Burden of caregivers of older people in Sri Lanka: an overlooked concern in the health care system

Abstract

Background

Caring for older people has become a significant public health concern in Sri Lanka due to the growing aging population. This has placed a heavy burden on family caregivers, particularly those caring for older individuals with multiple chronic conditions. Recognizing this challenge, the present study aimed to evaluate the psychometric properties of the Sinhala version of the 10-item short form of the Burden Scale for Family Caregivers (BSFC-s) and assess caregiver burden and associated factors among caregivers of older people aged over 65 years with multimorbidity.

Methods

The BSFC-s was cross-culturally adapted following standard guidelines and administered to consecutively selected 178 family caregivers involved in the long-term care of older patients (aged ≥ 65 years) who were regular attendees of medical and neurology clinics at the National Hospital Galle, Sri Lanka. The validated Sinhala version of the Short Form-36 questionnaire (SF-36) was also administered simultaneously. After two weeks, 60 caregivers were re-assessed using the BSFC-s. The psychometric properties, including reliability and validity, as well as floor and ceiling effects, were evaluated. The final version of the BSFC-s was then administered to 270 informal family caregivers in the Gampaha District, Sri Lanka.

Results

The study revealed that the internal consistency of the Sinhala version of the BSFC-s, measured by Cronbach’s alpha, was 0.90, with item-total correlations ranging from 0.38 to 0.82. The test-retest reliability, assessed using the intra-class correlation coefficient, was 0.99 (95% CI: 0.99-1.00; p < 0.001). Principal component analysis revealed two factors explaining 65.8% of the cumulative variance. BSFC-s scores showed a strong negative correlation with SF-36 scores (r = -0.81, p < 0.001), indicating a good concurrent validity. Neither floor nor ceiling effects were observed. Among the 270 family caregivers, 78.1% reported a high level of burden, while 21.1% reported moderate burden. Caregiver burden was significantly associated with low educational level (p = 0.032), low monthly income (p = 0.041), and unemployment status (p < 0.001) of the caregiver.

Conclusion

The Sinhala version of the BSFC-s demonstrated strong reliability and validity, making it a suitable tool for assessing caregiver burden in Sri Lanka. Most caregivers reported high levels of subjective burden, particularly those with lower socio-demographic status.

Trial registration

Not applicable.

Background

Population aging is a demographic phenomenon resulting from declining fertility rates and increasing longevity. While this demographic transition is global, it is more pronounced in low- and middle-income countries [1, 2]. Compared to younger individuals, older people are more prone to get impairments in physical, psychological, and social functions, and are therefore more likely to be frail and dependent. Apart from age-related bodily changes, multimorbidity and polypharmacy, which are prevalent in old age, also contribute to frailty and dependency observed in this population. These changes create an additional burden on caregivers. The relationship between the caregiver and care recipient is a crucial factor in determining caregiver workload and subjective burden [3].

“Caregiver’s subjective burden” is defined as a person’s subjective self-evaluation of feeling burdened. It is associated with various negative outcomes, such as health issues, increased mortality risk, and a higher risk of institutionalization for the debilitated individual, as well as influencing the caregiving style. The Burden Scale for Family Caregivers (BSFC) provides vital information about the negative aspects associated with caregiving and how providing care affects the caregiver’s health and quality of life [4, 5]. The short form of the BSFC (BSFC-s), consisting of 10 items, is a condensed version of the BSFC and provides a quick and practical way to assess the subjective burden of caregivers [6]. The stressful aspects, such as coping and social support including background and content, crucial stressors, secondary role strains, secondary intra-psychiatric pressures, and outcomes of caregiving are evaluated using a 4-point scale with the following values: “strongly disagree” (0), “disagree” (1), “agree” (2), and “strongly agree” (3) [6]. The items of the BSFC-s have been reported to adhere to a one-factor structure, with a Cronbach’s alpha of 0.92 for the entire scale [6]. Subsequent validations have reported adherence to a two-factor structure as well [7].

Family caregivers of older people have reported a higher burden than caregivers of individuals with other disabilities [8]. They play a vital role in providing long-term home care for disabled older people without remuneration, offering physical, emotional, social, and financial support to the best of their ability. In most cases, the primary care is provided by a close family member of the disabled older person. The majority of caregivers for disabled older individuals are typically their spouses and children. Accordingly, informal caregivers are often untrained and inexperienced.

