Chronic illnesses requiring home-based palliative care and their impact on families in an urban resettlement colony of Delhi, India

Abstract

Background:

Non-communicable diseases are a growing public health concern in India. However, limited knowledge of community-based need for palliative care has contributed to its poor access.

Objective:

To assess the community-based palliative care needs, social security access, and the economic burden on families requiring home-based palliative care.

Design:

A community-based cross-sectional study.

Methods:

The entire population of an urban resettlement colony was surveyed by trained field research workers to identify people requiring home-based palliative care, whose needs were confirmed by a physician trained in palliative care needs identification. Data were collected on sociodemographics, health status, disease details, access to social security schemes, and economic impact. People in need of home-based palliative care were referred for home-based care and social security guidance. Data were analyzed using R and geographically mapped with ArcGIS and Google My Maps.

Results:

Out of 43,267 individuals, 0.21% (2 per 1000) required home-based palliative care. The majority were elderly males (60%), with 51.11% illiterate and 55.56% previously unemployed. Neurological disorders, primarily stroke (67.8%), were the leading cause of disability. The average Barthel Index score was 33, indicating severe dependence in nearly 49% of participants. 62.22% of families reported a negative quality of life, and 34.44% incurred debt due to illness. 73.33% had ration cards, 50% received pensions, and only 15.56% had public health insurance. The mean out-of-pocket healthcare expenditure was 58.56% of their per capita income and 11.11% of their total family income.

Conclusion:

The study highlights the significant need for home-based palliative care in urban areas and the financial hardship families face. There is a need for community-based package development for palliative care service delivery followed by an evaluation of its effectiveness.

Introduction

Non-communicable diseases (NCDs) have surged to the forefront of public health concerns in India. India had an estimated 100.4 incident cancer cases per 100,000 people in 2022, which is expected to rise in the coming years. ¹ The incidence of other NCDs like chronic cardiovascular and respiratory diseases has also grown over the last few decades owing to an epidemiologic transition, adding to the burden of palliative care in the country. ² NCDs account for about 16,939 DALYs per 100,000 in India. ³ In 2017, it was estimated that approximately 4.7 million deaths had occurred in India due to NCDs, which comprised 49% of all-cause mortality. ³

Due to various sociodemographic and health system-related factors, a majority of patients with NCDs are diagnosed in the advanced stage of the disease.^4,5 While screening and early diagnosis would help in the early initiation of treatment, relieving suffering by providing palliative care to people with advanced stages of chronic diseases is vital.

Palliative care is a branch of medicine that aims to prevent and relieve physical, social, and spiritual suffering of patients suffering from chronic life-limiting illnesses and their caregivers. ⁶ Realizing the impact of palliative care on the quality of life of patients and caregivers, in 2014, the World Health Organization passed a resolution urging member states to provide palliative care services as a part of comprehensive care. ⁵

Palliative care was first introduced in India in the mid-1980s. ⁵ The National Program for Palliative Care (NPPC) was launched in 2012 to promote palliative care provision at the community and all health system levels. The National Program for Non-Communicable Diseases also includes palliative provision from district hospitals and medical colleges. However, despite having these programs in place, it is estimated that <4% of India’s population has access to palliative care. ⁷ With the rising incidence of chronic diseases, the need for palliative care is also expected to rise in India to address serious health-related suffering. ⁸

The existing literature on community-level palliative care needs in India ranges widely from 1.5/1000 population ⁹ to 43.1/1000 population. ¹⁰ Out of the existing eight studies, only three have focused on the needs of the urban population. ¹¹ Within the urban population, there is a paucity of information on the needs of the urban poor people. As India aims to achieve Universal Health Coverage by 2030, estimating the need for palliative care in diverse sociodemographic populations is crucial to ensure equity in care provision.

In this study, we assessed the community-based need for palliative care. Our secondary objective was to assess the social security of families with people requiring HBPC and assess the economic burden of the disease on the family.

Material and methods

Study design and setting

A community-based cross-sectional study was conducted in the field practice area of the Urban Health Center (UHC), Department of Community Medicine, Maulana Azad Medical College, Delhi, India, located at Gangavihar in the northeast district of Delhi. The study site was chosen as, besides being the field practice area of the medical college, it is also an updated demographic, developmental, and environmental surveillance site. ¹² The total study area consisted of Gangavihar, Gokalpuri, and Gokalpuri village, consisting of 9597 households with a total population of 54,614 people.

