Suicidal ideation and influencing factors among hospitalized older adults with functional impairments: a mixed-methods study

Abstract

Background

Compared to their healthy counterparts, hospitalized older adults with functional impairments face a heightened risk of suicidal ideation due to disease-related pain, loss of independence, and reduced social engagement, positioning them as a high-risk demographic for suicide. However, extant research has predominantly focused on generally healthy older populations, leaving a significant gap regarding those with substantial functional limitations. This study aims to investigate the prevalence and influencing factors of suicidal ideation among hospitalized older adults with functional disabilities.

Methods

We conducted a two-part mixed-methods study. After hospital manager approval, we surveyed the suicide ideation of 717 hospitalized older adults with functional impairments. We carried out participatory observation with 4 hospitals and semi-structured interviews with 10 hospitalized older adults with functional impairments to elicit detailed information on the process leading to suicidal ideation.

Results

The prevalence of suicidal ideation among hospitalized older adults with functional impairments was notably high (17.4%). The quantitative and qualitative findings corroborate and complement each other. Factors significantly associated with suicidal ideation (p < 0.05) included lower monthly income, bad family relationships, high experiential avoidance, high depressive symptoms, and low resilience. Through qualitative analysis, three overarching categories of influencing factors were identified: vulnerability factors, risk factors, and protective factors.

Conclusions

Given the high prevalence of suicidal ideation among hospitalized older adults with functional impairments, it is imperative for healthcare professionals to prioritize monitoring those with identified vulnerability factors, promptly recognize emerging risk factors, and actively leverage protective factors to implement tailored interventions aimed at mitigating suicidal ideation.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40359-025-03896-2.

Background

According to the United Nations’ World Population Prospects 2022 report, the global proportion of individuals aged 65 and above has risen from 6% in 1990 to 10% in 2022. This figure is projected to climb to 16% by 2050 [1]. In parallel with this deepening demographic aging and a consequent epidemiological transition characterized by a higher prevalence of chronic diseases, the number of older adults experiencing functional impairment, either partial or severe, has been increasing annually on a global scale. The World Health Organization estimates that approximately 14% of the world’s elderly population contends with moderate to severe functional disabilities [1]. In China, the population of older adults with functional impairments had reached 46.54 million by 2024. Forecast suggests a significant rise in this figure, with estimates indicating 76.11 million older Chinese adults will be living with functional impairments by 2030. This number is anticipated to grow to 120 million by 2050 [2].

Hospitalized older adults often suffer from severe physical dysfunction due to disease deterioration or surgical complications, transitioning from a previously abled state to a disabled state, leading to a high dependence on others in daily life, participation in activities, and social Interactions [3]. This loss of functional ability and change in roles brings about a significant psychological gap, making the elderly susceptible to low self-esteem, high experiential avoidance, feelings of hopelessness, and stigma associated with illness, which can exacerbate negative emotions such as anxiety and depression [4]. The unfamiliar hospital environment further compounds this distress by removing patients from their home settings, adding adaptive pressure and potentially triggering suicidal ideation (SI) [3]. SI involves explicit thoughts of ending one’s life without action and is a critical precursor to suicidal behavior [5]. While previous research has investigated SI either among general geriatric inpatients or among functionally impaired residents in long-term care facilities, for instance, Yeh et al. reported a prevalence rate of 14.7% in hospitalized older patients [6], and one study targeting moderately to severely impaired nursing home residents found a rate as high as 32.4% [7]. However, the prevalence and influencing factors of SI among hospitalized older adults with functional impairments during their stay have not been systematically examined.

Beghi proposed that hospitalized older adults with low self-esteem and high experiential avoidance are more likely to exhibit stronger SI [8]. Huang identified limited social interaction, loneliness, functional impairment, and depressive symptoms as significant risk factors [9], while socio-environmental factors such as living alone and low social support also contribute to poor mental health. In contrast, Zhang and Miao indicated that strong social support, positive psychological traits, and effective family care may mitigate SI risk [10, 11]. Nevertheless, SI arises from complex multifactorial interactions. In many cultures, suicide remains heavily stigmatized, viewed as shameful, sinful, or damaging to family honor. Older adults with functional impairments often internalize such beliefs, making open disclosure of SI difficult due to self-stigma and fear of disgracing their families [10]. Therefore, quantitative research alone may not be able to deeply explore the reasons behind the SI of older adults and the related influencing factors. Qualitative research can delve deeper and more comprehensively into the research results based on building trust with the older adults, validating and supplementing the results of quantitative research [12].

