Feasibility of stratification of households based on health needs in Mugalur village, Anekal taluk—An exploratory study

ABSTRACT

Introduction:

Stratifying households helps identify those at higher health risk and supports focused interventions. This study assessed the feasibility of creating a household stratification tool in rural Mugalur village, Anekal Taluk, to enhance community health services by classifying households according to their healthcare needs.

Methodology:

A stratification tool was developed by a brainstorming session with public health professionals, identifying key health conditions across age groups, along with relevant social and environmental risk factors. These were categorised into high, medium, and low priority, represented by red, yellow, and green colour codes, respectively. High-priority households were scheduled for monthly visits, medium-priority households once every two months, and low-priority households once every six months. The tool was administered to all households in the village.

Results:

Among 332 households, 18.1% were classified as red, 45.8% as yellow, and 36.1% as green. Chronic conditions were more common among older adults, while the presence of young children or pregnancy often increased household priority. Larger family size and lower socioeconomic status were significantly associated with higher priority status. Notably, several high-priority households were not receiving the recommended frequency of health worker visits, indicating gaps in service delivery.

Conclusion:

The stratification tool was feasible to implement in this rural setting. By identifying households with greater health needs, it can support community health workers in optimising visit schedules and improving the efficiency and equity of home-based care.

Introduction

India’s rural health system serves nearly 65% of the national population, approximately one billion people,[1] with a persistent burden of communicable diseases, malnutrition, and maternal–child health challenges, alongside a rapidly rising prevalence of non-communicable diseases (NCDs), all within a context of longstanding infrastructural constraints. Recent geocoded analyses confirm that primary health centres (PHCs) routinely exceed recommended population norms, serving an average of 33,800 individuals against the stipulated 30,000, with villages located a mean distance of 5.49 km from the nearest facility.[2] Compounding these access barriers, PHCs and sub-centres continue to face acute workforce shortages, inconsistent availability of essential medicines and diagnostics, limited staff training, and geographical challenges that impede reliable service delivery.[3]

In this context of constrained resources and growing multimorbidity, primary healthcare teams require simple, operational tools to distinguish households with greater health needs from those at lower risk, so that limited time and services can be directed more efficiently. Existing community health needs assessments in India and other low- and middle-income settings have largely generated aggregate village or facility-level data, but have not translated into systematic household-level prioritisation for outreach and follow-up.[4] NHM guidelines envisage one ASHA per 1000 population, yet monitoring reports show that in several states, each ASHA actually covers well over this norm.[5] In addition to high caseloads, a 2024 survey of 150 ASHAs found that 92% reported being overloaded with work, with associated burnout, high stress, and reduced job performance.[6] A 2022 time motion study concluded that ASHAs are ‘overburdened’ and effectively work seven days a week due to continuous programme and survey demands. Also, they were observed to focus predominantly on their key duties, including referral facilitation, promotion of institutional deliveries, and delivery of essential primary healthcare services, largely driven by performance-based incentives.[7] In this context, approaches that help ASHAs prioritise and focus their activities on households with the greatest health needs, without adding to workload, are especially important.

Household-level stratification provides a systematic approach to prioritising health services for vulnerable families, including those with chronic illnesses, elderly members, pregnant women, young children, and households facing socioeconomic disadvantage. However, the feasibility of implementing such stratification within routine public health practice remains insufficiently explored. A technical paper by the World Health Organisation identifies risk stratification and household mapping as core functions of community health worker (CHW) programmes, enabling the identification and follow-up of households at risk and the targeted provision of home visits to those most in need.[8] For practitioners, particularly ASHAs and primary healthcare teams in resource-constrained rural settings, a simple household stratification tool can support prioritisation of home visits, screening, and follow-up for households with higher health needs, thereby improving the efficient use of limited time and human resources.[9] For patients and families—especially those with chronic diseases, high-risk pregnancies, young children, or socioeconomic vulnerability—such risk-based stratification may facilitate timely care, early problem identification, and better linkage to government schemes, ultimately enhancing continuity, quality, and equity of primary health care at the household level.[10,11] These functions are explicitly recommended to enhance efficiency and equity in primary healthcare delivery, particularly in resource-constrained settings.