The caregiver’s burden is associated with both the characteristics of the care recipient and the caregiver [9]. A heavier burden is associated with lower functional levels, chronic health issues, lack of education, poor cognitive status, and psychological health problems of care recipients. Caregiver factors such as older age, female gender, lower levels of education, poorer health, and living with the care recipient are also closely linked to a greater level of caregiver burden [10]. Furthermore, psychological characteristics, caregivers’ belief systems, social support, familial cultural identity, familial faithfulness in adult children, and a sense of acceptance of the caregiver role and responsibility have been associated with a lower burden [10].

Sri Lanka has the highest percentage of older people in South Asia. Furthermore, the country’s older adult population is the third largest in Asia, following Japan and Singapore [11, 12]. With the expansion of its older adult population, Sri Lanka’s health and social care systems need to adopt new strategies to meet the growing demands of older people. Due to religious and cultural values, the care of older people is currently shared among family members. However, the shift from extended to nuclear families and the higher rate of female employment are likely to undermine these traditional family values, potentially making the care of older people an additional burden for families. Therefore, assessing the burden placed on caregivers of older people is of utmost importance.

In Sri Lanka, there is a scarcity of coordinated research in this area, partly due to the low priority given to the subject and the lack of valid and reliable measurement tools to assess caregiver burden. Assessing caregiver burden will help healthcare providers take steps to mitigate it by offering community support, respite care, and other interventions. A culturally adapted version of the BSFC-s will address this gap and encourage further research in this area.

Therefore, this study aimed to evaluate the validity of the Sinhala version of the BSFC-s using a group of family caregivers of individuals aged over 65 years with multi-morbidity attending medical clinics at the National Hospital Galle, Sri Lanka. Additionally, it aimed to assess caregiver burden and associated factors among a group of informal family caregivers selected from various geographical locations in the Gampaha District, Sri Lanka.

Methods

This study comprised two phases: the validation of the Sinhala version of the BSFC-s and the evaluation of family caregivers’ burden using the validated tool.

Phase 1

Study design

This validation study was conducted from January to December 2021.

Cross-cultural adaptation

The cross-cultural adaptation process closely adhered to the standard guidelines outlined by Beaton et al., following a structured and methodical approach [13]. The first step, forward translation, involved translating the original English version into Sinhala by two independent health professionals. This ensured that the nuances of the original language were preserved while capturing the cultural essence of the target language.

Next, the investigators synthesized these translations into a single, cohesive version. This step was essential for creating a unified scale that harmonized the interpretations of the two translators. The synthesized Sinhala version was then subjected to back translation, where two other independent health professionals retranslated it into English. This step ensured that the translated version remained consistent with the original, identifying any discrepancies or conceptual differences.

An expert group, consisting of three physicians, two nursing academics, and the forward and backward translators, meticulously reviewed all versions. Their goal was to ensure semantic, idiomatic, experiential, and conceptual equivalence between the original and the translated version. This phase also involved confirming content validity, resulting in the creation of a pre-final Sinhala version.

The pre-final version was then subjected to a trial phase for further refinement. Ten family caregivers of older patients from the medical wards at the National Hospital in Galle were interviewed to assess the clarity, understandability, and naturalness of the items. Based on their feedback, the final version was pretested among another 20 family caregivers to ensure its face validity. This thorough process ensured that the final version was not only linguistically accurate but also culturally appropriate and understandable to the target audience.

Administration of questionnaire

The finalized Sinhala version of the BSFC-s was administered to 178 family caregivers of older patients (> 65 years) who were regular attendees of the medical and neurology clinics at the National Hospital, Galle. Consecutive family caregivers who accompanied their older debilitated family members (patients) to the clinic were invited to participate in the study. Older patients with moderate to severe dependency in physical activities, as measured using the Barthel Index, were included. In total, 254 family caregivers were evaluated, with 178 caregivers providing care to older individuals with moderate to severe physical dependency. The validated Sinhala version of the Short Form-36 questionnaire (SF-36) [14] was also administered simultaneously.

During the administration of the questionnaire, the trained interviewer read the questions aloud to the caregiver and recorded their responses through direct interaction. To maintain consistency in the data collection process, no further explanations or rewording of the statements were provided. The BSFC-s is a 10-item instrument for measuring subjective burden in informal caregivers. Each item is a statement rated on a 4-point scale with the values: “strongly disagree” (0), “disagree” (1), “agree” (2), and “strongly agree” (3) [6]. A high degree of agreement indicates a higher subjective burden for the caregiver. The scoring ranges are: 0–4 (none/low burden), 5–14 (moderate burden), and 15–30 (severe/very severe burden) [6].