The " Diathesis-Stress Model” of suicide posits that suicidal behavior arises from the complex interplay between risk factors, protective factors, and individual vulnerability [13]. Specifically, predisposing characteristics such as vulnerability or maladaptive coping styles may amplify suicidal ideation (SI) when combined with risk factors like social isolation, depression, or high experiential avoidance. In contrast, protective factors including resilience, social support, and strong family relationships can promote adaptive coping, buffer stress, and reduce SI. Guided by this framework, this mixed-methods study examines the prevalence, demographic correlates, and influencing factors of SI in hospitalized older adults with functional impairment. Quantitative analyses identify rates and predictors, while participatory observation and semi-structured interviews explore lived experiences and psychological pathways. By integrating these approaches, the study aims to identify vulnerability, risk, and protective factors, offering practical insights to improve suicide risk assessment, prevention, and care in this clinical group.

Methods

Design

This study employed a convergent parallel mixed-methods design, comprising two integral parts (Fig. 1). In the first part, a structured questionnaire was administered to hospitalized older adults with functional impairments to identify factors associated with SI. The second part adopted a phenomenological approach, incorporating observational methods followed by in-depth, semi-structured interviews with the same participant group. This dual-faceted methodology allowed for a comprehensive exploration of both the prevalence and subjective experiences related to suicidal thoughts in this vulnerable population.

Fig. 1.

Flowchart of the mixed methods study with an explanatory sequential design SI

Participants

The inclusion criteria for hospitalized older adults with functional impairment were as follows: (1) age ≥ 60 years; (2) a Barthel Index (BI) score of < 100 points; and (3) provision of informed consent for voluntary participation. Exclusion criteria comprised: (1) presence of severe psychiatric disorders accompanied by cognitive impairment; (2) a diagnosis of temporary functional disabilities with an expected full recovery of daily living abilities within six months following short-term treatment; and (3) patients at the end stage of disease or otherwise unable to cooperate with the investigation due to medical conditions. The inclusion and exclusion criteria for the qualitative phase aligned with those of the quantitative component, with one additional inclusion criterion for the qualitative study: participants were required to have documented SI, as indicated by a score > 2 on the Beck Scale for Suicidal Ideation [14]. or the quantitative arm, three departments containing eligible patients were randomly selected from each of four public hospitals in Hunan Province between June and October 2024. The sample size was calculated using the formula: =(1-)/, Accounting for a potential 20% rate of invalid responses or attrition, a minimum of 476 participants were required. The qualitative study was conducted within the same four hospitals and employed a participatory observation approach. Using purposive sampling, participants who scored greater than zero in the fourth or fifth item on the Beck Scale for Suicidal Ideation in the quantitative survey.

Instruments

Quantitative research instruments

General demographic information

A self-developed demographic questionnaire was used to collect the participants’ age, gender, family residence, number of children raised, and relationship between caregivers and individuals, with a total of 16 entries.

SI

The evaluation of participants’ SI in the previous week was conducted via the Chinese version of the Beck Scale for Suicidal Ideation (BSI-CV). It has been validated with good reliability and validity in Chinese nursing home older residents with the Cronbach’s α of 0.89 [7].It used 19 items to evaluate the subject’s suicidal ideation in the last week. When item 4 (i.e., active suicide ideation) or item 5 (i.e., passive suicidal ideation) is answered with “weak” or “moderate to strong” (i.e., not 0), participants continue to finish the remaining items 6 ~ 19. The total scores are 0 to 38, with higher scores indicating stronger suicidal ideation. The Cronbach’s α for the BSI-CV was 0.928 in this study.

Experiential avoidance

The experiential avoidance of the participants was assessed via a Chinese adaptation of the Acceptance and Action Questionnaire-II(AAQ-II). The Chinese version of the AAQ-II was translated in 2013 and it has been validated with good reliability and validity in Chinese cancer patients with the Cronbach’s α of 0.94 [15]. The AAQ-II consists of 7 items, rated on a 7-point Likert scale ranging from 1 (never true) to 7 (always true), with higher scores indicating greater experiential avoidance. The AAQ-II has demonstrated adequate internal consistency and convergent and divergent validity [16]. The Cronbach’s α for the AAQ-II was 0.922 in this study.