In this context, the present exploratory study was undertaken to assess the feasibility of stratifying households by health needs in Mugalur village, Anekal taluk, to support risk-based prioritisation and equitable primary healthcare delivery. Mugalur is a rural village in the Bangalore Urban district, Karnataka, with a population of approximately 1,535 individuals across 332 households. Preliminary community engagement identified key health concerns, including acute respiratory illnesses (ARIs) among children, occupational health risks among agricultural workers, and chronic diseases among the elderly. Despite the relevance of household stratification frameworks in resource-limited settings, their operational feasibility in routine community health practice has not been systematically evaluated in this region. This study, therefore, aimed to assess the feasibility of household health-needs stratification, describe the distribution of households across health-need priority categories, examine associations with sociodemographic factors, and assess utilisation of government health schemes.

Methodology

This exploratory, descriptive, cross-sectional study was conducted during March–April 2018 in Mugalur village, Anekal taluk, Bangalore Urban district, Karnataka, India. Mugalur is a rural field practice area of the Department of Community Medicine, St. John’s Medical College. The village comprises 332 households with a total population of 1,535 individuals and represents a predominantly agrarian and daily-wage-based rural community. Primary healthcare services in the village are delivered through a Health and Wellness Centre Sub-centre and by ASHAs.

The study population consisted of all households in Mugalur village. For this study, a ‘household’ is usually a group of persons who normally live together and take their meals from a common kitchen unless the exigencies of work prevent any of them from doing so.[12] Households were eligible for inclusion if they had been residing in the village for at least six months and if the head of the household or the primary decision-maker provided informed consent. Temporary or transient households were excluded.

A universal (census) sampling approach was adopted, and all 332 households in Mugalur village were included in the study. This approach was chosen because the target population was small, clearly defined, and logistically feasible to cover in its entirety. The use of a census method eliminated sampling error and enabled a precise village-level assessment of population characteristics and the distribution of health needs.

The household stratification tool was developed through an iterative and collaborative process. A brainstorming session was held with senior public health professionals and epidemiologists, followed by consultation with senior faculty from the Department of Community Medicine. Health conditions and social risk factors relevant to the local context were identified and organised by age group. These indicators were then systematically prioritised and classified into three health-need categories based on their public health significance and disease burden, preventability and treatability, frequency and severity, and the potential for targeted intervention at the community level. The final output of this process was a colour-coded stratification matrix that grouped age-specific and social/environmental indicators into high (red), medium (yellow), and low (green) priority categories, each linked to a recommended frequency of home visits (monthly, once in three to six months, or routine). This matrix, shown in Figure 1, served both as a decisionsupport tool for health workers and as a visual aid for communicating household risk status during field implementation.

Figure 1.

The colour-coded matrix used for household stratification

Data were collected using a pre-tested, structured household questionnaire administered through face-to-face interviews conducted in Kannada by trained medical interns from the Department of Community Medicine. The questionnaire captured information on household composition, age and sex distribution, education and occupation of the household head, socioeconomic status using the Modified B.G. Prasad scale, and family structure. It also recorded the health status of all members, including self-reported chronic diseases such as diabetes and hypertension, disabilities, recent illnesses, and hospitalisations, as well as maternal and child health indicators such as pregnancy status, high-risk pregnancy features, immunisation status of under-five children, and nutritional status.

Government scheme utilisation was assessed by asking about enrolment in services such as Anganwadi nutrition and early childhood care, the Thayi card for maternal health, pension schemes, Pulse Polio campaigns, Ayushman Bharat-Arogya Karnataka health insurance, and Janani Suraksha Yojana, and by verifying available documentation where possible. The distance to the nearest health facility was recorded based on household reports and local mapping, categorised into four groups: within one kilometre, one to five kilometres, five to ten kilometres, and more than ten kilometres. Immediately after each interview and assessment, the stratification tool was applied, and a colour code—red for high priority, yellow for medium priority, and green/white for low risk—was assigned to the household based on the highest-priority member present.

Data from completed questionnaires were entered into Microsoft Excel and checked for completeness and internal consistency by comparing with the original forms when discrepancies were identified. Descriptive statistics, including frequencies and percentages, were used to summarise the distribution of households across priority categories, age-specific morbidity patterns, socioeconomic characteristics, and scheme utilisation. Associations between household priority category and sociodemographic variables, such as family size, socioeconomic status, education and occupation of the household head, family integrity, and frequency of health worker visits, were examined using the Chi-square test, with Fisher’s exact test applied when expected cell counts were less than five and a P value of less than 0.05 considered statistically significant. All statistical analyses were performed using JAMOVI.

Ethical approval for the study was obtained from the Institutional Ethics Committee of St. John’s Medical College, Bangalore, and written informed consent was obtained from all participating household heads or primary decision-makers before data collection. Confidentiality was maintained by assigning unique identification numbers to households, omitting personal identifiers from the analytic dataset.

