Introduction

The role of caregivers encompasses many domains, as caregivers provide physical, medical, financial and emotional support to patients entrusted to their care.1 However, this responsibility often comes at a cost, as caregivers may experience significant burden and adverse health outcomes.1 Caregiver burden is a term used to describe the physical, psychological, emotional, social and financial impact of caring for someone who is ill or has functional impairments.2 Although the definition could encompass all caregivers, it is often used to describe unpaid caregivers, who are usually family members of patients. These unpaid caregivers are often neglected or ignored and have been referred to as ‘the informal patient’.3 4

Chronic diseases, also referred to as ‘non-communicable diseases’ by the WHO, are defined by WHO as conditions ‘that are of long duration and generally slow progression, and are the result of a combination of genetic, physiological, environmental and behavioral factors’.5 The Centres for Disease Control and Prevention (CDC) defines them as conditions that last at least 1 year and require ongoing medical attention, limit activities of daily living or both.6 In the context of low–middle-income countries, a distinctive array of challenges, including gender norms, economic hardship, stigmatisation and non-inclusive governmental measures, exacerbates the caregiving experience.7 Addressing caregiver burden effectively requires informed strategies and policies grounded in empirical evidence to ensure that adequate support and services are provided.8

The prevalence of children with chronic medical conditions is high in Pakistan. Although prevalence for the most common individual disorders is available online, there is no grouped data on the prevalence of children with chronic medical conditions in the region. For instance, the incidence of childhood cancer in Pakistan is 100 per million, with approximately 8000 children diagnosed cancer every year.9 10 Similarly, the prevalence of paediatric epilepsy in Pakistan is 1% compared with the global prevalence of 0.6%.11 Furthermore, there are insufficient support resources tailored to the psychosocial needs of caregivers, which compound the challenges, potentially impacting the quality of care and overall family dynamics.12 13

Studies from other regions indicate varying levels of caregiver burden across different chronic conditions, emphasising the need for a comprehensive understanding of how specific illnesses influence caregiver experiences.14 15 There are a few studies reporting the caregiver burden in individual, specific chronic conditions, however there is little data related to pattern of chronic illness and its impact on the caregiver burden.16 17 This study aims to evaluate caregiver burden of children with chronic medical conditions in a tertiary care hospital, Karachi, Pakistan. It also aims to explore the various factors influencing caregiver burden and recommendations for potential interventions.

Methods

We conducted an explanatory sequential (Quantitative-Qualitative) mixed methods study over a duration of 6 months (November 2021 to June 2022) in the inpatient paediatric department of a tertiary care centre in Karachi, Pakistan.

Quantitative approach

The quantitative component adopted a cross-sectional design. We used non-purposive consecutive sampling, and all caregivers of patients aged 1–18 years with chronic medical illnesses, that fit the CDC definition of chronic diseases and consented for the study, were included.6 Written informed consent was obtained from the participants prior to the survey.

The survey included both caregiver and child variables. For caregivers, demographic details included the following categorical variables collected as ranges: age, average household income, number of family members and earning members. Discrete caregiver variables included gender, educational status, occupation, marital status and family type. Children demographic data included the following categorical variables collected as ranges: age, frequency of hospitalisations and monthly medical costs. Discrete variables included their gender and diagnosis. The burden score was evaluated using the Zarit Burden Interview (ZBI) scale.18 The scale has a validated Urdu version, which was modified for paediatric context. Permission to use the Urdu version and replace the phrase ‘your relative’ with ‘your child’ to make it relevant to the paediatric population was obtained through the official distribution portal.19 The ZBI consists of 22 items scored in 5-point Likert scale from 0 (never) to 4 (nearly always), except for the final item on global burden, rated from 0 (not at all) to 4 (extremely). The total score ranges from 0 to 88 with higher scores, indicating higher burden. Additionally, the burden scores can be categorised into severities based on the cut-offs suggested by the authors of the ZBI: no burden (0–20 points), mild burden (21–40 points), moderate burden (41–60 points) and severe burden (61–88 points).18 These cut-offs have been consistently used in the literature, including several validation studies.20–22 Sample size was estimated based on the result of a previous study in which the mean±SD of the ZBI caregiver burden was 39.04±15.14.23 Using the formula for estimating the mean of a quantitative variable, a margin of error of 5 points, 80% power, alpha of 0.05 and effect size (d) of 1.514 (0.1 of the given SD), a required sample size of 384 participants was calculated.