Data collection

A door-to-door survey was conducted on the entire population in the field practice area from 25 September 2024 to 10 January 2025. Informed consent was obtained from all the study participants. Households were screened consecutively by the 10 field research workers (FRWs), and only residents of the area were included in the study. The FRWs were given basic training by a trained physician on the assessment of palliative care needs. During the household screening, information on the name of the head of the family, their contact details, number of family members, number of people above 60 years of age in the family, and the presence of people affected by chronic disease in the family was collected. The screening for the need for palliative care was done by FRWs using three questions—(1) ‘Is there anyone in the house who is bedridden?’ or (2) ‘Is there any person in the house who needs help in activities of daily living?’ or, (3) ‘Is there anyone in the family who is not able to go for or quit work/education due to any physical chronic illness?’ A line listing of all households was prepared in which at least one member answered ‘yes’ to any of the three screening questions. We used this technique for multiple reasons. First, we wanted to generate evidence comparable to existing literature, which has commonly used the same three-question technique.^9,13,14 Second, as we aimed to cover a large population with the help of field workers, the three-question technique was easier compared to other tools to screen a large population using a door-to-door survey. The three questions took less time and required minimal training. Lastly, our primary objective was to identify the need for only HBPC. The three-question technique was therefore used as it helps in identifying people who are home or bed-bound. Using other tools would have given us a huge number of people in need of both outpatient and HBPC. The households found locked on two separate visits were excluded from the study.

The households in need of palliative care identified by FRWs were visited by a community physician trained in primary palliative care to confirm the need for palliative care. An individual requiring assistance with daily activities was evaluated for significant dependency using the Barthel Index, with scores ranging from 0 to 100. ¹⁵ The criteria for including people for HBPC services have been mentioned below in the section titled ‘Operational definitions’.

Social and economic vulnerability assessment of the patients was done using a pre-tested questionnaire (Supplemental File). The financial burden of the illness on the family was also assessed in terms of direct medical and non-medical expenditure on seeking healthcare in the past 6 months and financial adaptations in the family following the development of the disease and dependence of the person. Other details collected included sociodemographic details, disease and treatment details of the patient, the impact of the disease on the family, direct medical and non-medical out-of-pocket expenditure in the past 6 months, access to social security schemes, and characteristics of the primary caregiver of the patient. The patient and caregiver were also questioned about their awareness of palliative care by asking, ‘Have you heard of terms such as palliative care, end-of-life care, home care for bedridden individuals, or community-based care?’ Those who were unaware were provided with appropriate information, and their queries were addressed by the community physician trained in primary palliative care (P.S. and A.K.J.). The entire population was surveyed to avoid selection bias, and multiple visits were made to check empty houses to avoid non-response bias. Since the study site was a demographic site in which multiple surveys had been conducted previously by the same research team, a good rapport had been built with the community, reducing the chances of information bias.

Data were collected in Epicollect5. ¹⁶ The geolocation of the patients was also collected using the Epicollect5 application.

Patients identified with palliative care needs were given medical treatment at home and enrolled in the HBPC program initiated by the investigating institution. Through a public-private partnership model, in collaboration with a not-for-profit organization, patients were provided with their necessary medications free of cost. They were also made aware of the social security schemes they were eligible for and connected with the government-affiliated community health worker for support to enroll in the relevant schemes. When needed, patients were referred to a higher center for evaluation and management.

Outcome measures

• a. The proportion of participants needing community-based palliative care per Barthel Index.

• b. The proportion of households availing of any social security measures like pension for disability, old age, widowhood, or public health insurance or ration cards.

• c. The out-of-pocket healthcare expenditure on drugs, non-emergency medical care, investigations, and the cost of travel to the treatment facilities incurred by the families.

Operational definitions

1. Need for home-based care: The Barthel Index was used for the functional assessment of the patients and ranges from ‘0’ to ‘100’. Based on the total score, the participant’s functional status is categorized as follows: a score of 0–20 indicates total dependence, a score of 21–60 indicates severe dependence, a score of 61–90 indicates moderate dependence, a score of 91–99 indicates slight dependence, and a score of 100 indicates Independence. In this study, a participant scoring 60 or lower on the Barthel Index was considered to have marked dependence, indicating the need for community-based palliative care. ¹⁵ People who had a Barthel Index of more than 60 but were socially dependent and homebound, that is, needed another person’s presence to leave the house, due to psychological disabilities, were also considered to be eligible for home-based care.

2. Socioeconomic status: The socioeconomic status of the participants was assessed using the BG Prasad Scale and divided into five classes, from class I to class V. The classification is based on their per capita income, with households belonging to class I having a higher socioeconomic status. ¹⁷

3. Trained and untrained caregiver: Caregivers who were from a medical caregiving background (e.g., doctors or nurses) or had received training for caregiving were considered to be trained caregivers. All others were considered to be untrained caregivers.