Results

All 332 households in Mugalur village were successfully enrolled, yielding a 100% response rate and covering a total population of 1,535 individuals. The age distribution was 0–5 years (7%), 6–19 years (22%), 20–45 years (48%), 46–59 years (9%), and ≥ 60 years (14%). The population included 819 males (53.3%) and 716 females (46.7%), with a sex ratio of 0.87.

Among the households, 60 (18.1%) were classified as high priority (red), 152 (45.8%) as medium priority (yellow), and 120 (36.1%) as low priority (green), indicating substantial heterogeneity in health needs [Figure 2].

Figure 2.

Distribution of households by priority category

The age-wise distribution of major health conditions is shown in Table 1. Among children aged between 0 and 5 years (n = 112), recurrent ARIs were most common (27.7%), followed by incomplete immunisation (9.8%), undernutrition (4.5%), and chronic ear infection (2.7%). Among adolescents aged between 6 and 19 years (n = 333), recurrent illness (7.2%) and anaemia (3.0%) were most frequent. Among adults aged between 20 and 45 years (n = 732), diabetes, hypertension, and smoking were each reported in 2.7%, while alcohol dependency was reported in 2.1%. In adults aged between 46 and 59 years (n = 136), diabetes (22.1%), and hypertension (19.9%) predominated. Among the elderly (n = 206), hypertension (38.3%) and diabetes (29.1%) were most prevalent, with common sensory impairments.

Table 1.

Age-wise distribution of health condition

	Health condition	n	%
Children 0–5 years (n=112)	Recurrent ARI (≥3 episodes/year)	31	27.67
	Incomplete immunization	11	9.82
	Undernutrition	5	4.46
	Chronic ear infection	3	2.67
Children 6–19 years (n=333)	Recurrent illnesses	24	7.20
	Anaemia	10	3.00
	Disabilities	4	1.20
	School dropout	2	0.60
	Mental illness	1	0.30
Adults 20–45 years (n=732)	Diabetes	20	2.70
	Hypertension	20	2.70
	Alcohol dependency	15	2.05
	Smoking	20	2.70
	Recent ill-health/hospitalisation	6	0.82
	Gynaecological diseases	4	0.55
	High-risk pregnancy	4	0.55
	Mental illness	3	0.41
Adults 46–59 years (n=136)	Diabetes mellitus	30	22.06
	Hypertension	27	19.85
	Chronic illness	14	10.29
	Alcohol use	12	8.82
	Smoking	8	5.88
	Recent hospitalisation	2	1.48
	Cardiac disease	1	0.74
Elderly≥60 years (n=206)	Hypertension	79	38.34
	Diabetes mellitus	60	29.12
	Visual impairment (cataract)	27	13.11
	Hearing impairment	21	10.20
	Chronic illness	13	6.31
	Bedridden/severely disabled	5	2.43
	Mental illness	2	0.97

ARI=Acute respiratory infection

Figure 3 shows the socioeconomic status of the study households; most belonged to the middle (41.3%) and upper-middle (41.9%) classes. Table 2 presents health worker visits, and scheme utilisation. Most households (71.4%) were located within one kilometre of the sub-centre, and 68.1% reported monthly health worker visits. Utilisation of Anganwadi (48.49%) and Thayi card (45.78%) was relatively high, whereas Janani Suraksha Yojana (2.71%) and Ayushman Bharat-Arogya Karnataka (6.94%) enrolment was low.

Figure 3.

Distribution of households based on socioeconomic factors

Table 2.

Distribution of households by, health worker visits, and scheme utilisation

Category	Sub-category	Number of households	%
Health worker visit frequency	Once a month	224	68.1
	Once every two months	68	19.9
	Once every six months	36	10.8
	Once every year	4	1.2
Scheme/programme	Anganwadi	161	48.49
	Thayi card	152	45.78
	Pension schemes	146	44.58
	Pulse Polio	134	40.36
	Ayushman Bharat-Arogya Karnataka	23	6.94
	Janani Suraksha Yojana	9	2.71

Table 3 summarises associations with household priority status. Household size was significantly associated with priority category (χ² =17.354, df = 2, P = 0.001, Cramér’s V = 0.2286), with larger households more frequently classified as high or medium priority. Socioeconomic status was also significantly associated (χ² =15.40, df = 4, P = 0.004, Cramér’s V = 0.152).

Table 3.

