When the Street Becomes Home: Understanding Street-Connected Children in Odisha's Urban Centres

They are called "children in street situations" — a term the UN Committee on the Rights of the Child uses deliberately, because it describes a relationship with the street rather than a permanent identity. Some sleep on the street. Some work on it during the day and return to families in slums at night. Some have families on the street with them. And a smaller but significant number have effectively no family contact at all. What they share is that the street has become their primary social and economic environment — and that nearly every welfare system India has built assumes they live somewhere else.

Odisha's urban centres — Bhubaneswar, Cuttack, Rourkela, Sambalpur, Berhampur — have been growing fast. The state's urbanisation rate crossed 17 percent by the 2011 census and is estimated to have reached 20–22 percent by 2024, driven by industrial investment, port expansion, and the pull of state capital development under successive SMART City and AMRUT initiatives. Migration from the state's 30 districts — particularly from Bolangir, Kalahandi, Nuapada, Gajapati, and Koraput — feeds this urbanisation in waves. Children arrive with parents, follow parents who left, or arrive alone. The street absorbs all of them, and the welfare system barely notices.

This Convergence Note is for practitioners, district officials, and CSR managers who want to understand who these children actually are, what the existing scheme architecture looks like (and where it fails), and what an NGO working in this space actually needs to do.

Who Are Street-Connected Children in Odisha?

The government's most recent systematic count, conducted through Mission Vatsalya outreach data compiled by the Ministry of Women and Child Development as of February 2023, identified approximately 19,546 street children across India through Childline and shelter contact. Researchers and NGOs consistently argue this is a significant undercount — the methodology only captures children already in contact with services.

For Odisha specifically, a 2022 public expenditure study on Mission Vatsalya by the National Institute of Public Finance and Policy found that only 2 percent of the 93,865 identified vulnerable children in the state were institutionalised, which itself underscores the scale of the gap between identification and response. UNICEF's multi-city study on children in street situations, which included Bhubaneswar as a focal city, found that the city's response mechanisms — police referral, Childline activation, CCI placement — were functioning but dealing with a fraction of the actual population.

Three categories matter for programme design:

Children living on the street with their families. These are the most numerically significant group. Their parents are typically informal economy workers — daily wage labourers at construction sites, rag pickers, seasonal vegetable vendors, or workers at brick kilns on the urban fringe. The children are not abandoned; they are with adults who are themselves in precarious situations. They may attend an AWC sporadically, may have birth certificates, and may even be enrolled in school on paper. But actual school attendance is irregular because the family moves with labour demand, because the child contributes to household income, or because the journey to school from a pavement or under a bridge is practically impossible.

Children working on the street but sleeping elsewhere. These children often live in nearby slums with families. They are rag pickers, platform vendors, workers at roadside dhabas, domestic workers in nearby apartments, or engaged in seasonal trade. They may attend school in the morning and work in the afternoon. The Child Labour (Prohibition and Regulation) Amendment Act 2016 prohibits employment of children under 14 in all occupations and under 18 in hazardous ones, but enforcement in the informal urban economy is minimal. These children are statistically invisible because they are not on the street at night.

Children without family contact. The smallest group in absolute numbers but the most complex in terms of need. These children — some as young as eight — have either run away from home, been abandoned, become separated from families during migration or natural disasters, or exited the care of relatives with whom they were placed. They are more likely to be male, more likely to be in the 12–17 age bracket, and disproportionately likely to be from SC/ST communities. They are also the group most at risk of trafficking, substance abuse, and conflict with law. In the Bhubaneswar Railway Station catchment area, Childline India Foundation and partner NGOs have repeatedly documented that the 1098 helpline receives its highest call density from platform areas, temple complexes, and bus stands.

The Odisha Urban Context: Why It Matters Now

Bhubaneswar's population crossed 10 lakh by the 2011 census and is estimated at 15–17 lakh today. The Smart City Mission investments — new infrastructure, heritage zones, convention centres — have in some cases relocated informal settlements without adequate rehabilitation, increasing the street-living population. Cuttack, older and more densely packed, has a concentration of street-connected children near the old Barabazar market complex and the Mahanadi river ghats.