Qualitative approach

Semi-structured in-depth interviews (IDIs) were conducted to produce knowledge grounded in human experience, which might be overlooked in purely quantitative analyses. An interview guide was developed following a comprehensive review of literature pertaining to caregiver burden and chronic illnesses. Key topics included the overall caregiving experience, challenges, changes in routines and relationships as well as the impact on physical, emotional and mental well-being since the child’s diagnosis. Additionally, participants were asked about their perceived sources of support, coping mechanisms and unmet needs related to their caregiving responsibilities (the guide can be found as online supplemental file 1). The guide was translated to Urdu and used by the interviewers during all IDIs to ensure consistency across the interviews.

Participants for the IDIs were selected through purposive sampling from those who had completed the quantitative survey and exhibited a willingness to share detailed experiences beyond the scope of the survey. Invitations to participate in the IDIs were extended to caregivers who demonstrated a propensity for providing rich qualitative data during the survey process. Participants were explained about the procedure and the audio-recording of interviews, and written informed consent was obtained.

To ensure privacy and confidentiality, IDIs were conducted in a designated patient discussion room within the paediatric ward. For the duration of the interview, it was ensured that the child was accompanied by either a second caregiver if available (such as the participant’s spouse) or a nurse present constantly at the patient’s bedside. The language of the interviews with the participants was a combination of Urdu and English, as per the participants’ preferences. These audio-recorded interviews were transcribed verbatim by a trained member of the research team. Transcripts were translated to English by a trained member of the research team and then back-translated to Urdu by a second trained member to ensure linguistic accuracy and preserve the integrity of participants’ narratives. The approved English transcripts then underwent thematic analysis.

Initially, we aimed to interview 10% (n=40) of the participants from the quantitative survey. However, after conducting a preliminary analysis of the first 10 interviews, it became evident that theoretical saturation had been reached as subsequent interviews yielded no new themes beyond those already identified. Consequently, the decision was made to conclude the interview process after sessions with 19 participants.

Data analysis

Quantitative data were analysed using STATA V.15. Categorical variables are presented as frequencies and percentages. The numerical burden scores from the ZBI scale are presented as mean±SD and median (IQR). Univariate analyses were conducted using the Mann-Whitney U and Kruskal-Wallis tests to explore associations between caregiver and child variables with burden scores. Subsequently, a multivariable regression analysis was performed, including variables with a univariate p value less than 0.25. P<0.05 was considered statistically significant.

Qualitative data were analysed to the point of theoretical saturation, the point at which no new concepts are generated from the data.24 Thematic analysis was conducted to identify patterns, working systematically through the entire data set, giving equal attention to each item and exploring note-worthy aspects that may form the basis of themes across the data set. The themes were assigned unique codes and relevant information from each interview was added to the specific code.25 The analysis was conducted independently by two researchers. NVivo software was used to facilitate the organisation and management of the qualitative data.26

Patient and public involvement

Patients or the public were not involved in the design, or conduct, or reporting or dissemination plans of our research.

ResultsQuantitative results

A total of 383 caregivers completed the survey. Nearly half of the respondents were aged between 26 and 35 years old (48%, n=186), with the majority being the admitted child’s parent (93%, n=357), female (69%, n=263) and married (98%, n=374). Sixty-four per cent (n=247) of caregivers were unemployed, while 50% (n=192) had an average household income between 10 000 and 50 000 PKR. Family structures were predominantly non-nuclear or extended (52%, n=198), with at least two earning members in 52% of families (n=201).