4. Dependency ratio: The average number of economically dependent populations per 100 economically productive populations for the population belonging to families with at least one member in need of HBPC. ¹⁸

$$\begin{array}{l}(\mathrm{Number}\mathrm{of}people\mathrm{aged}0\mathrm{to}14+\mathrm{Number}\mathrm{of}\mathrm{people}\\ \mathrm{aged}\ge 65+\mathrm{People}\mathrm{in}\mathrm{need}\mathrm{of}\mathrm{HBPC}\mathrm{in}\mathrm{the}\mathrm{age}\\ \mathrm{group}\mathrm{of}15\mathrm{to}64)*100/\\ (\mathrm{Number}\mathrm{of}\mathrm{people}\mathrm{aged}15\mathrm{to}64-\mathrm{Study}\mathrm{participants}\\ \mathrm{in}\mathrm{the}\mathrm{age}\mathrm{group}\mathrm{of}15\mathrm{to}64)\end{array}$$

• 5. Total healthcare costs incurred per month: It is the sum of cost spent on drugs per month, the cost spent on non-emergency medical care over the past 6 months divided by 6, the cost spent on investigations over the past 6 months divided by 6, and cost of travel to and from the healthcare facility per month.

• 6. Distance from healthcare facility: The shortest route between the location of individual participants and the UHC using a road.

Statistical analysis

Statistical analysis was performed using R version 4.4.2. ¹⁹ A descriptive data analysis was done for the variables. Quantitative data were expressed using mean and standard deviation, median, interquartile range, and range, while qualitative data were summarized as frequencies and proportions with 95% confidence intervals. For geographical analysis, participant locations were geocoded and mapped using ArcGIS ArcMap software version 10.8.1 (Environmental Systems Research Institute) overlaid on the study area base map. ¹² The distance from the healthcare facility was calculated using Google Maps software based on the shortest road network between the two locations. ²⁰ The findings were reported using the STROBE checklist (Supplemental File). ²¹

Results

The FRWs screened 8956 (93.32%) out of 9597 households after excluding the locked houses. The included households constituted a total population of 43,267. In the initial survey by the FRWs, 0.34% of individuals were screened positive for the need for home-based care. After the second round of screening by trained physicians, 35% of the screened positive individuals were not found to be requiring home-based care as their Barthel Index was more than 60 and they were not socially dependent, 3.4% of the houses were locked even after two visits, and 2.72% of the participants denied consent. The final proportion of participants requiring home-based care was 0.21%, that is, 2 in every 1000 people, with a confidence interval of 0.17–0.25 for the population (Figure 1).

Figure 1.

Flowchart of participant selection.

Profile of patients and their caregivers

Profile of patients

The majority of the people requiring HBPC were males (54.44%) and aged 60 years or above (60%; Table 1). The majority (51.11%) of the participants were illiterate, with 55.56% being previously unemployed. Among the study participants, 50% were married, 25.56% were widowed, divorced, or separated, while the rest were unmarried. The majority (54.44%) of the participants lived in a joint family, and most (78.89%) of the families had more than or equal to five members. In most (76.67%) families, two or fewer members were earning. For families with people requiring HBPC, the dependency ratio was 70.82%. As per the BG Prasad Scale, the majority (38.10%) of the participants belonged to class III, followed by class II (22.62%) and class IV (20.24%; Table 1).

Table 1.

Distribution of demographic and socioeconomic characteristics of study participants in need of home-based palliative care.

Variable	Frequency (%)
Gender (N = 90)
Male	49 (54.44)
Female	41 (45.56)
Age (years; N = 90)
<15	7 (7.78)
15–59	29 (32.22)
⩾60	54 (60.00)
Mean ± SD	55.39 ± 22.76
Religion (N = 90)
Hindu	84 (93.33)
Muslim	4 (4.44)
Others	2 (2.23)
Education of participants (N = 90)
Illiterate	46 (51.11)
Primary school	18 (20.00)
Middle school	10 (11.11)
High school	11 (12.22)
Post-high school or higher	5 (5.56)
Previous occupation (N = 90)
Unemployed	50 (55.56)
Unskilled worker, semi-skilled worker	8 (8.89)
Skilled worker, arithmetic skill job	30 (33.33)
Semi-professional, professional	2 (2.22)
Marital status (N = 90)
Married	45 (50.00)
Widowed/divorced/separated	23 (25.56)
Unmarried	22 (24.44)
Family type (N = 90)
Nuclear	41 (45.56)
Joint	49 (54.44)
Number of family members (N = 90)
<5	19 (21.11)
⩾5	71 (78.89)
Mean ± SD	6.68 ± 3.17
Number of earning members (N = 90)
⩽2	69 (76.67)
>2	21 (23.33)
Socioeconomic status (BG Prasad Scale; N = 84)
Class I (⩾9130 INR)	8 (9.52)
Class II (4565–9129 INR)	17 (22.62)
Class III (2739–4564 INR)	32 (38.10)
Class IV (1369–2738 INR)	19 (20.24)
Class V (<1369 INR)	8 (9.52)

Profile of caregivers

It was found that nearly 94.45% of the caregivers in the families were untrained family members, while 2.22% were trained family members. In one family, the caregiver was a retired hospital staff member with experience in patient care, while in the other family, a nurse within the family was overseeing the care. Three families had hired caregivers; two were paid 15,000 rupees per month, and the third was paid 20,000 rupees per month. The majority (84.44%) of the caregivers were females, with a mean age of 46.48 ± 14.15 years. Among the caregivers, 30% were wives, 22.22% were mothers, and 20% were daughters-in-law. Most (83.33%) of the caregivers were unemployed (Table 2).

