Authors: Patama Vapattanawong, Ruttana Phetsitong
Categories: Research, Hearing impairment, Visual impairment, Sensory impairments, Dual sensory impairments, Caregiver burden, Older adults, Thailand
Source: BMC Geriatrics
Authors: Patama Vapattanawong, Ruttana Phetsitong
Visual and hearing impairments can increase dependence across the life course; however, their impact may be particularly pronounced in older adults due to age-related functional decline and increased care needs. This study examined how caregiver burden differs by type of sensory impairment, including vision impairment, hearing impairment, and dual sensory impairment.
A cross-sectional secondary data analysis was performed. Data were obtained from the 2021 Survey of Older Persons in Thailand, conducted by the National Statistical Office. The study sample comprised 3,190 pairs of older adults aged 60 and above and their primary caregivers. Visual and hearing impairments were defined based on self-reported functional difficulty with or without assistive devices. Caregiver burden was assessed using a self-reported single-item question on a 0–10 scale, with a higher score indicating greater burden, and was recoded into three no burden (score 0), low burden (scores 1–3), and high burden (scores ≥ 4), based on the median score. Generalized ordinal logistic regression, adjusted for caregiver background, older adults’ health status, physical environment, and socioeconomic factors within the older adults’ household, was used to examine the relationship between caregiver burden and sensory impairments.
Among older adults, 13.3%, 11.9%, and 27.0% had visual impairment, hearing impairment, and dual sensory impairment, respectively. Median burden score was 3 (IQR: 0–5), indicating low burden overall. Compared to caregivers of older adults without sensory impairments, those caring for individuals with hearing impairment were more likely to report higher levels of burden (AOR 1.55, 95% CI 1.24–1.94). Elevated odds of burden were also observed for caregivers of older adults with dual sensory impairments (AOR 1.29, 95% CI 1.08–1.54) and vision-only impairment (AOR 1.26, 95% CI 1.02–1.56).
Hearing impairment was associated with elevated caregiver burden, with the association being stronger than that observed for dual sensory impairments. This counterintuitive finding may reflect communication barriers, social isolation, and the less obvious nature of hearing loss, but it should be viewed with caution because of unmeasured factors like impairment severity and support availability.
The online version contains supplementary material available at 10.1186/s12877-026-07352-7.
Globally, the population aged 60 and over is projected to double by 2050, rising from 13.5% in 2020 to 22% [1]. A combination of longer life expectancy and declining fertility rates drives this trend. The growing aging population worldwide, including in Thailand, not only boosts the demand for healthcare services but also raises care needs. The World Health Organization reported that about two-thirds of older adults are likely to require long-term support and care from others to perform daily activities [2].
Caregivers play an essential role in supporting the health and well-being of older adults. Evidence suggests that caregiver characteristics and caregiving circumstances greatly influence care recipients’ outcomes [3, 4]. While caregiving can bring emotional fulfillment and personal meaning, providing opportunities for closeness and assurance of quality care [5, 6], it is also associated with substantial physical, psychological, and financial challenges [7, 8]. Persistent caregiving stress and perceived burden may lead to caregiver role strain or, in severe cases, increase the risk of older mistreatment [9, 10].
In Thailand, the need for care in households with older adults has been increasing [11]. Caring for older adults, especially those with chronic illnesses or disabilities, mainly falls on family members, with women, particularly daughters or daughters-in-law, providing most of the care [12–14]. This system is influenced by cultural norms of filial piety and intergenerational support, and in rural areas, it is further strengthened by limited access to formal care services [13, 15]. Many of these caregivers must balance employment and household tasks alongside their caregiving responsibilities, which imposes a significant burden [12].
Caregiver burden includes the stresses and challenges that arise when caring for chronically ill, disabled, or older adults. This covers physical, mental, emotional, social, and financial stress [16]. The burden varies depending on demographic, socioeconomic, health, and environmental factors, affecting both the caregiver and the older individual receiving care. Previously, research on caregiver burden has concentrated on caring for older adults with chronic illnesses or non-communicable diseases such as dementia, Alzheimer’s disease, and cancer [3, 16, 17], in both clinical and community settings. While caregiver burden is a widely studied construct, this study acknowledges critiques that this term may pathologize caregiving by implying the care recipient is “a burden.” Caregiving encompasses both challenges and rewards, including strengthened bonds and fulfillment of cultural values. This is particularly relevant in Thailand, where family caregiving is deeply rooted in filial piety and considered a normative responsibility. Nevertheless, understanding caregivers’ subjective feelings of difficulty is essential for identifying families who may need support. This study use of “caregiver burden” refers specifically to caregivers’ self-reported perceptions of difficulty rather than objective caregiving demands.