The ages of the admitted children ranged from 12 months to 17 years old. Among these, 58% (n=224) were men, and 36% (n=136) were aged between 6 and 13 years old, with the majority (31%, n=118) being diagnosed with the chronic condition during infancy (aged less than 12 months), and around half (51%, n=197) of them having had the diagnosis for the past 1–3 years. The most common diagnosis among children in our study was cancer (24%, n=92), followed by seizure disorders (14%, n=55) and endocrine disorders (12%, n=45).

The Urdu-translated version of the ZBI employed in our study demonstrated good internal consistency, with a Cronbach’s alpha of 0.865 (n=383). The responses to the questionnaire yielded an overall mean burden score of 35.35±15.14, with 17% (n=66) of respondents experiencing no burden, 46% (n=177) experiencing mild burden, 31% (n=118) experiencing moderate burden and 6% (n=22) experiencing severe burden.

Considering specific disease types, the highest mean burden score was observed for children diagnosed with congenital cardiac disease (42.97±15.47), followed by Cerebral Palsy (CP) (40.43±17.68) (table 1).

Table 1

Mean caregiver burden scores across disease types (n=383)

Results of non-parametric testing revealed significant associations between caregiver burden and several demographic factors (table 2). Caregiver age, educational status and marital status were significantly associated with burden (p<0.05). Furthermore, parents reported significantly higher burden scores than non-parent caregivers (p<0.001). Burden scores were also significantly different based on average household incomes (p<0.001) and number of earning family members (p=0.044).

Non-parametric testing of children demographics revealed that burden scores were significantly different based on the child’s current age (p<0.001), age at diagnosis (p=0.014), duration of illness (p=0.028) and number of hospital visits since diagnosis or onset of symptoms (p<0.001) (table 3).

Table 2

Caregiver demographics and their association with caregiver burden (n=383).

Table 3

Children demographics and their association with caregiver burden (n=383)

After performing multivariable analysis, the caregiver variables of ‘relationship with patient’ and ‘educational status’, along with the child variables ‘age at diagnosis’ and ‘number of hospital visits’ remained significant predictors of caregiver burden (p<0.05) (table 4).

Table 4

Multivariable regression analysis of caregiver burden predictors

Qualitative results

The IDIs with 19 caregivers explored their overall experiences, factors influencing caregiver burden and recommendations for potential support and improvement in care (figure 1). A detailed breakdown of the various themes and subthemes as well as the verbatim quotes from the participants can be found in the online supplemental materials of the manuscript (online supplemental file 2).

Figure 1

Themes and subthemes emerging from the IDIs. IDIs, in-depth interviews.

Challenges

All caregivers, regardless of their socioeconomic status, identified financial challenges as their primary concern, impacting various aspects of their lives, including treatment access, lifestyle adjustments, mental well-being and social interactions. Majority of the caregivers also reported significant psychosocial challenges, including the trauma of accepting the reality of their child’s condition and a lack of emotional support from the community. Adjusting to the child’s increasing dependency also posed challenges. This ranged from complete physical support to heightened care due to special dietary needs, treatment and symptom monitoring. Caregivers also identified distance as a challenge to accessing care. While some faced a problem commuting within the city due to logistical issues, others travelled across the country to access care.

Impact on lifestyle

Most caregivers expressed mental distress and noted a lack of time for self-care, attributing these challenges to the overwhelming responsibilities associated with their child’s care. Others attributed this lack of self-care to financial pressures. Despite these challenges, many caregivers maintained hope and a positive outlook, making special efforts to ensure their child’s comfort and normalcy for the family. However, caregivers of children with more severe and chronic conditions expressed distress and hopelessness. Their methods of coping varied, depending on duration, severity and chronicity of the child’s condition, frequency of hospital visits and financial stressors.

Impact on relationships

Relationship dynamics were notably affected, with caregivers reporting disruptions in their relationships with other children, often due to reduced attention and resources. Many noted a change in their emotional disposition, affecting their children’s behaviour in response to decreased parental attention. Financial challenges also led to educational setbacks for other children.