Table 2.

Distribution of sociodemographic characteristics of caregivers of the study participants in need of home-based palliative care.

Variable	Total (N = 90), n (%)
Presence of caregiver at home
Hired caregiver (maid)	3 (3.33)
Trained family member	2 (2.22)
Untrained family member	85 (94.45)
Gender
Male	14 (15.56)
Female	76 (84.44)
Age of primary caregiver (years)
20–29	9 (10.00)
30–39	24 (26.67)
40–49	18 (20.00)
50–59	17 (18.89)
60 and above	22 (24.44)
Mean ± SD	46.48 ± 14.15 years
Relation of the caregiver
Wife	27 (30.00)
Mother	20 (22.22)
Daughter-in-law	18 (20.00)
Husband	6 (6.67)
Son	6 (6.67)
Daughter	4 (4.44)
Maid	3 (3.33)
Other (grandchildren/sister-in-law/grandmother)	6 (6.67)
Occupation of caregiver
Unemployed	75 (83.33)
Employed	15 (16.67)

Medical profile of people requiring HBPC

Disease and symptom profile

The study population was found to have 53 different diseases (Supplemental Table 1). A list of their prescribed medicines was also collated, which contained 106 different medicines (Supplemental Table 2). Neurological disorders were the most common cause of chronic illness, accounting for 67.8% of all cases among the participants, followed by orthopedic conditions (8.9%) and old-age-related weakness with marked dependence (7.8%; Figure 2). Among neurological disorders, stroke with deficit was the most common cause of disability among 32.22% of the participants. Chronic NCDs like hypertension and diabetes mellitus were present among 44.44% and 18.89% of the participants, respectively.

Figure 2.

Distribution of the participants as per the primary diagnosis.

Pain was the predominant symptom, affecting more than half (51.11%) of the study participants, followed by weakness or tiredness (44.44%), low mood (15.56%), and shortness of breath (14.44%). A majority (61.11%) of the participants had lived with the illness for 5 years or less, and 30% had the illness for more than 10 years, with an average duration of illness being 8.57 years.

Disease severity

Nearly half (48.89%) of the participants had severe dependence, 37.78% had total dependence, while 13.33% had only moderate to slight dependence or were independent as per the Barthel Index. The average Barthel Index score of the people requiring HBPC was 33 ± 25.59 (Table 3). Most participants (42.22%) were wheelchair-bound, 33.33% required human support for mobility, and only 10% could walk independently but only within their house. Only four (4.44%) participants were found to have a pressure sore.

Table 3.

Disease severity of the study participants.

Variable	Frequency (%), n (%)
Barthel Index (N = 90)
Total dependence (⩽20)	34 (37.78)
Severe dependence (21–60)	44 (48.89)
Moderate dependence (61–90)	9 (10.00)
Slight dependence (91–99)	1 (1.11)
Independent (100)	2 (2.22)
Mean ± SD	33 ± 25.59
Ambulatory status (N = 90)
Predominantly wheelchair	38 (42.22)
Human support	30 (33.33)
Predominantly walker	4 (4.45)
Walking stick	9 (10.00)
No support	9 (10.00)
Pressure sore (N = 90)
Absent	86 (95.56)
Present	4 (4.44)

Healthcare-seeking behavior

Government health facilities were the major source of healthcare (45.56%), followed by private formal healthcare facilities (32.22%), while 23.33% were not under any regular treatment from any healthcare facility. It was also noted that 2.22% and 5.56% of the participants received care from informal healthcare providers and traditional medicine practitioners, respectively. Most (44.44%) of the participants visited healthcare facilities only in case of an emergency/acute illness, and 38.89% visited healthcare facilities once a month or more frequently. Other participants had follow-ups scheduled once in 2 months (7.78%), once in 3 months (7.78%), or once in 6 months (1.11%). None of the study participants knew about palliative care, end-of-life care, home care for bedridden individuals, or community-based care, and reported never hearing about such services before.

Out-of-pocket expenditure

On analyzing the health-related expenditure of the participants, it was found that the participants incurred a median expenditure of 800 (0, 2000) INR on medications, and the cost of travel to the healthcare facilities was 250 (0, 500) INR per month. Non-emergency medical care and investigations led to a median expenditure of zero rupees, with a maximum expenditure of 50,000 and 6000 INR, respectively (Table 4). The total monthly health-related expenditure averaged 2203.39 ± 2840.06 INR, with a median expenditure of 1018.34 INR. These healthcare costs represented a substantial proportion of participants’ income, with monthly drug expenses alone accounting for an average of 46.49% of per capita monthly income and 9.13% of total monthly family income. The total healthcare costs, including direct medical and non-medical care, accounted for an average of 58.56% of per capita monthly income and 11.11% of total monthly family income, with a maximum of 168.33% of the total monthly family income (Table 4).