The Short Form-36 survey (SF-36) assesses the health-related quality of life (HRQoL) of individuals based on 36 items. It measures eight health domains: physical functioning, role limitation due to physical problems, bodily pain, general health perception, vitality, social functioning, role limitation due to emotional problems, and mental health. In this questionnaire, each domain is assigned a score ranging from 0 to 100 using the original coding algorithm [15].

The minimum required sample size for the study was determined by multiplying the number of variables in the BSFC-s (10 items) by 15 [16], and then adding an additional 50% to account for non-respondents and incomplete questionnaires. This method of sample size calculation was used to ensure that the study had adequate statistical power and validity in measuring the constructs of interest. Furthermore, sampling adequacy was assessed in the Factor Analysis (FA) using the Kaiser-Meyer-Olkin (KMO) measure [17].

Moreover, a subgroup of 60 randomly selected subjects completed the Sinhala version of the BSFC-s two weeks after the initial administration, under the same conditions.

Statistical analysis

The basic characteristics of the subjects who participated in the evaluation of psychometric properties were presented as frequencies (percentages) or mean (SD). Psychometric properties (reliability and validity) were evaluated. Data analysis was conducted using SPSS version 25.0. The Kolmogorov-Smirnov test was performed to assess the distribution of the data, revealing that the dataset was not normally distributed (p < 0.001). Therefore, the data were log-transformed to achieve a near-normal distribution, allowing the use of parametric tests in the analysis. Statistical significance was considered at a p-value of < 0.05.

Psychometric properties analysis

The reliability and validity of the BSFC-s were rigorously assessed using several statistical methods.

Test-retest reliability was assessed using the intra-class correlation coefficient (ICC), comparing the overall scores from two consecutive administrations of the BSFC-s to ensure consistency over time [17]. Internal consistency was assessed using Cronbach’s alpha, which measured the reliability of the overall scale. A Cronbach’s alpha value of 0.7 or higher was considered acceptable, indicating a satisfactory level of internal consistency [17]. Construct validity was evaluated using both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA). EFA was conducted using Principal Component Analysis (PCA) with Varimax rotation and Kaiser normalization. This process helped determine whether the underlying item structure aligned with the intended design of the instrument [17]. The KMO measure and Bartlett’s test of Sphericity were applied to assess sampling adequacy and the appropriateness of FA. The correlation matrix was also examined to further explore item relationships. The number of components extracted was determined based on several criteria: the Scree plot, the percentage of variance explained by each component, and the Kaiser-Guttman rule, which includes components with Eigenvalues greater than one. Items were considered representative of a component if their individual item loading was ≥ 0.40 [17].

CFA was conducted using AMOS 23 software. The root mean square error of approximation (RMSEA) and comparative fit index (CFI) were examined to assess the model’s fit. The cut-off values for acceptable model fit used for this study were RMSEA < 0.08 for a good fit; CFI ≥ 0.90 for an acceptable fit [18]. Path diagram of the model was drawn.

Concurrent validity was assessed using Pearson’s correlation, comparing the overall scores of the BSFC-s with the domain scores of the SF-36 to determine the degree of correlation between the two measures [17].

Finally, floor and ceiling effects were assessed to evaluate the distribution of scores. These effects were considered significant if more than 15% of the population had either the lowest (floor) or the highest (ceiling) score, which could indicate a lack of sensitivity in the scale for that portion of the population.

Phase 2

Study design, sample and setting

A community-based cross-sectional study was conducted from January to December 2023 in three selected Divisional Secretariat (DS) Divisions of the Gampaha district, Sri Lanka, including Negombo DS Division, Minuwangoda DS Division, and Dompe DS Division, which respectively represent urban, semi-urban, and rural communities.

Older patients with debilitation (bedbound) were identified, and information was gathered from the Social Service Officer assigned to the selected DS areas in the Gampaha district, with permission from the Divisional Secretary. Subsequently, the investigator visited the households after arranging appointments with the caregivers. Only those caregivers who expressed their willingness to participate in the study were visited, and this process continued until the required sample size was achieved.

Main family caregivers of chronically ill, disabled, or bed-bound (≥ 3 months) older family members (≥ 65 years) participated in the study (n = 270). Family members who were only temporarily looking after the patients were excluded.