Besides chronic illnesses, visual and hearing impairments are among the most common disabilities in older adults [18, 19]. In Thailand, community-based studies have reported that nearly one in five older adults experience visual impairment, mainly caused by treatable conditions such as cataracts and refractive errors [20, 21]. Similarly, the occurrence of hearing impairment increases significantly with age, affecting more than half of older adults in some regions [22, 23]. These sensory limitations can reduce mobility, independence, and social participation, often leading to increased caregiving needs [24].
Most existing evidence on sensory impairment and caregiver burden has been generated in Western contexts, where formal care services are more widely available. In non-Western settings such as Thailand, caregiving is predominantly provided by family members at home and is shaped by cultural norms of filial responsibility, intergenerational co-residence, and limited access to formal long-term care. Although the prevalence and impact of visual and hearing impairments among older adults in Thailand are increasingly recognized, research comparing the specific caregiver burdens associated with each type of sensory impairment remains limited. Current studies mainly focus on the clinical or epidemiological aspects of visual or hearing loss separately, with limited exploration of how these impairments affect caregivers’ daily tasks, stress levels, and overall well-being. Additionally, in the Thai context, where caregiving is primarily carried out by family members at home, understanding whether specific impairments result in significantly higher caregiving demands is essential for informing culturally appropriate interventions. Therefore, this study aims to assess the level of caregiver burden and examine whether and how the burden differs when caring for older adults with varying levels of visual and/or hearing impairments. Addressing this gap is crucial for developing targeted caregiver support programs, shaping health policy, and guiding resource allocation to improve outcomes for both older adults and their caregivers.
This study was a cross-sectional analysis, utilizing secondary data from the 2021 Survey of Older Persons in Thailand (SOPT), a large-scale, nationally representative household survey conducted by the National Statistical Office (NSO). The survey employed a stratified multi-stage sampling design to ensure national representativeness across Thailand’s regions and urban/rural areas. It covered 83,880 sample households. Key data on all pre-older adults (aged 50–59) and older adults (aged 60 and above) in each household, including demographic and socioeconomic characteristics, health status, healthcare behaviors, financial support, and access to food, clothing, and supplies, were collected. The survey also gathered data on caregivers of older adults from the screening question that asked each older adult, “Who is currently the main caregiver providing your daily care?” The answers ranged from no caregiver or self-care to various types of caregivers, depending on their relationship to each older adult, including both formal and informal caregivers. Then, those identified by older adults as the main caregiver were asked a set of questions about their tasks and the burden they face.
The sample in this study consisted of older adults aged 60 and above who reported having a caregiver. Initially, there were 4,092 older adults. Caregivers identified themselves as secondary caregivers (218 cases), and those who could not be approached (564 cases) were excluded. After removing cases with missing data for the main variables of interest, 3,190 older adults and their primary caregivers remained for analysis (see Fig. 1).
Fig. 1Flow chart of analytic sample selection
Caregiver burden, a multidimensional concept, refers to the emotional, social, financial, physical, and spiritual strains that caregivers experience. It was the first time a question asking the caregiver of an older person about caregiver burden had been added to the Thailand older person’s survey. The primary caregiver was asked in the “How much of a burden or difficulty do you feel caring for the older person places on you?” He/she rated his/her burden on a scale from 0 (= no burden) to 10 (= highest burden), with 1 representing the lowest. From the initial analysis, it was found that the median score for caregiver burden was 3 (with the 25th and 75th percentiles at 0 and 5, respectively), indicating a right-skewed distribution. To address this skewness, we divided the burden score into three no burden (score 0), low burden (scores 1–3), and high burden (score ≥ 4). This categorization using the median as a cut-point ensures statistical appropriateness and interpretability.