Majority of the caregivers spoke positively about their relationship with their spouse, with most acknowledging their partners as significant sources of support. However, some caregivers experienced strain in their relationships, particularly when partners failed to understand the caregiver’s changed behaviour due to stress. Responses about relationships with friends and family varied. While some received support from relatives in the care of their children, others related an almost complete absence of relatives in their life following their child’s diagnosis. A recurrent reason was their inability to be present at social events due to increased caretaking demands. Financial difficulties sometimes strained relationships, particularly when borrowing money was involved.

Caregiver health

Following their child’s diagnosis, caregivers reported a significant impact on their health, experiencing worsening of pre-existing medical conditions due to stress. Mood changes, including anxiety, depression and anger, were commonly mentioned.

Several caregivers described various coping strategies for mental stress, ranging from prayer to family outings. However, many expressed a complete absence of time for personal mental and emotional self-care.

Most caregivers were reluctant to seek professional psychiatric help. Many reasoned that the financial burden following their child’s diagnosis left little room for any extra expense, while others mentioned a lack of time. They denied feeling any hesitation in accessing psychiatric help due to sociocultural stigmas. Prioritising their child’s needs, most caregivers overlooked the impacts on their own physical and mental health.

Attitude towards child

An overwhelming majority of the responses discussed positive changes in their family and social circle’s attitudes towards their child postdiagnosis, noting an increase in love and care. However, in a few families, a child’s need for special attention and support was also accompanied by what caretakers felt was an increase in judgement and change in perception of their child among their distant relatives and family members. Despite many positive experiences, a few caregivers faced distressing negative reactions within their social circles.

Support and resources

Overall, the greatest support in every domain identified by caregivers was their spouse and partners. Additionally, many found solace in their religious faith, which provided emotional strength during challenging times. Many caregivers living in joint families discussed this as a major source of support. Hospitals were also mentioned as a major source of support for some families, offering financial assistance for treatment and long-term management, either through partial discounts or fully covering costs. A unique finding was one caregiver mentioning social media as the biggest support in their journey in the form of treatment funding. However, not all caregivers experienced such support. Some faced the challenge alone, without any external assistance, struggling to manage hospital visits and afford medications.

Hospital services

Most caregivers praised the medical team and care management during hospital visits, highlighting the emotional availability and support provided by doctors and their teams. However, there were suggestions for improving inter-team coordination and increasing access to screening tests, such as antenatal testing for Down syndrome as well as the need for health education and awareness among patients and their families. Waiting times were another concern, with calls for more efficient and streamlined procedures, especially for families who frequently visit the hospital.

Discussion

The intricate dynamics of caregiving, especially for children with chronic medical conditions, present a multifaceted challenge that extends beyond the immediate medical needs of the child. Our study offers comprehensive insights into the caregiver burden of children with chronic illnesses, using both quantitative and qualitative approaches. We found that the role of the caregiver was majorly taken up by women, particularly young and middle-aged mothers, most of whom were unemployed homemakers, a demographic in line with prior literature.14 27 Our study yielded a mean burden score of 35.35±15.14, with nearly half (46%) of the caregivers considering the burden to be mild, while 37% noting experiencing moderate-to-severe burden. A study conducted in Iran also reported a similar overall mean ZBI score of 39.04±15.14, while results from the USA found moderate caregiver burden (mean: 52.55±16.8).23 28 This difference between white and non-white caregivers has been previously attributed to different degrees of knowledge of medical illnesses, varying sociocultural expectations, family-centred values and healthcare system dynamics.29 30

The common diagnoses in our study, such as cancer, seizure disorders and endocrine disorders, align with global trends.31 We found the highest burden among caregivers of children with congenital cardiac diseases, followed by CP and musculoskeletal disorders. Piran et al reported the highest caregiver burden in CP, attributing it to the wide array of challenges posed by the condition, encompassing motor, mental, emotional and communication impairments, which not only intensify caregiving demands but also strain finances and disrupt family routines.14 Congenital cardiac diseases, often identified early life, are associated with a looming uncertainty of complications such as heart failure, pneumonia and brain abscesses, which can amplify anxiety and stress for caregivers, fearing potential life-threatening episodes.32 Similar to CP, children with congenital heart conditions often face dietary and activity constraints, demanding unwavering attention from caregivers who may find reduced opportunities for leisure and social interactions, along with constant preoccupation with the child’s health and a relentless cycle of physical and emotional strain.