Association health-need priority categories with family size and socioeconomic status

Category	Family size	High priority	Medium priority	Low priority	P
Family size	≤4 members	10 (12%)	47 (55%)	29 (34%)	P=0.001
	>4 members	50 (20%)	73 (30%)	123 (50%)
	Total	60 (18.1%)	152 (45.8%)	120 (36.1%)
Socioeconomic status	Lower	3 (9.7%)	7 (23%)	21 (68%)	P=0.004
	Middle	35 (26%)	43 (31%)	59 (43%)
	Upper	22 (13%)	70 (43%)	72 (44%)
	Total	60 (18.1%)	152 (45.8%)	120 (36.1%)

No statistically significant associations were observed between household health-need priority category and the gender of the household head (P = 0.81), family integrity (P = 0.68), or the age (P = 0.63), education (P = 0.57), and occupation (P = 0.51) of the household head. Similarly, no significant association was found with the frequency of health worker visits (P = 0.28) or distance to the nearest health facility (P = 0.53).

Discussion

This community-based exploratory study in rural South India successfully demonstrated the operational feasibility of a colour-coded household stratification tool and identified substantial heterogeneity in health needs within a largely low-income population. The findings reveal that 18.1% of households in Mugalur village were classified as high-priority (red), 45.8% as medium-priority (yellow), and 36.1% as low-priority or no-risk (green), indicating that the stratification approach effectively differentiated households across a spectrum of health vulnerabilities. This distribution has direct implications for resource allocation in resource-constrained rural primary healthcare settings, where the traditional approach of uniform service delivery cannot accommodate the multifaceted health challenges that exist within communities.[13]

In this study, both household size and socioeconomic status were significantly associated with health-need priority category, with effect sizes indicating small-to-moderate but meaningful relationships (Cramér’s V = 0.23 for family size and 0.15 for SES). Larger households were more often classified as high- or medium-priority, which is consistent with evidence that crowding and bigger family size are linked to higher burdens of infectious disease, resource strain, and poorer self-rated health, particularly in low-resource settings. The association between lower SES and higher priority category aligns with well-documented socioeconomic gradients in morbidity, life expectancy, and access to quality care in rural India, where poorer households are more likely to experience illness and to receive lower quality services when they do seek care.[14] Taken together, these findings support the inclusion of simple sociodemographic markers, such as family size and SES, within household stratification tools to help identify clusters of vulnerability, while recognising that they should complement, rather than replace, direct assessment of clinical and social risk. Studies from rural Uttar Pradesh have shown that ASHA home visits during pregnancy and the perinatal period improve institutional delivery, care-seeking, and equity between socio-economic groups, highlighting the value of structured outreach to vulnerable households.[15]

Within this overall pattern of household vulnerability, the age-specific morbidity profile reveals important dimensions of the epidemiological transition in this setting. Among children 0–5 years, 27.67% were reported by caregivers to have recurrent episodes of ARI, highlighting that preventable infectious conditions remain important contributors to early childhood morbidity in this setting. The ARI proportion is lower than the 40–55% one-month ARI prevalence reported in several rural community-based studies,[16] which typically record any ARI episode over a short recall period, suggesting that differences in case definition and recall interval, focusing here on children perceived as having frequent or repeated illness, likely contribute to the lower estimate. At the same time, the persistence of nearly 10% incomplete immunisation despite functioning government programmes indicates residual gaps in routine service delivery and defaulter tracking. According to NFHS5, in Karnataka, 84% of children aged between 12 and 23 months are fully immunised, 13% are partially immunised, and 3% are not immunised at all, indicating that approximately one in six children remain incompletely immunised despite high overall coverage.[17] In our study, 9.82% of children under-five were incompletely immunised, which is broadly consistent with these state-level data and suggests that even in a village with functional services, a residual subgroup of children continues to miss one or more doses. It supports the need for more targeted outreach to high-risk households identified through stratification. Community-based surveys in rural Karnataka, including Kolar district, report high prevalence of hypertension and diabetes among adults and older people, similar to the substantial NCD burden observed among households in Mugalur.[18] In our study, diabetes and hypertension each affected 2.70% of adults aged between 20 and 45 years, rising to 22.06% and 19.85% in those 46–59 years and 29.12% and 38.34% among those aged ≥60 years. The hypertension figures are comparable to rural geriatric studies reporting one-third to one-half of the elderly with elevated blood pressure, whereas the diabetes prevalence in the ≥60 year group appears higher than the 12–24% often reported in similar settings, which may reflect differences in case ascertainment, local lifestyle factors, or recent screening efforts. These patterns underscore the need to strengthen sub-centre level screening and long-term management of hypertension and diabetes, with a particular focus on the late middle-aged group as a critical window for preventing complications in older age.