Depression

The depression of the participants was assessed via a Chinese adaptation of the Geriatric Depression Scale-15(GDS-15). The scale has been validated by Chinese scholars [17]. The Geriatric Depression Scale (GDS) was created with the express purpose of assessing depressive symptoms in older adults [18]. The scale consists of a total of 15 items. The higher the score, the higher the level of depression. The Cronbach’s α for the GDS-15 was 0.930 in this study.

Resilience

The resilience of the participants was assessed via a Chinese adaptation of the Connor Davidson Resilience Scale (CD-RISC). It was used in the current study, which measures 25 items of five dimensions of resilience. The Cronbach’s α of this scale in the elderly population in Chinese nursing home is 0.92 [7].Participants were asked to respond to each item on a 5-point Likert scale, from 0 (not true at all) to 4 (true all the time). The higher the score, the higher the level of resilience. The Cronbach’s α for the CD-RISC was 0.980 in this study.

Qualitative research instruments

An initial interview guide was developed based on the study objectives and a review of extant literature, structured around the " Diathesis-Stress Model” of suicide. This draft was subsequently refined through research team discussions and consultations with a panel of three experts (comprising a hospital psychiatrist, a senior geriatric nurse, and a psychological counselor). Furthermore, a pilot interview was conducted with two hospitalized older adults with functional disabilities from the same hospital. We set up the final guide based on the previous pilot testing and final interview guide is shown in Table 1.

Table 1.

Interview questions and corresponding main topics

Main topics	Interview questions
Causes of suicide ideation	What changes have you felt in your body since this hospitalization?
	After you were unable to act on your own because of the illness, how did you feel about the change in your mindset from before?
	What do your loved ones, friends, and colleagues think of your condition? What is their attitude towards your inability to act on your own?
Risk factors	Have you ever felt hopeless lately? What causes you to feel this way?
	When did you start having suicidal thoughts? What is the process?
Protective factors	What’s stopping you from putting that idea into action? What specific help do you expect?
	What do you think is the greatest value of your life?
	What’s next for your condition?

Ethic approval

The study protocol was approved by the Ethics Committee of Hunan Normal University (No. 2024 − 172). Written informed consent was obtained from all participants prior to enrollment, in accordance with the principles of the Declaration of Helsinki. Throughout the study, subjects’ emotions will be closely monitored. If significant fluctuations are observed, the study will be immediately suspended. Emotional support will be provided and the head nurse and chief physician notified. Resumption of the study will be determined based on emotional recovery. Persistent distress will necessitate referral to a mental health specialist for appropriate intervention.

Data collection

In part one, the questionnaire was administered by three nursing researchers to identify protective and risk factors for SI in hospitalized older adults with functional impairment. The researchers distributed the questionnaires to hospitalized older adults with functional impairment at the end of each morning meeting and collected them promptly after completion. This study was conducted in Hunan Province.

In part two, data were collected through participatory observation and semi-structured face-to-face interviews. Researchers, assuming the role of assistant nurses, built trust with patients, families, and caregivers during clinical activities. Eligible respondents were invited in person, with study purpose and confidentiality explained. After obtaining consent, interviews were scheduled in private meeting rooms at mutually agreed times. Given the sensitive nature of the research topic, all interviews were conducted by two researchers who had received formal training in qualitative research to ensure the quality of the interviews. Given the sensitive topic, all interviews were conducted by two formally trained qualitative researchers. Using purposive sampling, 10 hospitalized disabled older adults from four hospitals were included. Data saturation was achieved after the 10th participant. The entire process was recorded, and non-verbal behaviors were documented. The files were encrypted for storage. The semi-structured interview lasts 40–60 min.

Data analysis

In this study, qualitative and quantitative data were integrated and reported narratively. For the quantitative component, statistical analyses were performed using IBM SPSS Statistics 26.0 (IBM Corp., Armonk, NY, USA). Categorical variables were summarized as frequencies and percentages. Continuous data following a normal distribution were described using means and standard deviations, while those not normally distributed were expressed as medians and interquartile ranges. Group comparisons were conducted using independent two-sample t-tests, one-way ANOVA, or Kruskal-Wallis H tests, as appropriate. Suicide ideation (dichotomous: presence/absence) served as the dependent variable in a multivariate logistic regression model, which included variables showing statistical significance (P < 0.05) in univariate analyses. Statistical significance was defined a priori as a two-sided = 0.05.