This study defined sensory impairments as visual and hearing impairments. These impairments in older adults can significantly hinder their daily activities and social participation [3], often leading to dependence on others. They were evaluated using two questions in the 2021 SOPT. The first question was “Can you see clearly?” with response (1) 1a) clearly without glasses or lenses, 1b) clearly with glasses or lenses, (2) not 2a) not clearly without glasses or lenses, 2b) not clearly even with glasses or lenses, and (3) unable to see at all. Response 1 was classified as “clear vision,” while the other responses were classified as “impaired vision or blindness.” The second question was “Can you hear clearly?” with response (1) 1a) clearly without a hearing aid, 1b) clearly with a hearing aid, (2) not 2a) not clearly without a hearing aid, 2b) not clearly even with a hearing aid, and (3) unable to hear at all. Response 1 was defined as “clear hearing,” and responses 2 & 3 were defined as “impaired hearing or deafness.” Then, these groupings were combined to form the main outcome variable, called “combined visual and hearing impairments,” with four (1) no impairments, (2) clear vision but impaired hearing or deafness, (3) impaired vision or blindness but clear hearing, and (4) dual sensory impairments. This categorization approach prioritizes functional ability in daily older adults who could see or hear clearly with assistive devices (glasses, lenses, or hearing aids) were classified as having clear vision or hearing, as these devices effectively address their sensory limitations and enable functional daily activities.
Covariates were selected based on the Pearlin stress process model [25] and the Yates stress appraisal process model [26] to consider the factors that can induce burden when caring for older adults. They are multi-factorial and interrelated, including the caregiver’s demographic and socioeconomic characteristics, caregiving hours, physical environment, household wealth, and the older adults’ health status.
Regarding caregiver factors, the age and gender of the primary caregiver at the time of the survey are important considerations. Caregiver age was categorized into two groups (< 60 years and ≥ 60 years). It’s well recognized that gender differences influence caregiving outcomes [27]. Education level refers to the highest level of education the primary caregiver has received. Caregivers with higher levels of education tend to have lower physical health burden but higher mental or emotional health burden than those with lower levels of education [28, 29].
A caregiver’s residence can reflect the relationship between the caregiver and the older person they care for. In Thailand, the primary caregiver living in the same household as the older person is often a relative. Such a bond may lead to role conflicts for the caregiver and increase their burden compared to that of a non-relative caregiver [16, 30]. Therefore, the caregiver’s residence served as a proxy for the relationship in this study.
Caregiver training programs enhance knowledge and skill proficiency, thereby strengthening caregivers’ confidence and competence in handling caregiving duties. As a result, it can help reduce common feelings of frustration, uncertainty, and burden associated with caregiving [31, 32]. This study utilized the question from the 2021 SOPT, “Have you (caregiver name) ever received formal training in caring for the older person?” as another variable.
The number of older adults being cared for is another factor that affects the burden on caregivers. Caring for more than one older person simultaneously indicates an increased burden [16]. The more care-receiving people in the household, the greater the chance that the caregiver will experience stress. This variable is derived from the total number of older people being cared for by the primary caregiver. In addition, the number of caregiving hours also affects the burden on caregivers [16]. This study measured the caregiving hours from a “In the past week, how many hours did you (caregiver name) spend on older person(s) being cared for?”
Quitting a job or taking time off work to provide care for an older person can have financial, emotional, and social burdens on the caregiver [33]. This study utilized two “Did you (caregiver name) quit a job to take care of the older person(s) or not?”, and “Currently, what is your (caregiver name) average monthly income?”. The 10,000 Baht threshold represents the approximate average monthly income for Thai older adults in 2021 (equivalent to approximately 300 USD/month; US$ at 2021 prices). This study categorized older adults’ income into three no income, 1–9,999 Baht/month (below average), and ≥ 10,000 Baht/month (average or above).
Besides caregiver factors, the older person’s characteristics, including age, gender, and dependency level, are crucial factors affecting the stress or tension of the caregiver [11, 34, 35]. This study used the Modified Barthel Index Scale, scoring ten activities of daily living, to measure the dependency level of older adults [36, 37]. Based on the total score, which is equal to 20, it was categorized into three scores 0–4 (dependence), scores 5–11 (partial dependence), and scores 12–20 (independence).
The home modifications (e.g., handrails) were considered indicators of the physical caregiving environment, and the household’s socioeconomic status was also considered as a covariate in this study. Recently, a study in Thailand confirmed that the presence of a handrail in households with older adults was associated with a reduced physical burden on caregivers [38]. Regarding socioeconomic status, this study employed the household wealth index as a proxy since it was a more accurate measure of the long-term wealth of households than household income or consumption [39, 40]. The index was calculated from the total sample of 83,880 households based on household characteristics and asset ownership, using principal component analysis (PCA) to convert these characteristics into scores [41]. Households were then classified into five quintiles based on their index scores, with quintile 1 representing the poorest households and quintile 5 representing the wealthiest households.
