and using pharmacotherapy, psychotherapy and rehabilitation training in combination has been proven to be effective in decreasing depressive symptoms and improving the quality of life in stroke survivors.24 25 However, the low awareness of the risks of combining psychotherapy and medication in cases with mild symptoms and without cognitive disorders and communication impairment was concerning. Previous studies have also reported similar findings of inadequate awareness and knowledge regarding PSD treatment among stroke survivors, as well as their family caregivers.26 27 Therefore, developing educational interventions for stroke survivors, family caregivers and healthcare professionals on PSD treatment options and engaging in patient-centred communication to dispel misconceptions were crucial to address the issue of low awareness of treatment risks and benefits.

In the attitude dimension, the item which most participants agreed signified the heightened awareness of the impact of PSD on mental and cognitive aspects among patients and their family members. Prior studies indicated that PSD was frequently overlooked and undertreated,28 29 and that patients who had a stroke might not receive the necessary support for their mental health requirements.30 Studies have established an increasing appreciation of the importance of addressing the psychological needs of patients who had a stroke through interventions like cognitive–behavioural and problem-solving therapy.31 32 Nevertheless, obstacles to accessing mental healthcare, including stigma and resource shortages, posed substantial obstacles that need to be resolved. However, it was concerning that a significant proportion of respondents suffered from PSD and self-blame, resulting in feelings of being a burden on others. This finding was consistent with previous studies that have shown that PSD was a prevalent and debilitating problem among stroke survivors and their caregivers.33 34 Addressing this issue required education and support for patients and their family members regarding the nature of stroke and PSD, including risk factors, symptoms and treatments.35 Psychological counselling and therapy,36 as well as lifestyle interventions37 such as exercise, healthy diet and social support, could be helpful in reducing the risk and severity of PSD and improving overall mental and physical health in patients who had a stroke.

The findings in the practice dimension indicated that a majority of participants expressed their willingness to actively cooperate with the treatment. This result was in line with previous research demonstrating high levels of treatment adherence and compliance among people with depression.38 Additionally, those participants who were more actively engaged in treatment, such as with exercise intervention, demonstrated greater improvements in depressive symptoms and overall functioning.39 Furthermore, 74.03% of the participants recognised the importance of continuing the treatment for a minimum of 4–6 months even after their depressive symptoms had resolved. This outcome aligned with clinical guidelines that recommend the continuation of treatment for several months postremission to prevent relapse.40 Continuation of depression treatment, involving medication, psychotherapy or their combination, was crucial to consolidate treatment gains, prevent relapse and improve long-term outcomes. Patients with a history of recurrent or chronic depression were particularly vulnerable to relapse, necessitating the need for ongoing treatment. Thus, educating patients and their family members on the importance of continuing treatment and promoting adherence to guidelines were critical for ensuring successful treatment outcomes. Besides, Chinese medicinal preparations were among the most accepted therapies, likely due to their popularity in East Asian countries and their long history of use in traditional Chinese medicine.

Specifically, stroke survivors and their family members with higher knowledge scores of PSD exhibited more favourable attitude towards its treatment and management, as well as better practice for managing the condition. These findings suggested that educating both stroke survivors and their family members about PSD might improve overall attitude and practice towards its management, ultimately enhancing the quality of life. Previous study has similarly emphasised the importance of knowledge in fostering positive attitude and practice towards mental health issues.41

Besides, the positive association between attitude and practice was only observed in the SEM analysis, suggesting that other variables or underlying factors adjusted in the SEM might contribute to this association. This result emphasised the need for multivariate perspective when analysing relationships between KAP scores. Furthermore, the study revealed significant associations between KAP scores and items within each KAP dimension. Addressing these specific items through targeted interventions could lead to more effective improvements in PSD management. Notably, socioeconomic factors, such as retirement status and income, might influence knowledge and awareness of PSD among stroke survivors and their families. Retirement imposed detrimental effects on knowledge level due to reduced exposure to recent developments, inadequate access to continued education or training and a lack of motivation to stay informed. Additionally, lower income levels restricted access to educational resources, prioritising basic necessities over education or training, and insufficient exposure to diverse experiences or perspectives, thus leading to insufficient knowledge. Targeted educational interventions should be developed to address the impact of socioeconomic factors on PSD knowledge and awareness, with a focus on accessibility and relevance to stroke survivors, families and healthcare professionals. This could be achieved through workshops, seminars and online resources while ensuring accessibility to individuals from all socioeconomic backgrounds through community-based approaches and resources in multiple languages.

This study has several limitations. First, the data collected through self-administered questionnaires are prone to recall bias and social desirability bias, potentially affecting the accuracy of the reported KAP related to PSD. Second, while certain demographic factors (retirement, monthly income) were found to be associated with KAP, the study did not account for potential confounding variables, limiting the ability to establish direct causal links. Additionally, the study’s findings may not be generalisable beyond the specific geographic area (Heilongjiang Province), limiting their applicability to populations with different cultural, socioeconomic or healthcare contexts.

This post was originally published on