Table 4.

Out-of-pocket healthcare expenditure and its proportion to household income.

Variable	Mean ± SD	Median (IQR)	Range (Min, Max)
Cost spent on drugs for the last month	1650.33 ± 2258.26 INR	800.00 (0.00, 2000.00) INR	0.00, 12,000.00 INR
Cost spent on seeing the doctor in the last 6 months for non-emergency care (N = 90)	911.11 ± 5399.01 INR	0.00 (0.00, 0.00) INR	0.00, 50,000.00 INR
Cost of travel to the treatment facility (N = 90)	341.67 ± 484.55 INR	250.00 (0.00, 500.00) INR	0.00, 3300.00 INR
Cost spent on investigations in the last 6 months (N = 90)	357.22 ± 1215.11 INR	0.00 (0.00, 0.00) INR	0.00, 6000.00 INR
Total healthcare costs incurred per month (N = 90)	2203.39 ± 2840.06 INR	1018.34 (300.00, 3000.00) INR	0.00, 15,000.00
Cost spent on drugs per month as a percentage of per capita income (N = 80)*	46.49 ± 90.28%	21.43 (0.00, 61.11) %	0.00%, 600.00%
Cost spent on drugs per month as a percentage of total monthly family income (N = 80)*	9.13 ± 22.95%	3.17 (0.00, 9.05) %	0.00%, 166.67%
Cost spent on travel to healthcare facility per month as a percentage of per capita income (N = 80)*	9.03 ± 11.57%	5.00 (0.00, 12.00) %	0.00%, 50.00%
Cost spent on travel to healthcare facility per month as a percentage of total monthly family income (N = 80)*	1.45 ± 1.91%	0.78 (0.00, 2.00) %	0.00%, 10.00%
Cost spent on non-emergency medical care per month as a percentage of per capita income (N = 80)*	1.64 ± 7.08%	0.00 (0.00, 0.00) %	0.00%, 50.00%
Cost spent on non-emergency medical care per month as a percentage of total monthly family income (N = 80)*	0.30 ± 1.41%	0.00 (0.00, 0.00) %	0.00%, 10.00%
Cost spent on investigation per month as a percentage of per capita income (N = 80)*	1.40 ± 6.34%	0.00 (0.00, 0.00) %	0.00%, 50.00%
Cost spent on investigation per month as a percentage of total monthly family income (N = 80)*	0.23 ± 1.18%	0.00 (0.00, 0.00) %	0.00%, 10.00%
Total costs incurred per month as a percentage of per capita income (N = 80)*	58.56 ± 103.79%	31.00 (8.65, 85.45) %	0.00%, 750.00%
Total costs incurred per month as a percentage of total monthly family income (N = 80)*	11.11 ± 25.01%	4.99 (1.17, 11.63) %	0.00%, 168.33%

^*N = 80, as six participants were unwilling to share their monthly family income, and four had zero income.

Impact of illness on the family

It was found that 62.22% of the families suffered due to the participant’s illness, with 42.22% experiencing deterioration in the quality of their diet, 42.22% stopped celebrating festivals and family events, and the caregivers’ jobs were affected in 31.11% of the families. Other impacts included members dropping out of education (14.44%) and neglect of the illness of other family members (13.33%). In addition to the above, several other impacts were reported by the families. These included displacement from their village due to financial pressure, being forced to change houses due to inability to pay rent, and a significant decrease in leisure and recreational activities. One family reported that they reduced medication doses due to cost, and one family reported a suicide attempt related to the disease burden on the family. There were also reports of household members using water and electricity judiciously, and a family reported that their child was not able to start schooling due to the ongoing illness in the family.

It was found that 34.44% of the families had incurred debt, and 87.10% of the time, it was illness-related. The families with debt had an average debt of nearly 1,80,000 INR, ranging from a minimum of 7500 INR to a maximum of 5,00,000 INR. Nearly half of families (48.88%) had difficulty affording food, and five (5.56%) families could not afford their daily meals. Only 37.78% of the families could afford the medications for the treatment without difficulty, and 21.11% could not afford their medicines. It was found that 23.33% of caregivers had to quit or take extended/more frequent leaves from work.

Distance from the healthcare facility

There was a higher density of participants in Gokalpuri (46.67%) than in Ganga Vihar (41.11%) and Gokalpuri Village (12.22%). The mean distance from each household to the UHC was 550.22 ± 287.51 m, with a median of 471.50 m. On subgroup analysis, children under 15 years (n = 7) had the highest mean distance of 595.57 ± 167.87 m, followed by adults aged 15–59 years (n = 29) at 582.14 ± 360.56 m, while elderly patients aged 60 years and above (n = 54) had houses at a mean distance of 527.20 ± 256.06 m from the UHC. Female patients (n = 41) lived farther from the dispensary, with a mean distance of 599.68 ± 348.59 m compared to males (n = 49) at 508.84 ± 219.46 m.