This study performed two steps of analysis. Initially, descriptive statistics, including counts and percentages, were used to disclose socio-demographic characteristics and factors of caregivers and older adults, stratified by sensory impairment status (no impairments, clear vision but impaired hearing or deafness, impaired vision or blindness but clear hearing, and dual sensory impairment) as shown in Table 1. In addition, bivariate analysis with a chi-square test was used to explore the differences between each independent variable and the level of caregiver burden. Then, the ordinal logistic analysis was performed to examine the association between the level of caregiver burden (no burden, low burden, higher burden) and visual and hearing impairments, accounting for the ordered nature of the burden outcome. The proportional odds assumption was tested using the Brant test. Since the assumption was violated (p < 0.05), generalized ordinal logistic regression with the autofit option was employed to relax the proportional odds assumption where necessary, allowing for different effects across outcome categories when appropriate. Five hierarchical logistic models were developed to assess this Model Unadjusted model (combined visual and hearing impairments only), Model Added care recipient characteristics (gender, age, dependency level), Model Added caregiver demographics and caregiving context (caregiver gender, age, education, number of older adults cared for, caregiving hours), Model Included caregiver socioeconomic factors (residence, elderly care training, employment changes, income), and Model Full model with added environmental and household factors (home modifications, household wealth index). Additionally, a sensitivity analysis treating caregiver burden as a continuous variable and performing linear regression on the full model was conducted to verify the findings from the ordinal logistic regression. All statistical analyses were carried out using STATA/SE 17.0 [42], where the value of alpha was set at 0.05 for statistical significance.
Among the total of 3,190 primary caregivers, 75.4% were female, 66.3% were under 60 years old, approximately half of them (55.4%) had education at the primary level or below, and 98.0% of them lived in the same household as the older person. Regarding economic status, 56.8% earned less than 10,000 Baht per month (see Table 1).
In terms of caregiving, 89.9% of primary caregivers had never received training in caring for older adults. Most of them (88.7%) cared for only one older person. Seven-point 1% had to quit their jobs to care for the older individual. Regarding the number of caregiving hours, 53.8% dedicated 1–8 h, while 20.7% provided more than 20 h in the past week (see Table 1).
A total of 35.1% and 30.6% of caregivers provided care to older adults aged 80–89 and 70–79 years, respectively. Meanwhile, 21.2% cared for individuals aged 60–69, and 13.1% provided care to those aged 90 years and above. The majority of caregivers (61.6%) provided care to female older adults. Regarding the functional status of care recipients, 67.8% of caregivers assisted socially bound older adults, while 18.0% and 14.2% cared for homebound and bedridden individuals, respectively. Additionally, nearly half of the caregivers (47.9%) provided care to older adults with clear vision and clear hearing, while 27.0% cared for older adults with both vision and hearing impairment (see Table 1).
Table 1Characteristics of the studied samples by sensory impairment status (N = 3,190)CharacteristicsTotalSensory impairment status (%)Number%No impairments(n = 1,527)Clear vision but impaired hearing or deafness (n = 378)Impaired vision orblindness but clear hearing (n = 425)Dual sensory impairments(n = 860)Characteristics of caregiverGender Male78424.626.924.122.821.6 Female2,40675.473.275.977.278.4Age < 60 years2,11666.363.763.866.472.2 ≥ 60 years1,07433.736.436.233.727.8Highest level of education Primary or lower1,76655.451.754.859.560.0 Secondary96630.330.733.328.529.1 Higher than secondary45814.417.611.912.010.9Residing with an older adult