The minimum required sample size was calculated using n = z² p (1 – p)/d² equation [19], with the 81.1% of caregiver burden reported [20].

Data collection and tools

An interviewer-administered questionnaire, consisting of a demographic questionnaire and the validated Sinhala version of the BSFC-s, was used. Socio-demographic information of both the caregiver and the care receiver was collected, including details such as age, gender, marital status, and the relationship between the caregiver and care receiver. To ensure consistency in data collection, the same interviewer-administered approach used in Phase 1 was employed.

Statistical analysis

The Kolmogorov-Smirnov test was conducted to assess the distribution of the data, revealing that the dataset was not normally distributed (p < 0.001). Consequently, a log transformation was applied, resulting in a distribution closer to normal.

Descriptive statistics were presented using the mean, standard deviation, frequencies, and percentages. To identify the factors associated with caregiver burden, parametric tests (Independent Sample t-tests and One-way ANOVA tests) were applied. All data analysis was conducted using SPSS version 25.0, with statistical significance set at p < 0.05.

Ethical clearance

Ethical clearance for the study was granted by the Ethics Review Committee of the Faculty of Medicine, University of Ruhuna, Sri Lanka (Ref No: 2020.P.122, dated 22.10.2020). Prior to administering the questionnaire, written informed consent was obtained from all participants in both phases. The purpose and procedures of the study were clearly explained to participants in both phases, ensuring their understanding and voluntary participation. Administrative clearance was obtained from the Director of the National Hospital Galle for Phase 1 and from the Divisional Secretaries of the respective DS areas (Negombo, Dompe, and Minuwangoda) for Phase 2 prior to commencing data collection.

Results

Phase 1

Basic characteristics of caregivers

The mean (SD) age of the participants was 50.9 (16.8) years, with the majority being female (56.7%), earning a monthly income of less than 20,000 LKR (59.6%), educated up to the GCE Ordinary Level (69.1%), and married (80.3%). The majority (59.5%) provided care for one of their parents, while 29.3% were caring for a spouse. In 80.9% of instances, the caregiver and care receiver lived in the same household (Supplementary Table 1).

The mean (SD) scores of the Sinhala version of the BSFC-s and SF-36 were 11.7 (5.8) and 63.5 (15.3), respectively. Based on the BSFC-s score, the majority of caregivers experienced moderate caregiver burden (61.8%, n = 110), while the rest reported mild (19.1%, n = 34) or severe (19.1%, n = 34) caregiver burden.

Furthermore, higher mean (SD) scores were observed in the four items, “My life satisfaction has suffered because of the care” 1.39 (0.89), “Since I have been a caregiver my financial situation has decreased” 1.38 (0.81), “My relationships with other family members, relatives, friends and acquaintances are suffering as a result of the care 1.29 (0.90) and I often feel physically exhausted 1.20 (0.82) (Table 1).

Table 1 Distribution of burden of informal caregivers (n = 178)

Psychometric properties of BSFC-s

Reliability

The internal consistency of the scale, measured by the overall Cronbach’s alpha, was 0.90, with item-total correlations ranging from 0.38 to 0.82. The test-retest reliability, evaluated with the ICC (95% CI), was 0.99 (0.98-1.00) (p < 0.001).

Construct validity

The EFA with PCA revealed the presence of two factors with eigenvalues exceeding 1 (Fig. 1), explaining 65.8% of the cumulative variance (with each factor accounting for 33.7% and 32.1%, respectively). The factor loadings were high, ranging from 0.59 to 0.92. Factor extraction is shown in Table 2.

Fig. 1

Scree plot of factor extraction

Table 2 Rotated component matrix of FA

The CFA yielded a chi-square goodness-of-fit p-value of < 0.001, indicated that there is a significant difference between the hypothesized model and the observed data. The CFI and RMSEA values were 0.751 and 0.231, respectively, suggesting that the model does not fit well with the gathered data well. The corresponding path diagram is presented in Fig. 2.

Fig. 2

Path diagram of the model

Concurrent validity

The Sinhala version of the BSFC-s score showed a strong correlation with the overall SF-36 score (r = -0.81, p < 0.001) and moderate to strong correlations with the domain scores (r = -0.38 to -0.63, p < 0.001) (Table 3).