Considering participants’ primary diagnoses, a patient with psychological disorder (n = 1) was located the farthest from UHC at a distance of 965 m, followed by those with orthopedic conditions (n = 8) at 704.25 ± 576.34 m, while those with cardiovascular disease (n = 2) had the shortest mean distance at 346.50 ± 50.20 m (Table 5). The participants with neurological conditions, like stroke, seizure disorders, mental retardation, hypoxic ischemic encephalopathy, etc. (n = 61), were located at an average distance of 531.97 ± 231.84 m from the UHC.

Table 5.

Distance of the households from the dispensary and its subgroup analysis by age, gender, and primary diagnosis of the participants.

Variable	Frequency (%)	Mean ± SD	Median (Q1, Q3)	Min, Max
Distance
Overall	90 (100.00)	550.22 ± 287.51	471.50 (375.00, 696.00)	190.00, 2000.00
Distance by age
<15	7 (7.78)	595.57 ± 167.87	629.00 (420.00, 730.00)	390.00, 858.00
⩾60	54 (60.00)	527.20 ± 256.06	463.50 (367.00, 738.00)	190.00, 1000.00
15–59	29 (32.22)	582.14 ± 360.56	494.00 (372.00, 673.00)	231.00, 2000.00
Distance by gender
Female	41 (45.56)	599.68 ± 348.59	513.00 (377.00, 836.00)	230.00, 2000.00
Male	49 (54.44)	508.84 ± 219.46	470.00 (372.00, 608.00)	190.00, 1000.00
Distance by primary diagnosis
Cardiovascular disease	2 (2.22)	346.50 ± 50.20	346.50 (311.00, 382.00)	311.00, 382.00
Lung disease	4 (4.44)	531.25 ± 327.70	439.00 (319.00, 743.50)	247.00, 1000.00
Malignancy	3 (3.33)	393.33 ± 243.34	277.00 (230.00, 673.00)	230.00, 673.00
Neurological disorder	61 (67.78)	531.97 ± 231.84	473.00 (392.00, 629.00)	190.00, 1000.00
Old age-related weakness and marked dependence	7 (7.78)	633.00 ± 272.90	520.00 (462.00, 918.00)	247.00, 1000.00
Orthopedic condition	8 (8.89)	704.25 ± 576.34	454.00 (371.00, 869.00)	246.00, 2,000.00
Psychological conditions	1 (1.11)	965.00 ± NA	965.00 (965.00, 965.00)	965.00, 965.00
Renal disease	4 (4.44)	510.50 ± 327.50	435.50 (254.00, 767.00)	234.00, 937.00

Access to social security services

A majority (73.33%) of the participants had a ration card, 50.00% were beneficiaries of a pension scheme, only 15.56% had access to public health insurance, and no one was availing of any private health insurance. Among the participants receiving pensions, 18.89% were receiving disability pensions, 16.67% were receiving old age pensions, 3.33% were receiving widow pensions, and 11.11% were receiving job pensions. Only one of the participants, with cancer, was receiving assistance (HBPC) from an NGO that focuses on cancer palliative care. While examining overall access to social security schemes (ration cards, pensions, public health insurance), 46.67% of the population had access to at least one scheme, while 41.11% had access to any two, 3.33% had access to all three (ration card, insurance, and pension), and 8.89% had no access to any social security measures.

Discussion

The community-based screening and assessment for HBPC in a population of 43,267 individuals identified that 0.21% (approximately 2 per 1000 people) needed home-based care. Our need estimation was similar to other studies from northern India and lesser than the need estimated in studies from southern India. ¹¹ While population-based needs specifically for HBPC are lacking, in 2004, the WHO estimated that approximately 4 per 1000 adults (over 15 years) and 1 per 1000 children (under 15 years) worldwide required palliative care. ²² Knaul et al. ²³ reported that the global burden of serious health-related suffering requiring palliative care by 74% between 1990 and 2021, with 80% of that burden being concentrated in low- and middle-income countries (LMICs). The burden was reported to be rising almost twice as rapidly in LMICs compared to high-income countries (HICs). ²³

The age distribution of our study population was similar to other studies from India, with a majority being above 60 years of age. ¹¹ Most of these individuals were elderly males, illiterate (51.11%), and unemployed at the time of the onset of the disabling illness (55.56%). Many participants lived in joint families, with 78.89% having five or more family members. This was in contrast to studies from southern India, where the majority of the patients lived in nuclear families.^14,24 Similar to other need assessment studies, socioeconomic status varied, with most participants in the lower-middle class range (class III). ¹¹ Socioeconomic status is particularly important to study in this vulnerable population, as it has been found to correlate with greater end-of-life suffering. ²⁵ Similar demographics of patients in need of palliative care have also been reported from HICs with a high need in the elderly population. ²⁶