Yes3,12598.097.599.298.498.1 No652.02.60.81.71.9Have received training in caring for older adults (Trained caregiver) Never (No)2,86789.990.192.389.788.5 Ever (Yes)32310.19.97.710.411.5Number of older adults being cared for 12,82888.789.285.789.788.5 >136211.410.814.310.411.5Quit job for caregiving Not quit2,96392.993.592.691.192.9 Quit for less than 1 year260.81.20.50.50.5 Quit for more than 1 year2016.35.46.98.56.6Caregiving hours in past week < 12648.39.86.68.56.3 1–81,71553.854.459.054.849.8 9–2055217.315.713.818.121.3 > 2065920.720.120.618.622.7Income (Baht/month)^§^ No income38212.012.312.79.912.2 1–9,9991,81356.853.157.764.759.3 10,000 or more99531.234.729.625.428.5Characteristics of older adultAge (years) 60–6967521.233.613.816.24.8 70–7997730.634.725.938.421.6 80–891,12035.125.343.933.249.7 90+41813.16.416.412.224.0Gender Male1,22538.440.345.232.934.7 Female1,96561.659.754.867.165.4Dependency level Independence2,16267.875.170.964.055.4 Partial dependence57518.015.017.718.623.3 Dependence45314.210.011.417.421.4Number of handrails in the home No handrail1,11334.934.331.838.435.6 One1,39443.743.945.243.542.7 More than one68321.421.723.018.121.7Relative household wealth Quintile 1 (poorest)68621.518.522.226.424.1 Quintile 269021.621.522.022.121.5 Quintile 351216.114.517.715.118.6 Quintile 467621.223.319.620.018.8 Quintile 5 (wealthiest)62619.622.318.516.517.0^§^ Average exchange rate in 31.44 Baht = 1 US
It was found that the mean burden score of the primary caregivers for older adults was 3.22, with a standard deviation (SD) of 2.83. In addition, the median score was 3, with the 25th and 75th percentiles at 0 and 5, respectively, indicating a right-skewed distribution. After categorizing burden scores, 26.7% of primary caregivers reported no burden, 28.2% reported low burden, and nearly half (45.1%) experienced a higher burden level. Bivariate analyses revealed a significant association between caregiver burden and multiple caregiver characteristics (p < 0.001 for age, education, quitting job, caregiving hours, and income). Notably, caregivers aged 60 years and older, with a primary education or less, and who had quit their job for more than a year, experienced a higher burden than other groups. Those who provided care for less than one hour in the past week and those with an average monthly income of 10,000 Baht or more reported a lower burden than others (see Table 2).
Table 2Caregiver burden by caregivers’ characteristicsCaregivers’ characteristicsLevel of caregiver burden (%)No burden(score = 0)Low(score = 1–3)High(score > 4)TotalOverall26.728.245.1100.0Gender Male27.628.144.4100.0 Female26.528.245.3100.0Pearson chi^2^(2) = 0.3789 p = 0.827Age < 60 years28.029.442.6100.0 ≥ 60 years24.225.750.1100.0Pearson chi^2^(2) = 16.2518 p < 0.001Highest level of education Primary or lower25.725.948.4100.0 Secondary26.029.444.6100.0 Higher than secondary32.334.333.4100.0Pearson chi^2^(4) = 34.3551 p < 0.001Residing with the older adult Yes26.928.145.0100.0 No16.930.852.3100.0Pearson chi^2^(2) = 3.3150 p = 0.191Have received training in caring for older adults (Trained caregiver) Never (No)27.028.244.8100.0 Ever (Yes)24.227.948.0100.0Pearson chi^2^(2) = 1.5728 p = 0.455Number of older adults being cared for 126.628.045.4100.0 >127.929.043.1100.0Pearson chi^2^(2) = 0.6796 p = 0.712Quit job for caregiving Not quit27.528.843.7100.0 Quit for less than 1 year30.838.530.8100.0 Quit for more than 1 year14.916.968.2100.0Pearson chi^2^(4) = 47.9858 p < 0.001Caregiving hours in past week < 143.228.828.0100.0 1–824.829.246.0100.0 9–2023.727.548.7100.0 > 2027.625.646.7100.0Pearson chi^2^(6) = 52.6643 p < 0.001Income (Baht/month)^§^ No income25.428.346.3100.0 1–9,99926.325.847.9100.0 10,000 or more28.032.539.5100.0Pearson chi^2^(4) = 21.5474 p < 0.001**** Statistically significant at p-value < 0.01, ^§^ Average exchange rate in 31.44 Baht = 1 USD
The burden levels among caregivers also varied significantly with almost all characteristics of the older adults, including age, gender, visual and hearing impairments, level of dependency, and household wealth index (see Table 3).