While our study did not explicitly examine the correlation between burden scores and the intensity of care required for different illnesses, it is a crucial aspect in understanding caregiver burden comprehensively. The intensity of care, healthcare access and insurance coverage can significantly influence the caregiving experience and caregiver burden. In Pakistan, over 58% of healthcare costs are out-of-pocket, as medical insurance coverage is in the single digits, which has been attributed to low insurance literacy among the population and to a lack of empaneled private providers.33 34 The high rate of out-of-pocket expenditure stems from a low level of government spending on health, as successive governments in Pakistan have allocated much lower amounts of their budgets to healthcare, even when compared with other developing countries.35 Factors such as the availability of healthcare resources, including specialised medical care, therapies and support services, can impact the caregiver’s ability to manage the child’s condition effectively.

Several associations with caregiver burden emerged in our study. Considering caregiver factors, we found that older participants had higher levels of burden. A study conducted on caregivers of adults with intellectual disabilities reported higher levels of worry and lower quality of life among older caregivers and suggested that older caregivers tend to lose their social support with time.36 Additionally, older caregivers might face their own age-related health challenges, thereby compounding the stress and demands of caregiving. Our study found no difference in burden among employed and unemployed caregivers, a result also reported by Piran et al.14 A potential explanation for this could be that the majority of our respondents were women, primarily homemakers, who might not be the primary earners in their households and thus, their employment status might not have a direct bearing on the household’s overall socioeconomic standing. This further aligns with our finding that caregivers from households with higher monthly incomes reported significantly lower burdens, an association consistently observed in the literature.14 23 37

Interestingly, we found no significant difference in burden based on caregiver gender. Prior literature is also conflicted in this regard, with several studies reporting significantly higher burden among women and mothers while others have not discerned any gender-specific associations.14 23 38 39 However, our qualitative insights suggest that irrespective of gender, caregivers grapple with similar emotional challenges, shaped by societal expectations and perceptions, acceptance of the child’s condition and the inherent demands of caregiving.

Our study found that caregivers of children diagnosed with chronic illnesses at younger ages tended to have significantly higher burden. This may be attributed to the increased care demands of younger children, who naturally require more attention and supervision. Additionally, an early diagnosis implies a prolonged journey with the illness, leading to extended caregiving durations. This notion is further corroborated by our finding that burden among caregivers increased with longer durations of the illness and with increased visits to the hospital. A study by Boztepe et al also found that younger patients diagnosed with leukaemia were associated with higher burden among caregivers.40 Another study reported contrasting findings, reporting no significant associations between perceived burden and children’s age or duration of disease.23 The authors suggested that caregivers may learn to better cope with the associated burdens. However, they also recognised that elevated burden levels could persist among these individuals, driven by concerns about their child’s disease course, prognosis, quality of life and uncertainties surrounding their future. It is also likely that a prolonged duration of illness may lead to an increase in the overall financial demands of care, thereby contributing to the overall caregiver burden.

The qualitative findings of our study serve to further illuminate the lived experience of caregivers. The recurring theme of financial strain appeared to transcend socioeconomic boundaries, being evident across various backgrounds. Interviews with the caregivers revealed that financial strain is complex, stemming not only from the direct costs of medical treatment but also indirect costs that often go unnoticed, such as transportation to and from medical facilities and potential loss of income due to missed workdays. Our study resonates with global trends and prior literature underscoring the financial challenges faced in the care of children with chronic conditions and emphasises the need for systemic interventions through financial aid, subsidised medical care or community support initiatives.41 42 It is reasonable to infer that our study’s sample is a representative sample of the country, as majority of our patients were in the PKR 10 000 to PKR 50 000 income bracket, in line with the national average household income of PKR 41 545 documented by the Pakistan Bureau of Statistics.43

Our qualitative results also provide valuable insights into potential avenues for intervention. Caregivers emphasised the need for improved coordination among medical teams and provision of more comprehensive information to caregivers about their child’s medical condition. Indeed, the role of health education in improving patient and caregiver outcomes is well established in the literature and employing future care strategies by clinicians regarding education of caregivers about disease information, prognoses and avenues for support can aid to reduce caregiver burden.44 45 Accessibility to essential medications closer to home was another concern, addressing the added strain of travelling long distances for treatments.