In the second part, within 24 h after each interview, the audio content was transcribed verbatim and reviewed and analyzed by two researchers. NVivo 12.0 software was used for categorization and coding, in combination with Colaizzi’s seven-step method for data analysis. Any discrepancies that arose during transcription and analysis were resolved by consulting a third researcher. The organized data were returned to the participants for verification. Guba and Lincoln’s criteria, including credibility, conformability, and dependability [19], to guarantee the trustworthiness of this study. The credibility of the study was established through triangulation and member checking. Methodological triangulation (integrating qualitative and quantitative findings) and researcher triangulation (multiple researchers in coding consensus) were employed. Before analysis, the authors immersed themselves in the data and actively sought participant feedback on the findings and interpretations, ensuring that the results accurately reflected the participants’ perspectives. To improve confirmability, researchers enhanced quality control by writing reflective journals. To enhance dependability, one external nursing professor reviewed the data, leading to adjustments in the codes and categories based on her feedback.

Results

Quantitative results

Characteristics of participants

Among the 717 hospitalized older adults with functional impairment investigated in this study, there were 403 males (56.2%). The majority of the older adults were aged 61 to 70 years (mean age = 69.7 ± 7.31 years), and a total of 381 (53.1%) lived in rural areas. Heart and brain diseases were the most common conditions (55.0%), and 8.4% of the participants were classified as having the most severe level of functional impairment. Those with urban and rural residents’ health insurance accounted for the majority (43.8%). For detailed information about the older adults, see Appendix 1.

Multifactor logistic regression analysis of SI in hospitalized older adults with functional impairment

First, taking SI as the dependent variable, a one-way analysis was conducted. The results showed that differences in monthly income, education level, place of residence, family relationships, work status before illness, degree of disability, and medical payment methods had statistical significance (P < 0.05, Appendix 1). Including the aforementioned variables in a multivariate logistic regression analysis, the results indicated that a monthly income of less than 3000 CNY, poor family relationships, severe disability, and high levels of depression were risk factors for SI in hospitalized disabled elderly individuals; while having employee medical insurance and high psychological resilience were protective factors against SI (Table 2).

Table 2.

Multifactor logistic regression analysis of SI in hospitalized older adults with functional impairment

Variables	N (%;±)	B	SE	P	OR(95%CI)
Monthly income
>9000	87(12.10)	−2.810	1.240	0.023	0.060(0.005 ~ 0.720)
6001 ~ 9000	225(31.40)	−3.210	1.320	0.015	0.040(0.003 ~ 0.530)
3001 ~ 6000	286(39.90)	−3.980	1.520	0.009	0.019(0.001 ~ 0.360)
≤ 3000(Ref)	119(16.60)
Education level
College and above	113(15.80)	−0.015	0.245	0.951	0.985(0.610–1.592)
High school	80(11.20)	−0.120	0.270	0.656	0.887(0.524–1.520)
Middle school	256(35.70)	0.280	0.195	0.152	1.323(0.902–1.941)
Primary and below(Ref)	268(37.30)
Place of residence
Urban	336(46.90)	0.180	0.165	0.275	1.197(0.867–1.653)
Rural(Ref)	381(53.10)
Family relationships
Bad	47(6.60)	2.950	0.710	<0.001	19.112(4.752–76.881)
Ordinary	294(41.00)	−0.035	0.172	0.839	0.966(0.689–1.354)
Good	376(52.40)
Pre-disabling working conditions
Retire	329(45.90)	0.140	0.160	0.380	1.150(0.841–1.572)
Incumbency	11(1.50)	−0.320	0.700	0.648	0.726(0.184–2.861)
Farmers(Ref)	377(52.60)
Whether it is a first-time disability
No	484(67.50)	−0.110	0.175	0.529	0.896(0.636–1.262)
Yes(Ref)	233(32.50)
Degree of disability
Extremely severely disabled	60(8.40)	1.650	0.390	<0.001	5.207(2.427–11.169)
Severely disabled	69(9.60)	1.480	0.345	<0.001	4.392(2.233–8.639)
Mid disability	119(16.60)	1.020	0.290	<0.001	2.773(1.569–4.902)
Mild disability(Ref)	469(65.40)
Payment Methods
Medical insurance for residents	314(43.80)	−0.870	0.265	＜0.001	0.419(0.249–0.705.249.705
Employee medical insurance	279(38.90）	−0.977	0.255	<0.001	0.419(0.249–0.705)
Privately pay(Ref)	124(17.30)
Hospitalization duration	10.40 ± 6.80	1.773	0.216	<0.001	5.888(3.857–8.987)
Experiential avoidance	26.34 ± 9.69	1.506	0.298	<0.001	4.508(2.516–8.078)
Depression	5.50 ± 4.84	1.306	0.200	<0.001	3.69(2.494–5.461)
Resilience	77.12 ± 28.74	−1.874	0.264	<0.001	0.154(0.092–0.257)