Table 3Caregiver burden by older adults’ characteristicsOlder adults’ characteristicsLevel of caregiver burden (%)No burden(score = 0)Low(score = 1–3)High(score ≥ 4)TotalOverall26.728.245.1100.0Age (years) 60–6926.431.042.7100.0 70–7924.828.846.5100.0 80–8926.527.645.9100.0 90+32.523.743.8100.0Pearson chi^2^(6) = 13.4476 p = 0.036Gender Male23.827.448.7100.0 Female28.628.642.9100.0Pearson chi^2^(2) = 12.4291 p < 0.01*Combined visual and hearing impairments No impairments30.229.040.9100.0 Clear vision but impaired hearing or deafness20.628.850.5100.0 Impaired vision or blindness, but clear hearing25.425.449.2100.0 Dual sensory impairments24.027.848.3100.0Pearson chi^2^(6) = 28.4282 p < 0.001Dependency level Independence30.832.636.6100.0 Partial dependence17.720.262.1100.0 Dependence19.017.064.0100.0Pearson chi^2^(4) = 196.0258 p < 0.001Number of handrails in the home No handrail27.227.745.1100.0 One27.028.244.8100.0 More than one25.328.845.8100.0Pearson chi^2^(4) = 0.9765 p = 0.913Relative household wealth Quintile 1 (poorest)24.626.449.0100.0 Quintile 225.924.649.4100.0 Quintile 329.327.343.4100.0 Quintile 425.731.243.1100.0 Quintile 5 (wealthiest)28.931.339.8100.0Pearson chi^2^(8) = 22.3382 p < 0.01**, ** Statistically significant at p-value <.05, and <.01 respectively
Regarding the primary exposure of interest, caregiver burden varied significantly by older adults’ sensory impairment status (chi² = 28.43, p < 0.001). Caregivers of older adults with clear vision and hearing reported the lowest burden (40.9% higher burden), while those caring for individuals with hearing impairment alone reported the highest burden levels (50.5% higher burden). These bivariate associations informed the subsequent multivariable analyses examining the independent effects of sensory impairments on caregiver burden.
To examine whether and how the burden on caregivers, classified as no burden (score = 0), low burden (score = 1–3), and higher burden (score > 3)—differs when caring for older individuals with varying levels of visual and/or hearing impairments, ordinal logistic regression was initially attempted. However, the Brant test indicated a violation of the parallel lines assumption (p < 0.05), requiring the use of generalized ordinal logistic regression with the autofit option to accommodate non-proportional odds where necessary [43].
Across all models (Table 4), the pattern and magnitude of associations remained consistent, demonstrating the robustness of findings. In both the unadjusted model (Model 1) and the adjusted models (Models 2–5), caregivers of older adults with hearing impairment consistently showed the highest burden, followed by those caring for individuals with dual sensory impairments, then those with vision-only impairment, compared to the reference group (those with clear vision and hearing).
Table 4Unadjusted and adjusted odds ratios (OR, AOR) and 95% confidence interval (CI) from generalized ordinal logistic regressionCaregiver burdenModel 1(Unadjusted model)Model 2(Adjusted model)Model 3(Adjusted model)Model 4(Adjusted model)Model 5(Adjusted model)OR95% CIAOR95% CIAOR95% CIAOR95% CIAOR95% CICombined visual and hearing impairments of older adults No impairments (Reference) Clear vision but impaired hearing or deafness1.53***(1.24, 1.89)1.62***(1.20, 2.01)1.58***(1.27, 1.97)1.57***(1.26, 1.96)1.55***(1.24, 1.94) Impaired vision or blindness, but clear hearing1.36**(1.11, 1.66)1.31*(1.06, 1.61)1.29*(1.04, 1.59)1.27*(1.02, 1.57)1.26*(1.02, 1.56) Dual sensory impairments1.36***(1.16, 1.59)1.31**(1.10, 1.55)1.29**(1.08, 1.54)1.30**(1.09, 1.55)1.29**(1.08, 1.54)N3,1903,1903,1903,1903,190Log-likelihood-3395.6635-3281.8168-3252.0323-3234.8613-3229.3261LR chi^2^26.630254.330313.900348.240359.310Prob > chi^2^0.0000.0000.0000.0000.000Pseudo R^2^0.0040.0370.0460.0510.053OR Unadjusted odds ratio, *AOR *adjusted odds ratio, *, **, *** = statistically significant at p-value < 0.05, <0.01, and < 0.001 respectivelyModel 2 adjusted for gender, age, and dependency level of the older adultsModel 3 adjusted by adding gender, age, the highest level of education of the caregiver, the number of older adults being cared for, and the number of caregiving hours on top of Model 2Model 4 adjusted by adding the caregiver’s resident, received training in elderly care, quitting job, and caregiver income on top of Model 3Model 5 adjusted by adding the number of handrails in the older adult’s home and the household wealth index on top of Model 4
The fully adjusted model (Model 5) showed the best fit, based on the highest log-likelihood and likelihood ratio chi-square (LR chi2), and revealed a clear hierarchy of caregiver burden by impairment type. After controlling for caregiver demographics, caregiving context, socioeconomic factors, and environmental characteristics, caregivers who cared for older adults with clear vision but impaired hearing or deafness were 1.55 times more likely to experience burden than those caring for older adults with both clear vision and hearing (95% CI 1.24–1.94, p < 0.001). Meanwhile, caregivers caring for older adults with both impairments were 1.29 times more likely to experience burden than those caring for older adults with clear vision and hearing (95% CI 1.08–1.54, p < 0.01). Caregivers caring for older adults with impaired vision or blindness but with clear hearing were 1.26 times more likely to experience burden than those caring for older adults with both clear vision and hearing (95% CI 1.02–1.56, p < 0.05). A sensitivity analysis indicated that the magnitudes and directions of the regression coefficients for sensory impairment types remained consistent with the results from Model 5 of the ordinal logistic regression (see sensitivity analysis results in the Supplementary Material 1).