Table 3 Correlation between BSFC-s score and SF-36 scores

Floor and ceiling effect

Six (3.3%) subjects had the lowest score (0 points), while none had the maximum score (30 points). Therefore, neither a floor nor a ceiling effect was found.

Phase 2

Socio-demographic characteristics of the family caregivers

The mean (SD) age of the family caregivers was 51.5 (13.0) years. Among the participants, 202 (74.8%) were female family caregivers, and nearly half of them, 118 (43.7%), were aged between 46 and 60 years. The majority of the study participants, 159 (58.9%), were married, and 40.7% (n = 110) had only completed the Ordinary Level of education. Additionally, 92 out of 270 participants (34.1%) were unemployed, and 87 (32.2%) were engaged in part-time jobs. Regarding income, 39.6% (n = 107) had a gross monthly family income between 10,000 and 29,999 LKR. Nearly three-quarters of the participants, 185 (68.5%), were children of older people, and the majority, 234 (86.7%), lived with the older individuals in the same household (Supplementary Table 2).

Socio-demographic characteristics of the care recipient (older people)

The mean (SD) age of the care recipients was 82.1 (9.9) years. Among them, 92 (34.1%) were in the 85–94 years age category, and 159 (58.9%) were females. All 270 older individuals were suffering from one or more morbidities that resulted in bed confinement. Of these, 100 (37.0%) experienced age-related morbidities, while 80 (29.6%) suffered from neurological-related conditions such as paralysis, spinal cord injuries, dementia, and other neurological disabilities. The majority of older individuals received Allopathic treatments (n = 161, 59.6%) (Supplementary Table 3).

Caregiver burden

The mean (SD) caregiver burden score was 17.73 (4.21). The caregiver burden for each item is shown in Table 4. High mean scores were seen in the items indicating the financial burden (Since I have been a caregiver my financial situation has decreased), psychological burden (Sometimes I don’t really feel like “myself” as before) and physical burden (I often feel physically exhausted) respectively. The majority of participants agreed or strongly agreed with the given statements.

Table 4 Caregiver burden of family caregivers (n = 270)

The majority of participants reported a high level of burden (n = 211, 78.1%), while 57 caregivers experienced moderate subjective burden (n = 57, 21.1%), and a smaller number of caregivers reported low subjective burden (n = 2, 0.7%).

Factors associated with caregiver burden

None of the characteristics of care recipients were associated with caregiver burden (p > 0.05). However, there was a significant difference in caregiver burden across the subgroups of education level (p < 0.032), gross monthly income (p = 0.041), and employment status (p < 0.001) of the caregivers. Higher burden was observed among caregivers with lower education (ordinary level education) (18.39 ± 4.02), unemployment (19.05 ± 3.25), and low monthly income (< 10,000 LKR) (18.49 ± 3.48) (Table 5).

Table 5 Association between caregiver burden with care recipient factors and caregiver factors (n = 270)

Discussion

The current study demonstrates that the cross-culturally adapted Sinhala version of the BSFC-s is a valid tool for measuring subjective burden in Sinhala-speaking informal caregivers in Sri Lanka. The findings reveal that caregiver burden among informal family caregivers is notably high, and this burden is significantly associated with low socioeconomic status, including lower education levels, unemployment, and lower monthly income.

Validity of the scale

The process of cross-cultural adaptation was carefully conducted following a standard protocol to ensure the Sinhala version of the BSFC-s accurately reflects the original scale’s conceptual qualities. Throughout this process, we focused on achieving semantic, idiomatic, experiential, and conceptual equivalence with the original version. While not quantified, content and face validity were ensured through extensive discussions with an expert group and feedback from a focus group of caregivers. This rigorous adaptation process ensured that the scale captured the multifaceted nature of caregiver burden, including physical, social, psychological, and economic dimensions, in a way that is culturally relevant and meaningful. As the scale addresses various aspects of caregiving stress, such as coping strategies, social support, primary stressors, secondary role strains, and psychological impacts, it provides a robust estimation of subjective caregiver burden, encompassing the diverse challenges caregivers face.

The FA of the Sinhala version of the BSFC-s revealed a two-factor structure, with Eigenvalues greater than 1.0 and all 10 items demonstrating factor loadings above 0.59. This finding contrasts with the original one-factor structure observed in the initial version of the scale. Despite this deviation, the internal consistency of the overall scale was high, with a Cronbach’s alpha of 0.90, and the item-total correlations for individual items were also notably high. These results suggest that the items collectively measure a unified construct of “total subjective burden.” Even with a single item, the scale successfully captures the overall subjective caregiver burden. Therefore, it can be concluded that all items in the scale contribute significantly to the assessment of the total subjective caregiver burden, ensuring a comprehensive evaluation of the caregiver’s experience.