Assessment of the primary caregivers’ profiles revealed that caregivers were predominantly untrained family members (94.45%), mostly female (84.44%), with wives and mothers as the most common primary caregivers. This gender disparity in caregiver burden has been previously highlighted globally and in India.^9,27,28 Most caregivers were also unemployed (83.33%). In our study, only two families had trained caregivers, and only one patient was receiving HBPC from a non-profit palliative care organization, highlighting the poor access to trained caregivers and appropriate home-based care. This is despite the NPPC in India that recommends the provision of HBPC for those in need from the public health system. ²⁹

The health profiles of participants were dominated by neurological disorders (67.8%), with stroke being the most common cause of disability. The inclusion of only patients in need of HBPC could also be the reason why neurological diseases were the most commonly reported illness in our study population. The global evidence, however, highlights the need for palliative care to be dominated by chronic diseases such as cardiovascular diseases (38.5%), cancer (34%), chronic respiratory diseases (10.3%), AIDS (5.7%), and diabetes (4.6%). ⁶ In a study from Slovenia, 94% of the patients receiving palliative care were reported to have cancer. ³⁰ Similarly, in Ireland, cancer and cardiovascular diseases were identified to make up for almost half of the palliative care needs during the final stages of life. ²⁶ Our findings, however, were similar to another study from Delhi that also looked at the need for home-based care, which also reported neurological deficit and osteoarthritis with marked dependence to be the top two diseases requiring palliative care. ⁹ Similarly, another study from the North Indian state of Haryana that looked at the HBPC needs reported neurological deficit as the most common condition requiring palliative care. ³¹ However, studies from the southern part of the country reported a different disease profile with cardiovascular disease and old age with frailty being reported as the leading diseases requiring HBPC.^9,14,24 This could be due to a higher mean age of the population in South Indian states, since cardiovascular diseases and frailty are more common in old age. ³⁰ This highlights the importance of regional evidence generation for palliative care needs globally at the national and subnational levels.

On comparing the list of primary diagnoses with the Lancet serious health-related suffering (SHS) study, it was found that hemorrhagic fevers, HIV, leukemia, liver diseases, birth trauma, congenital malformations, and malnutrition were not identified among the study participants. Also, old age-related weakness and psychological conditions were identified to be contributing to the need for palliative care among our study participants, which were not reported by the Lancet SHS study. ³²

The list of medications taken by the study participants was compared with the essential medicines recommended by several key organizations: the International Association of Hospice and Palliative Care (IAHPC), ³³ the World Health Organization Essential Drug List (WHO EDL), ³⁴ the National List of Essential Medicines (NLEM) of India, ³⁵ and the Delhi Essential Medicines List (EML) for primary healthcare centers. ³⁶ The comparison revealed that the NLEM India was the most comprehensive, encompassing 50 out of 106 medicines, followed by the Delhi EML with 40 medicines, and the WHO EDL, which covered 35 medicines. Notably, only 8 out of the 33 essential medicines recommended by the IAHPC were prescribed to the study participants. While pain (51.11%) and shortness of breath (14.44%) were two common symptoms reported by the study participants, access to opioids, which can be used both for pain and shortness of breath, was poor in the community. No pharmacy in the entire region of the population under study was licensed to stock and dispense oral morphine, nor was it available at the government-run or any private primary healthcare centers in the region. At the community level, this reflected the larger problem of poor access to opioids for symptom control in India due to strict regulations, poor awareness, and misconceptions among healthcare providers. ³⁷

Approximately 62% of families endure significant hardship, with members compelled to abandon their education, leave their jobs, forgo festival celebrations, and experience a decline in the quality of their diets. This particular aspect of the challenges faced by patients’ families and caregivers has received limited attention, as most research has primarily focused on the psychosocial impact of illness. Regarding social security, while most had access to ration cards (73.33%) and pensions (50%), fewer had access to public health insurance (15.56%). Access to health insurance in this population was significantly lower compared to the general population of Delhi (25%) as per the National Family Health Survey—5. ³⁸ A sizable portion of our study population (8.89%) had no access to any social security scheme, highlighting their socioeconomic vulnerability. Studies from other regions of the world have highlighted the impact of socioeconomic deprivation on the quality of life, which further impacts the quality of death of individuals, stressing the need for healthcare models that meet the needs of those experiencing socioeconomic deprivation at the end-of-life. ³⁹

The total direct (medical and non-medical) out-of-pocket healthcare expenditure comprised an average of 58.56% of their per capita income and 11.11% of their total monthly family income, with the cost of medicines being the biggest expense. Many families also faced socioeconomic challenges, such as debt and difficulty affording food, with caregivers often having to reduce or quit work due to caregiving responsibilities. A study done in the United States had similar findings as our study, with participants suffering from chronic life-limiting illnesses suffering financial burdens due to their illness. They also identified expenditure on medicines being the driving force for the financial burden. ⁴⁰

We also assessed the distance of the patients from the health center. With the center catering to a population of nearly 45,000 people, a majority of the patients were found to be within walking distance of around 500 m from the center. Under the National Health Mission in India, there should be one primary healthcare center for every 20,000–30,000 population. However, despite geographical proximity and the existence of a national program, all but one of the study participants were not receiving HBPC. This could be due to the lack of demand from the community due to a lack of awareness as highlighted in our study. The proximity of health centers in densely populated urban areas makes it possible to deliver primary HBPC to patients in the community through the primary healthcare center. However, there is a need to raise awareness within the community about the benefits of palliative care.