Notably, hearing impairment demonstrated the strongest independent association with caregiver burden, with effect sizes remaining statistically significant even after extensive adjustment for confounders. These findings suggest that sensory impairments significantly increase caregiver burden, with hearing impairment posing the greatest challenge, possibly due to communication barriers and associated behavioral changes in care recipients.
This study presents the first nationally representative comparison of caregiver burden by types of sensory impairment in Thailand and describes differences in caregiver burden across impairment groups.
The results revealed that, regarding caregiver burden, the average burden score for the caregivers in this study was only 3.22 out of 10, which is relatively low. This can be explained by the fact that caregiving is often framed as “Bun Khun” (Thai word), a sense of moral obligation and gratitude to one’s parents or older people. This relatively low average burden may partly reflect cultural norms in Thailand, where caregiving is often viewed as an expression of gratitude (“Bun Khun”) and moral responsibility toward older family members. Furthermore, in Thai society—a Buddhist society—where caregivers are typically relatives living in the same household, providing care for the older adults is often voluntary rather than mandatory. It also reflects gratitude towards benefactors and a way of accumulating spiritual merit [44, 45]. At the same time, it is also a social expectation, resulting in a low or even no perceived burden. There are studies to support that gratitude to parents acts as a buffer, mitigating the adverse effects of stress while amplifying the positive influence of evaluative factors on caregiver burden [46, 47]. The low prevalence of caregiver burden found in this study is consistent with [48–52] and contradictory to other research in Thailand and some Southeast Asian countries [11, 38, 53].
Among older adults receiving care, 27% had both visual and hearing impairments, followed by 13.3% with only visual impairment and 11.9% with only hearing impairment. The rest, about half, had no impairments at all. The proportion of older adults with these impairments is higher than in developed countries like the United States, but lower than in developing countries such as Mexico and India [54–56]. This intermediate level is consistent with Thailand’s status as a middle-income country in the midst of an epidemiological transition, the Age of Degenerative and Man-made Diseases.
Studies have shown that caring for older adults with multiple impairments can be more demanding for caregivers than caring for those with only one impairment [54, 57, 58]. However, the findings of this study differ. Using generalized logistic regression, the relationship between visual and hearing impairments among older adults and the level of caregiver burden (measured in three categories) was examined. Hearing impairment, but not visual impairment, was found to be most strongly associated with caregiver burden, followed by the combination of both impairments, while visual impairment alone was least associated. This result supports the idea that hearing impairment in older adults places a greater burden on caregivers than visual impairment, consistent with the study in disabled older people in Japan [59]. The likely reason is that caregivers require more skill or effort to care for older adults with hearing impairments than those with visual impairments, even when performing similar care activities, especially in communication [60].
Older adults with visual impairments may require assistance from caregivers to perform daily tasks or participate in special activities, such as offering support or guidance while walking or reminding them to exercise caution when handling potentially hazardous objects. This type of care may not demand much effort, as seniors can hear and understand what the caregiver says. This differs from communicating with an older person who has hearing impairments. Although that individual can see well, if communication is not face-to-face, they might not hear or may hear unclearly, which can lead to misunderstandings. This can cause frustration for both the caregiver and the older adult [61–63], especially if the caregiver has other responsibilities, such as household chores, in addition to caring for the senior. Additionally, hearing impairment in older adults can increase their risk of various dangers because they might be unable to hear sounds or warning signals. Therefore, caregivers must be careful not to let the older adults out of their sight, which adds another burden to their responsibilities.