Qualitative results

Interviews were conducted with 10 hospitalized elderly individuals with disabilities, aged between 65 and 80 (mean age = 71.50 ± 9.87), comprising 4 females and 6 males. The average interview duration was 50 min. Through thematic analysis, three influencing factors, eight core themes, and fifteen sub-themes regarding suicidal ideation were identified.

Vulnerability factors

Theme 1: avoidance behavior patterns

Observational records indicated avoidance behaviors in participants P1, P2, P4, and P8. Semi-structured interviews further revealed distinct aspects of such avoidant tendencies. (See Appendix 2 for the observation form.)

1. Sub theme: Social–emotional Avoidance

   After the onset of functional impairment, older adults often transition from a state of health to one of dependency, requiring assistance in daily activities. This shift evoked feelings of shame and stigma, leading to social withdrawal.

   P5: “I haven’t been in touch with my former colleagues since I got sick. We were all so active and well-presented in our younger days. Now that I’m old and can hardly move, reaching out would just make me a laughingstock.”

   P7: “I don’t want to contact my old army comrades. I haven’t even read the messages they sent after I was hospitalized… I just can’t bring myself to. I used to be so strong—none of them would believe what’s happened to me. I can’t bear the pity or the shame.”

2. Sub theme: Avoidance of seeking medical treatment.

   Limited understanding of their conditions, combined with traumatic experiences such as the loss of relatives, led some participants to delay or avoid medical intervention.

   P7: “The doctor told me I have an aortic dissection. I read about it online and it seemed like a death sentence. What’s the point of living like this?”

   P8: “They found a tumor in my brain and want to operate. But what’s the use? My eldest son had a tumor too—he had the surgery and still passed away. I’d rather not go through that. I’m ready to go.”

Theme 2: economic burden and cultural repression

• 3.Sub theme: Burden of disease

  High medical expenses placed severe financial and psychological strain on participants and their families, substantially reducing their overall quality of life.

  P2: “When I was in the ICU, with all those tubes—they said it cost ¥10,000 a day. Our family doesn’t have that kind of money.”

  P10: “We were already struggling financially. Now my daughter’s mother-in-law has cancer, and that’s another big expense. With my condition on top of that, we can barely make ends meet.”

• 4.Sub theme: Suicide stigma bondage

  Influenced by traditional Chinese values, suicide is widely perceived as an act of cowardice, selfishness, or escape. This cultural stigma discourages open discussion of SI, as individuals fear bringing shame to their families.

  P9: “Our family considers suicide taboo. I wouldn’t want the neighbors talking badly about us or pointing fingers at my grandchildren.”

Risk factors

theme 1: disease burden and loss of control

Nursing documentation noted pain management interventions for P4 and P8; the latter was observed hitting his head at night due to pain. Interviews revealed experiences of physical suffering and loss of autonomy.

• 5.Sub theme: Burden of disease.

  Participants endured both physical and emotional pain, including sleep disturbances and side effects from medications, which diminished their quality of life and heightened SI.

  P4: “The bloating and pain keep me up at night… My hair’s all gone (rubbing head)… I can’t do anything. I’m just a waste of space.”

  P8: “The headaches are unbearable… I’m drowsy all day. The meds help me sleep but wreck my stomach. I’m stuck—nothing helps.”

• 6.Sub theme: Loss of physical function.

  Functional impairment led to loss of autonomy and self-perception as a burden.