While some of our findings are similar to another study from Delhi, ⁹ our study offers more depth to the understanding of the needs of the patients by reporting detailed socioeconomic assessment, and impact of disease on the families both in terms of out-of-pocket expenditure and impact on access to food, education, and other services along with the geospatial distribution of the patients around the healthcare center. These valuable insights can help guide policymakers in creating holistic policies for this vulnerable population. First, it highlights the substantial burden of diseases requiring palliative care in an urban resettlement colony in Delhi. Despite being the national capital, the unmet need for palliative care remains significant, revealing a critical gap in the implementation of the NPPC. Under this program, Accredited Social Health Activists (ASHAs), India’s community health workers, are responsible for identifying bedridden individuals and others in need of palliative care within the community. ASHAs are expected to refer these patients to the Community Health Officer, provide basic home-based care, and ensure continuity of care by accompanying patients to their initial healthcare visits. However, several barriers have hindered the successful implementation of this program, including a lack of dedicated funding, poor awareness of palliative care among both the community and healthcare providers, and the overburdening of community health workers. Addressing these barriers is crucial for the successful execution of the NPPC. Our study also supports the possibility of the implementation of NPPC through the Departments of Community Medicine of Medical Colleges in India. As of 2024, India has 780 medical colleges. Therefore, a significant proportion of the population could be provided HBPC through these medical colleges and their peripheral primary care centers run by medical officers and community medicine physicians. This could also be done through public-private partnerships between medical colleges and not-for-profit palliative care organizations.

Second, our study reveals the socioeconomic vulnerabilities of patients requiring palliative care. To address these challenges, there is an urgent need to expand social security schemes, such as health insurance, pensions, and food security, to include individuals from lower socioeconomic backgrounds in need of palliative care. Since the cost of medications presents a significant financial burden for families of patients with chronic diseases requiring palliative care, it is essential to extend the government-funded health insurance schemes to cover outpatient care, including medication costs. Ensuring the availability of good-quality generic medications would also help in reducing the financial burden of the disease. This could be ensured by improving access to the Indian government’s Janaushadhi Kendras, which are pharmacies selling only government-approved generic medicines.

Third, empowering caregivers of patients with chronic life-limiting illnesses through vocational rehabilitation is essential. Additionally, new social security programs should be introduced to protect families from generational poverty by ensuring continued education for children in families facing financial hardship due to chronic illnesses.

Lastly, it is crucial to ensure a dignified death for patients with chronic life-limiting illnesses through advance care planning and the creation of Advance Medical Directives (AMDs). AMDs would also protect families from further financial strain caused by unnecessary medical interventions during the final stages of life.

This study has several strengths. First, we conducted a household-to-household survey, the largest of its kind in India, with a high response rate to estimate the need for palliative care in the community. Second, the need for palliative care was assessed by a trained physician using standardized tools. Third, this is the first study to assess the need for palliative care and evaluate the disease’s socioeconomic impact on the family. Despite these strengths, the study has some limitations. Our study reports the need for only HBPC, and not the entire spectrum of palliative care needs in the community, which might have a higher burden than our reported needs. We used the Barthel Index, which is a surrogate marker for palliative care needs, and we also did not estimate the SHS experienced by the study population. However, the decision to estimate only the HBPC needs was taken after considering the limited resources at the community level and the pressing need to prioritize the needs of home or bed-bound patients. Our methodology was also guided by our aim to generate data comparable to existing studies from India to push for policy action. We did not estimate the indirect out-of-pocket expenditure caused by the disease burden, nor did we assess the impact of the disease on the psychological well-being of the patient and caregiver. There is also a possibility of information bias regarding expenditure on healthcare and family income. The study’s findings may have limited generalizability as it was conducted in an urban area of northeast Delhi, which may not be representative of other settings.

Conclusion

The need for HBPC was found to be 2 per 1000 people, with the majority of patients being elderly males, suffering from neurological disorders, and requiring significant assistance. The socioeconomic impact of their illness was found to be substantial, with most families facing financial strain, particularly in covering healthcare costs. Despite some access to social security services, a significant proportion of participants remain without adequate support. The findings highlight the need for improved healthcare access, better social security measures, and comprehensive support for families caring for individuals with chronic illnesses. There is a need for more regional-level HBPC needs assessment in India to design relevant policies.