In adults and older persons, NCDs, especially diabetes and hypertension, dominated the morbidity profile, mirroring patterns described from rural Karnataka and other South Indian settings. In our study, diabetes and hypertension each affected 2.70% of adults aged between 20 and 45 years, rising to 22.06% and 19.85% in those 46–59 years and 29.12% and 38.34% among those aged ≥ 60 years. These elderly hypertension figures are comparable to reports from rural Karnataka, where approximately one-third to over half of the geriatric population have elevated blood pressure, but the diabetes prevalence in our ≥ 60 year group is higher than the 12–26% reported in several rural Karnataka and South Indian geriatric cohorts.[19] This suggests a particularly high chronic disease burden in this village and reinforces the need to prioritise NCD screening and long-term management at subcentre level. Kshatri et al. (2025)[20] developed the Elderly Health Status Assessment and Screening (EHSAS) tool, a brief CHW-administered screening tool to identify older adults requiring comprehensive geriatric assessment. Similarly, studies on high-risk pregnancies in India have demonstrated that risk stratification can be integrated into routine care to prioritise follow-up of high-risk women.[21] Unlike studies such as EHSAS that focus on specific population groups, this work applies community-based, CHW-delivered assessment at the household level across life-course domains. Unlike high-risk pregnancy studies that use risk stratification for a single clinical cohort, this study extends a simple green–yellow–red framework to entire households to guide overall home-visit prioritisation. Unlike population-level NCD risk banding approaches, it operationalises colour-coded risk at the micro (household) level, demonstrating the feasibility of integrating such a tool into routine field practice and ASHA-led outreach planning. The study’s strengths include its universal sampling of all households in the village, providing a complete population snapshot free from selection bias, and the development of a simple, teachable stratification tool directly applicable to routine health worker practice. However, limitations must be acknowledged: This exploratory study was conducted in a single rural village and used a cross-sectional design, which limits the generalisability of the findings and prevents assessment of causal relationships or prediction of future health outcomes. The stratification tool relied on selfreported morbidity and household head recall without clinical verification or biomarker data, so some conditions may have been under- or over-reported. The colour-coded priority categories were applied by study staff under close supervision in a teaching institution field practice area, and feasibility, fidelity, and acceptability may differ when implemented at scale by routine health workers with competing responsibilities.

This household-level stratification approach carries significant implications for public health systems and practice. For frontline health workers and service providers, the tool offers a feasible mechanism to concentrate preventive and curative efforts on the minority of households bearing a disproportionate share of morbidity burden, thereby optimising the use of constrained human resources through need-based rather than uniform allocation of home visits, screening, and clinical follow-up. For health system managers and policy-makers, risk-stratified visiting schedules are consistent with established epidemiological principles of targeted intervention and align with global and national priorities for comprehensive, equitable primary health care. The systematic review of CHW programmes has identified household mapping and population risk stratification as core functions of CHW programmes, emphasising their role in improving both service efficiency and health equity.[22] Integration of household health-needs stratification into routine ASHA and sub-centre data systems could be accomplished with minimal additional resource investment through mobile-based applications or simplified paper tools, enabling real-time risk assessment and visit planning at the point of contact. However, future research should include prospective cohort or implementationresearch designs to assess whether stratified visiting schedules actually improve service coverage, disease control, and equity in health outcomes compared with usual practice. Multi-site studies in diverse rural and peri-urban settings are needed to validate and refine the prioritisation criteria, including integration with digital tools used by ASHAs and sub-centres. In addition, qualitative work with CHWs, supervisors, and households could help optimise the usability, workload implications, and acceptability of the stratification tool before wider scaleup.

Conclusion

This exploratory study successfully developed and implemented a simple, colour-coded household stratification tool and demonstrated its feasibility for identifying households according to healthcare need in rural South India. The stratification classified 18.1% of households as high priority (red), 45.8% as medium priority (yellow), and 36.1% as low priority (green). Family size and socioeconomic status emerged as significant predictors of priority status, while the predominant health problems identified were recurrent ARI and incomplete immunisation in young children, high-risk pregnancy in women, and uncontrolled hypertension and diabetes in older adults, providing a concrete basis for age-specific and household-targeted interventions. Critically, current health worker visits were not systematically targeted to households with greater health burden, representing an opportunity to improve efficiency and equity of service delivery through the proposed stratified visiting schedule: monthly for high priority, bi-monthly for medium priority, and semi-annually for low priority households.

Non-author contribution

The authors thank the CHWs and staff of the Community Health Training Centre, Mugalur, for their support in data collection and community mobilisation.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.