The CFA suggested that the model does not fit well, possibly due to factors such as the non-normal distribution of the data. However, this does not limit the use of the tool in the local context, as the tool has demonstrated acceptable validity and reliability through EFA, proving that it is meaningful and relevant to the local population.

The majority of caregivers in the study group reported experiencing a moderate burden, likely associated with providing care to individuals with moderate physical dependency. This indicates that the scale effectively captures the total subjective burden faced by caregivers, particularly in terms of the presence and severity of caregiver burden. Apart from that the higher mean scores observed in the items such as “My life satisfaction has suffered because of the care”, “Since I have been a caregiver my financial situation has decreased”, “My relationships with other family members, relatives, friends and acquaintances are suffering as a result of the care and I often feel physically exhausted. These four items generally reflect the financial, psychological, social, and physical aspects of burden, respectively. Therefore, we can assume that the scale effectively evaluates all aspects of a caregiver’s burden across a broad spectrum.

Furthermore, the scale’s completion time was brief, making it suitable for use in busy clinical settings. We assumed that the actual caregiver burden is a predictor of HRQOL, and this concept was used to assess the concurrent validity of the scale. We observed a strong correlation between the total scores of the Sinhala version of the BSFC-s scale and the overall SF-36 score, as well as its individual domains. Among the individual domains of the SF-36, the HRQOL specific to the psychological health domain showed stronger correlations with the burden scale score compared to the other domains.

The findings of the current study are largely consistent with those of the original validation study, which reported a reliability (Cronbach’s alpha) of 0.92 [6]. Furthermore, Pendergrass et al. demonstrated strong convergent and discriminant validity, along with a Cronbach’s alpha of 0.92, confirming that the BSFC-s is a valid tool for measuring subjective burden in informal caregivers [21].

Previous studies have shown varying factor structures of the scale in different ethnic groups. Konerding et al. [7] also found a two-factor model with English, Finnish, and Greek caregivers. In the same analysis, English caregivers were more likely to endorse items related to impairments in individual well-being, while Finnish caregivers showed a stronger tendency to endorse items related to conflicts. Greek caregivers in the same study showed a stronger tendency to endorse items addressing impairments in physical health, indicating that the factor structure is likely to vary based on ethnicity. In contrast, in our study, the financial, psychological, social, and physical aspects were more frequently endorsed.

Although many scales, such as the Zarit Burden Interview [22], Modified Caregiver Strain Index [23], and Caregiver Difficulties Scale [24], are used to assess the subjective burden of informal caregivers, the BSFC-s offers several advantages. These include a shorter structure, a more comprehensive assessment of multiple aspects of caregiver burden, and easier interpretation. As a result, it has been translated into 20 different languages and is used worldwide (www.caregiver-burden.eu).

A reliable measure of subjective caregiver burden is essential for both clinical practice and research. It helps in the early detection of caregivers with high burden, who may be vulnerable to physical and mental health instability. Furthermore, the scale can be used in caregiver counseling, training family caregivers, and supporting self-help groups.

Caregiver burden

In the second phase of the study, we observed that a majority of participants reported a high level of caregiver burden, which was closely linked to their low socio-economic status. Caregivers with low income, limited education, and unemployment were predominantly affected, creating a vicious cycle. The ongoing economic crisis in Sri Lanka exacerbates these challenges. As the country’s financial situation worsens, families’ monthly incomes have decreased, particularly in poorer communities, while healthcare and living expenses have risen. This economic instability directly impacts families caring for debilitated older individuals, placing additional strain on their well-being. In some cases, caregivers have been forced to leave their jobs due to caregiving demands, further compounding their financial difficulties.

Moreover, caregivers are often the children or spouses of older individuals, and many have sacrificed their employment to provide full-time care. As a result, they experience heightened psychological and economic burdens. Additionally, caregivers with lower educational backgrounds tend to have diminished coping abilities, resilience, and competence in caregiving, which increases their overall burden and further impairs their well-being.

Many caregivers experienced physical, psychological, and economic strain, which may lead to poor health outcomes, psychiatric conditions, financial instability, and disrupted family relationships. This holistic impact on caregivers can ultimately damage the relationship between the caregiver and the care recipient, potentially escalating to violence or neglect of the care recipient.