  P9: “I can’t move my legs. The doctors say I won’t regain sensation. I used to love dancing in the square—now I’m just lying here waiting to die.”

  P10: “Besides being paralyzed on the left, everything hurts. I have diabetes so I can’t eat what I want. Everything tastes bitter now.”

Theme 2: broken social ties

• 7.Sub theme: Family support lapses.

  Absence of spousal support and reluctance among children to provide care left some participants feeling abandoned.

  P1: “When my husband was alive, he helped me with everything. Now my eldest son hardly visits even though he lives nearby. My second daughter is always ‘too busy.’ I could starve to death and no one would notice.”

• 8.Sub theme: Feelings of helplessness and despair.

  Inadequate emotional and practical support contributed to loneliness and existential despair.

  P4: “My children are at work all day. I just lie here staring at the ceiling. Even TV has lost its meaning. I’m just counting down the days.”

Theme 3: existential despair

• 9.Sub theme: Self-loathing

  Dependence on others for basic needs and loss of bodily control (e.g., incontinence) resulted in shame and reduced self-worth.

  P2: “I’m bedridden—completely useless. Just a burden on my children.”

  P9: “I can’t move my legs… (hitting legs) … I used to dance every day. Now I’m a waste.”

  P4: “The pain and bloating… (holding stomach painfully) … I can’t get up. I eat, pee, and crap in bed. I have to have my kids change my diapers. I’ve lost all dignity.”

• 10.Sub theme: Loss of dignity

  Cultural narratives equating disability with loss of value reinforced self-directed stigma.

  P2: “I turned off my phone after being hospitalized. I’m too ashamed to face my colleagues.”

  P7: “I was a serious table tennis player. Now I can’t do anything. I’m afraid my old friends would laugh—I don’t answer their messages.” (Scrolls through old photos.)

Protective factors

Observations noted that P5 received emotional support from her granddaughter, especially during moments of physical struggle.

Theme 1: medical & social support

Nursing records indicated that most patients with moderate dependency or higher required Level I care, including hourly rounds.

• 11.Sub theme: Medical Support.

  As a vulnerable inpatient population, functionally impaired older adults received closer medical supervision. Some were advised about financial aid programs.

  P2: “The department director mentioned a insurance plan that could reduce costs—he even gave my daughter a website.”

  P6: “The doctors encourage me every time I’m here. They say I’m improving. It helps me stay positive.”

• 12. Sub theme: Social support.

  Positive reinforcement from neighbors and relatives alleviated depressive symptoms. Access to social resources also promoted psychological well-being.

  P3: “My neighbors help carry things for me when I’m back home.”

  P8: “A neighbor helped refer me to this hospital in the city and assured me we’d find effective treatment.”

Theme 2: perception of meaning in life

• 13.Sub theme: Perception of the value of life.

  Despite feelings of being a burden, reflecting on life achievements helped mitigate hopelessness.

  P1: “I raised three children for the Liu family. The youngest is now a professor—that’s my life’s greatest accomplishment.” (Showed photos and awards.)

  P2: “My purpose was raising two devoted children. Thinking of them helped me through the worst moments.”

  P5: “I taught so many students back in the day. Some still visit. My granddaughter says she’ll take me in after she marries.” (Smiled.)

• 14.Sub theme: Resilience.

  Participants drawing on inner strength could reframe challenges positively, thereby alleviating depressive and SI.

  P10: “I made it through chemotherapy—this isn’t terminal. My daughter says there’s hope as long as I’m alive.”

  P8: “My daughter said she can’t lose me too. Thinking of my kids kept me fighting.”

  P9: “The doctors said with treatment I could have several more years. My granddaughter still wants me to cook for her someday. Maybe rehab can help my legs.” (Light laugh.)

Theme 3: driven by family responsibility

• 15.Sub theme: Intergenerational emotional connection.

  Functional decline often led to self-loathing and depression, yet family support helped restore a sense of purpose and reduce SI.

  P2: “My daughter begged me through tears, saying, ‘I’ve already lost my mother—I can’t lose you too.’”

  P5: “My granddaughter told me even if I can’t walk, I can still knit sweaters for her and be there at her wedding. It gave me a reason to keep going.” (Participant showed a knitted sweater.)