Prior research supports the efficacy of non-pharmacological interventions in reducing caregiver burden. For example, it was found that diaphragmatic breathing resulted in lower levels of anxiety, stress, depression, burnout and decreased ZBI scores among caregivers of paediatric palliative patients.46 Similarly, cognitive behavioural therapy resulted in lower anxiety levels in mothers of children diagnosed with cancer.47 Multicomponent interventions, including psychoeducation and support group, offer a comprehensive approach to addressing caregiver burden.48 Physicians treating patients requiring caregivers should be trained to recognise and address the emotional needs of caregivers and advocate for caregivers keeping in mind the important deleterious effects of caregiver burden. Training programmes should emphasise the effectiveness of non-pharmacological interventions to benefit both patients and caregivers.

Strengths and limitations

The major strength of our study lies in its mixed-methods design, which provides a comprehensive understanding of caregiver burden amidst chronic childhood illness within the context of a lower middle-income country. The use of a validated assessment tool ensures the robustness of burden measures in our study, while the inclusion of diverse paediatric illnesses addresses a critical gap, presenting comprehensive data on burden across various chronic conditions.

However, several limitations warrant acknowledgement. First, the cross-sectional design of the survey may not fully capture the dynamic and evolving nature of the caregiving experience. Specifically, we were unable to assess how changes in the child’s health status due to ongoing treatment impact caregiver stress and burden, nor could we track how accumulating medical expenses over time might exacerbate burden levels. To address these limitations, future research could adopt a longitudinal approach, allowing for a deeper exploration of these complex relationships. Next, our single-centre setting may limit generalisability to other regions or healthcare settings in Pakistan. Additionally, we have used the Centers of Disease Control and Prevention’s definition of chronic diseases. We recognise that this broad definition includes diseases with a wide range in disease course, severity of clinical outcomes and includes both diseases that are associated with acute-on-chronic exacerbations and diseases with a more chronic, indolent course. This wide range can cause fluctuations and differences in the required level of caregiver dependence.

While the use of a validated tool in the ZBI improves the reliability of our findings, a patient survey-based questionnaire and interviews may be subjected to recall or social desirability biases. Additionally, approaching caregivers during inpatient hospital stays, when stress levels may already be elevated, could potentially influence their responses to the ZBI questionnaire. To mitigate these potential biases, future studies should consider conducting surveys in the home environment or during routine outpatient follow-up visits, which may offer a more accurate depiction of caregiver burden.

While our survey inquired about the presence of multiple earning members within families and the total number of household members, aiming to capture potential sources of support that might alleviate caregiver burden, we acknowledge the importance of also assessing the availability of medical, paramedical or non-medical personnel at home to aid in the child’s care. Recognising the potential impact of such assistance on caregiver burden, future studies could benefit from incorporating these additional factors into their assessments. Finally, we recommend that future surveys include inquiries about the schooling status of children, as caregivers of school-going children may have opportunities for respite while their child is in school. This inclusion could provide valuable insights into potential strategies for alleviating caregiver burden associated with continuous caregiving responsibilities.

Conclusion

The findings of our study hold profound implications for the healthcare administrations, policymakers and clinicians in the realm of paediatric chronic illnesses and warrant the need for a holistic approach to care especially in the context of low and middle-income countries. The reported caregiver demographics emphasise the socioeconomic vulnerabilities that can exacerbate the caregiving burden, while qualitative insights into caregivers’ experiences present actionable areas for intervention. Alongside the resonance with global literature suggesting universal themes regarding caregiving and burden, our study necessitates the consideration of regional nuances, particularly in developing countries like Pakistan, for tailored interventions. The integration of these insights into policy and practice is essential to provide caregivers with the support and resources they need to navigate the demanding journey of caring for such children.