In Sri Lanka, where caring for older people, especially parents and grandparents, is highly valued, failing to provide proper care is often seen as a social failure, and caregivers are blamed as a result. This cultural expectation forces many caregivers to bear the burden, despite facing significant challenges, which ultimately leads to a poor quality of life. These are plausible explanations for the high burden observed among caregivers, although they are often reluctant to express it openly.

Studies from Thailand [20] and Nepal [9] have shown a high level of caregiver burden; however, the current study shows a significantly higher burden than these countries. Owing to the Asian culture, which values the care of older people, this could be one reason for similar findings. Furthermore, the lack of professional knowledge and hands-on skills in elderly care, inadequate family support, insufficient professional social support, the tendency to hide stressors due to social stigma, and the increased stress levels of caregivers might be other contributing factors.

We identified low education, unemployment, and lack of education as factors associated with a high level of caregiver burden. A study conducted in Thailand identified significant predictors of caregiver burden, including the caregiver’s age, education level, relationship to the care recipient, and income level [20]. A study conducted in China identified that those who reported caregiver stress were more likely to experience insufficient caregiving abilities, economic challenges, or scheduling conflicts [25]. These study findings highlight that individuals with low socioeconomic status often experience a higher burden. Therefore, measures to enhance the well-being of caregivers must be given greater focus.

Another study from China found a lower level of caregiver burden in contrast to the current study. Social support has been identified as a predictor of lower caregiver burden, potentially reducing depressive symptoms among caregivers [26]. This is an imperative aspect to reconsider in the Sri Lankan context in empowering caregivers.

Strengths, limitations and recommendations

The current validation study includes an adequate sample size determined by statistical calculation, and the study participants were from both rural and urban populations in Southern Sri Lanka. Since the National Hospital Galle is the largest tertiary care hospital in Southern Sri Lanka, providing healthcare to the entire region, patients and their caregivers from different socioeconomic backgrounds were included in the study. Furthermore, the focus of caregiving extends beyond older adults; therefore, the applicability of the BSFC-s should extend to other situations, such as caregivers of people with dementia and children with disabilities, among other important caregiver groups. Furthermore, cross-validation of the tool by testing it with another independent sample from the same population is also important, since the tool has shown poor model fit in CFA.

The assessment of caregiver burden would benefit from a larger sample size and serial evaluations conducted at different time points during caregiving in a longitudinal study. Additionally, a qualitative approach could be employed to further explore caregivers’ perspectives and lived experiences, providing deeper insights into their challenges and inner views. These considerations highlight some limitations of the second phase of the study and suggest potential directions for future research. However, our study was conducted in a representative population, covering rural, semi-urban, and urban communities. The findings, which reveal a high level of caregiver burden, emphasize the urgent need for targeted interventions to support caregivers, especially those from lower socio-economic backgrounds, to break the cycle of burden and enhance both their quality of life and the care they provide. Addressing emotional and psychological stress, as well as physical strain, would be significantly improved through supportive actions. Authorities should focus on increasing caregivers’ self-efficacy by providing them with adequate knowledge, skills, and social support.

Furthermore, it is essential for the country’s health system to adopt a “care for caregivers” approach, which includes enhancing social support, monitoring caregivers’ health, and implementing structured interventions to reduce their burden and alleviate depressive symptoms. Ensuring that caregivers are respected and not neglected is crucial for improving their well-being and the overall quality of care they deliver. In addition to the current scope of the study, it would have been valuable to examine care-related factors such as the duration of caregiving, perceptions of caregiving responsibilities, the impact of caregiving on daily life, sources of caregiving stress, and the availability of caregiving resources. Thus, a more comprehensive study that includes these aspects is recommended for a deeper understanding of the caregiving experience.

Conclusions

The cross-culturally adapted Sinhala version of the BSFC-s demonstrated satisfactory reliability and validity, indicating its suitability for assessing the subjective burden of caregivers of older people in Sri Lanka. It is a feasible and economical measure to estimate the total subjective burden of caregivers, which can be used in both research and clinical practice to enhance the quality of life of both informal family caregivers and care recipients. The high burden of caregivers is associated with poor socio-economic status. This finding helps the country’s health system identify individuals at high risk of burden and facilitates the implementation of measures aimed at empowering caregivers, the silent heroes of the health system.