Integration of mixed research results

In the analysis of influencing factors, the results of quantitative research validate the core themes of qualitative research, while the sub-themes further supplement the results of quantitative research. For instance, in terms of vulnerability factors, qualitative research identifies social-emotional avoidance and avoidance of seeking medical treatment as refinements of high experiential avoidance. The stigma of suicide further extends this factor, with specific results shown in Fig. 2.

Fig. 2.

Integrated results from the mixed-methods study

Disucussion

High prevalence of SI among hospitalized older adults with functional impairment

This study is the first to examine the prevalence of suicidal ideation (SI) among hospitalized older adults with functional impairments. Using a mixed-methods approach grounded in vulnerability, risk, and protective frameworks, it provides a analysis of contributing factors beyond previous unidimensional assessments.

Findings revealed that the prevalence of SI in this population was 17.4%, which is notably higher than the 8.6% reported by Xiong et al. among general hospitalized older adults, a discrepancy likely attributable to our focus on functionally impaired individuals. Multiple studies have established a significant correlation between physical functional impairment and increased SI, with greater severity of disability being associated with higher SI [3]. However, this rate remains lower than the 32.8% observed in long-term care residents with functional impairments [7], possibly due to institutional factors such as restricted social interaction and limited family contact, which may intensify isolation [7, 20]. In contrast, hospitalized older adults often receive more consistent professional and social support, potentially buffering emotional distress [11]. The prevalence of suicidal ideation found in this study is also significantly higher than the 12% reported in European countries and the 17% in the United States, where comprehensive social welfare and accessible professional care may better mitigate the psychological impact of disability [21, 22]. In contrast, our rate is lower than figures often reported in other Asian countries, such as Japan and South Korea [23, 24]. Although these countries share Confucian values that emphasize family support, they experienced earlier and more rapid societal aging. Furthermore, having undergone modernization earlier, they have seen a more thorough breakdown of traditional family support, leading to stronger feelings of loneliness among the elderly, may contribute to our observed lower rate of suicidal ideation.

Vulnerability factors

The mixed-methods findings consistently identified poor economic status as a vulnerability factor for suicidal ideation (SI) among hospitalized older adults with functional impairments in Hunan Province, China. Although China has established a universal basic medical insurance system, disparities in coverage and reimbursement rates persist. Certain medications, long-term post-discharge rehabilitation exercises, and professional nursing care-related costs are often not fully covered [25], thereby imposing a substantial financial burden on older individuals with limited monthly income. Quantitative results further indicated that longer hospitalization duration was associated with higher levels of SI, as prolonged hospitalization may heighten uncertainty and fear regarding disease prognosis while increasing out-of-pocket expenses. Patients endure physical suffering, economic pressure, and psychological distress during treatment, compounding their sense of burden and hopelessness, thereby increasing SI risk [26]. Although China’s Long-Term Care Insurance (LTCI) covers rehabilitation and care costs, its pilot implementation remains limited to four cities in Hunan Province. Many families face barriers including low policy awareness and complex application procedures, restricting access to professional and financial support [27], Expanding LTCI coverage and enhancing primary care services, such as home-based nursing, health monitoring, and family caregiver training could reduce readmissions and alleviate economic strain [28].

Quantitative results indicated that higher experiential avoidance predicted SI. Functionally impaired older adults often adopt cognitive or behavioral withdrawal to evade negative experience [29]. While temporarily adaptive, chronic avoidance diminishes regulatory flexibility and fosters cognitive rigidity, intensifying hopelessness and SI [30, 31]. Qualitative findings revealed that some patients displayed medical avoidance through denial or distraction, while others expressed treatment futility beliefs, e.g., ‘I read online this illness can’t be cured.’ Such behaviors exacerbate disease progression and psychological distress. Clinicians should provide clear illness and treatment information, develop economically feasible care plans, and share recovery narratives to bolster confidence [32].

Risk factors

Quantitative analysis confirmed that greater severity of functional impairment significantly increases SI risk in older adults, aligning with existing evidence [33]. Severe disability leads to loss of autonomy, social isolation, and physical suffering, collectively reducing quality of life. Qualitative findings further identified disease burden, persistent pain, and treatment uncertainty as key contributors to hopelessness. Some patients also reported considering suicide as a means to seek familial attention, with inadequate family support intensifying despair [34].Therefore, primary interventions should focus on symptom management, such as alleviating pain and dyspnea, and reducing illness-related uncertainty through patient education [35]. Concurrently, actively fostering family support through regular visits and emotional interaction can strengthen social connectedness. However, caregivers themselves are at risk of psychological distress; thus, psychological support and respite services should also be provided to alleviate their burden.