Interestingly, having dual sensory impairments caused less caregiver burden than just hearing impairment alone, challenging assumptions that multiple impairments always increase caregiver stress. This surprising finding indicates that the link between impairment type and caregiver burden is more complicated than simple addition and needs further study. Several explanations may account for this counterintuitive finding. First, older adults with severe dual impairments may be more bedridden, potentially reducing the physical and emotional demands associated with supporting mobility, transportation, and social participation that caregivers of individuals with hearing loss alone must manage. Second, dual impairment may be more readily recognized as a significant disability by family members and healthcare providers, potentially leading to greater acceptance of dependency, more proactive seeking of external support services, and implementation of environmental adaptations. In contrast, hearing impairment alone may be underestimated or viewed as a normal part of aging, resulting in less formal support and greater reliance on family caregivers. Third, the communication barriers created by hearing impairment may be particularly frustrating and emotionally taxing for caregivers when the older adult’s other sensory and cognitive functions remain intact. Repeated communication breakdowns, social withdrawal, and the constant need to facilitate interactions may create chronic stress. When vision is also impaired, expectations for complex communication and social participation may be lower, potentially reducing this source of caregiver frustration. Finally, the presence of dual impairment may trigger different family coping mechanisms, support-seeking behaviors, or acceptance of professional care services compared to single sensory impairment.
This study aimed to explore the relationship between caregiver burden and visual and hearing impairments in older adults. Several limitations should be noted. First, a single-item burden measure, although practical for large surveys, may not effectively capture the multidimensional nature of caregiver burden, as validated tools such as the Zarit Burden Interview do [64]. Additionally, the survey did not capture information about the number of caregivers in the household or whether primary caregivers received assistance from other family members. The presence of caregiver networks and shared caregiving responsibilities may significantly influence burden levels.
Second, there is no standard method for categorizing caregiver burden into three groups. Third, the assessment of visual and hearing impairments depends on the older person’s perceptions rather than objective tests, such as visual acuity or audiometry. The perception of blurred vision or blindness tends to be more accurate than perceptions of muffled hearing or being unable to hear. Studies have shown that self-assessment of hearing impairment often underestimates the actual level rather than overestimates it. Moreover, our measure did not capture the severity or degree of sensory impairment. There is substantial heterogeneity within our ‘impaired’ categories, ranging from mild loss manageable with communication strategies or assistive devices to severe impairment that prevents most social interaction. The relationship between impairment severity and caregiver burden warrants further investigation.
Fourth, there was strong evidence supporting the association between cognitive impairment or dementia and sensory impairments [65–67]. However, the 2021 SOPT did not assess cognitive impairment or dementia in older adults. This prevented the analysis models from including it as another control variable. These limitations may influence the odds ratios in generalized ordinal logistic models. Future research should incorporate objective sensory testing with assessment of impairment, validated multidimensional burden measures, assessment of caregiver support networks, and longitudinal designs to examine causal relationships and trajectories of burden over time.
In conclusion, this study confirms that caregiver burden varies significantly by type of sensory impairment among older adults in Thailand. Caregivers of older adults with hearing impairment alone experience the highest burden, potentially due to communication barriers and associated psychosocial challenges that intensify caregiving demands. These findings have important implications for policy and practice on older persons in Thailand:
Strengthening the Implementation of Universal Health Coverage: Although the government provides free hearing aids and cataract surgery through universal health coverage, many older adults remain underserved due to a lack of awareness or insufficient caregiver support. This study highlights the need for community education programs that target both older adults and their caregivers, as well as hearing assistive device training for caregivers to enhance device use and effectiveness. It also stresses the importance of mobile screening initiatives to reach homebound older adults with sensory impairments.Including sensory disabilities in long-term care Current frameworks should clearly integrate sensory impairment assessments and specialized support services. This involves training healthcare workers and primary care managers to identify and handle communication issues, creating caregiver programs specifically for families managing hearing impairments, and establishing referral pathways between healthcare providers, caregivers, and community support services.Advocate for disability-inclusive aging policies that differentiate between various types of impairments, rather than treating older adults with disabilities as a uniform group. Develop targeted interventions based on impairment types and associated caregiver burden levels, or establish specialized caregiver training programs that emphasize communication strategies and the use of assistive technology.
Finally, research should develop culturally appropriate burden assessment tools for routine clinical and community use.
Supplementary Material 1.
Supplementary Material 2.