Older adults with elevated depressive symptoms are also at increased SI risk, in line with findings that major depression predicts suicidality in this population [36]. Those with functional impairments often hyperfocus on physical suffering and interpret stressors through a negative cognitive lens, impeding adaptation and intensifying depressive affect. Interview participants also expressed that self-directed aversion and loss of dignity exacerbated SI. Loss of control over basic functions and dependence on nursing care evoke shame and helplessness [28], fostering low self-esteem and self-disgust, which may escalate into SI [37]. For this vulnerable subgroup, healthcare providers should facilitate processes that help reconstruct self-identity and enhance self-worth despite functional limitations [38].

Protective factors

Mixed-methods analysis identified resilience and social support as significant protective factors against SI in this population, consistent with prior evidence [11]. Older adults with higher resilience are better able to draw upon past experiences to derive a sense of meaning and agency, thereby counteracting feelings of hopelessness when confronted with disability-related distress. Rather than resorting to avoidance, they tend to engage with negative emotions through adaptive and constructive appraisal [39]. intergenerational emotional bonds emerged as a key protective mechanism, with participants often citing concerns about causing family grief as a deterrent to acting on SI. Neuroimaging studies support this observation, indicating that envisioning relatives’ distress activates the dorsomedial prefrontal cortex, which may inhibit amygdala-mediated despair and reduce suicidal intent [40].

Robust social support also buffers against SI by providing psychological resources to cope with illness and functional decline [41]. Reliable support networks alleviate loneliness and helplessness, while also fostering resilience through both psychosocial and neurobiological pathways: social connectedness can interact with genetic factors such as the 5-HTT genotype, promoting a neural environment conducive to resilience [42]. Thus, establishing a comprehensive social support system is essential. We recommend strengthening community-based primary care and expanding home- and community-centered long-term services. Concrete measures should include financial and material aid, such as home modification subsidies, assistive device support, and economic relief, to reduce family burdens [43].

Limitations

This study has several limitations. First, the cross-sectional design limits causal inferences regarding factors associated with SI. Longitudinal studies tracking SI dynamics during the transition from hospital to community are needed. Second, bedside interviews and audio recording in clinical settings may have induced social desirability bias, potentially affecting response authenticity. Third, although semi-structured interviews were suitable for this vulnerable population, more open formats in future studies could better capture participant-driven perspectives. Finally, the generalizability of findings may be limited as data came solely from Changsha. However, given the large population of older adults with functional impairments and their severe suicide risk, this research remains valuable for guiding future interventions and studies.

Conclusion

The study reveals a high prevalence of suicidal ideation (SI) among hospitalized older adults with functional impairment, influenced by vulnerability, risk, and protective factors. Clinicians should prioritize patients with increased vulnerability such as low income or high experiential avoidance and promptly identify risk factors including poor family relationships, severe disability, and depressive symptoms. Meanwhile, reinforcing protective factors, such as expanding long-term care insurance coverage and enhancing psychological resilience, is essential for tailored interventions. Healthcare administrators should establish structured suicide prevention protocols to reduce SI and improve quality of life in this population.

Authors’ contributions

QT and XFW designed the study and developed the structured questionnaires. RHW, HYL, JHZ, XPX, and PPH were responsible for data collection and analysis. The initial draft of the manuscript was written by QT, with XFW and Pingping He providing critical revisions and substantial intellectual input to subsequent versions. All authors reviewed and approved the final manuscript.

Funding

This study was supported by the Scientific Research Fund of Hunan Provincial Education Department-Basic Education (Y20230728); the Scientific Research Fund of Hunan Provincial Education Department-Higher Education (202401000501); and Ministry of Education Industry-University Cooperative Education Program (241200541113826).

Data availability

The datasets used or analyzed during the current study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Biomedical Research Ethics Committee of Hunan Normal University (Approval No. 2024 − 172) on March 11, 2024. In accordance with the principles of the Declaration of Helsinki, written informed consent was obtained from all participants prior to enrollment.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.