Rourkela's street-connected children have a different demographic profile from the coastal cities. The steel economy creates a particular kind of secondary informal employment — scrap dealing, waste sorting, ancillary vending — that pulls child labour. Sambalpur, at the gateway to western Odisha, receives children from migration-prone Bolangir and Bargarh districts, sometimes as part of family units, sometimes as children placed with distant relatives who themselves have no stability.

Berhampur (Brahmapur) in Ganjam, while not a Smart City, functions as a major labour marshalling point for migrants to Surat and other textile manufacturing cities. Children of migrant families who did not travel with parents end up with grandparents or neighbours whose capacity to care for them is limited. These children — technically "at home" — often end up on the streets of Berhampur when that care arrangement breaks down.

What connects all these urban contexts is the absence of a robust identification mechanism that catches children before they reach the crisis threshold at which Childline activation or CWC referral becomes necessary.

The Scheme Architecture: What Exists, and the Gap Between Existence and Reach

Mission Vatsalya (formerly ICPS)

The flagship national scheme for child protection, Mission Vatsalya (launched 2021 as successor to the Integrated Child Protection Scheme of 2009) is the primary framework for street child response in Odisha. It operates through the Odisha State Child Protection Society (OSCPS), with District Child Protection Units (DCPUs) in all 30 districts as the implementing arm.

Key components relevant to street-connected children:

• Open Shelters: Residential facilities for street children that allow flexible entry/exit without formal institutionalisation. These are specifically designed for street situations — a child can access food, health services, and education without CWC admission.
• Childline 1098: The 24-hour emergency outreach helpline, integrated into Mission Vatsalya, is the primary mechanism for identifying and responding to children in distress on the street.
• Sponsorship and Foster Care: For children with families but at risk — a monthly support payment to prevent family breakdown-driven street entry.
• Child Care Institutions (CCIs): Observation homes and children's homes for children who cannot be in family or community settings.

The NIPFP working paper on Mission Vatsalya in Odisha (2024) identified significant structural weaknesses: a 30 percent shortfall in Protection Officers and a 39 percent shortfall in counsellors across district-level DCPUs. In practical terms, this means DCPUs are understaffed relative to mandate, and the open shelter model — which requires consistent outreach to function — is often reduced to a passive waiting facility.

Mission Vatsalya's allocation in Odisha was the highest among the 22 child protection schemes tracked in the state's Child Budget for FY 2023–24, reaching approximately Rs 1,472 crore in GoI allocation for 2024–25. Budget is not the problem. Absorptive capacity, staffing, and NGO partnership are.

The Juvenile Justice Act 2015 and the CWC

The JJ Act 2015 provides the statutory framework for responding to both children in conflict with law and children in need of care and protection. Street children, by definition, qualify as CNCP under Section 2(14) if they are found without a home and without means of subsistence. The Child Welfare Committee in each district has the authority to admit, transfer, and order care arrangements for such children.

The CWC pathway works for children who come to the attention of the system — police pickup, Childline contact, NGO referral. It does not work for children who exist below the visibility threshold of formal outreach.

The UNICEF study on children in street situations flagged a critical system gap specifically relevant to Bhubaneswar: the absence of standard operating procedures for police personnel at the thana level in handling CISS encounters. Police are often the first responders, and without SOP training, responses range from appropriate referral to casual dismissal to, in the worst cases, treating the child as a suspect rather than a person in need of protection.

PM CARE for Children

Launched in 2021 to support children orphaned by COVID-19, this scheme provides monthly financial support, health insurance under Ayushman Bharat, education support, and a lump-sum at age 23. While not designed specifically for street children, orphaned street-connected children or children who lost parents during the pandemic and subsequently ended up in street situations are eligible. District Collectors are the designated authority for identification and approval.

POSHAN Abhiyaan and ICDS

Anganwadi Centres are technically accessible to all children under six, including those in street situations. In practice, AWC access requires address registration, which street-connected families cannot provide. Some Odisha districts have experimented with mobile AWC services, but this is not standardised.

Scheme Convergence: What an NGO Needs to Navigate

An NGO working with street-connected children in an Odisha urban centre needs to operate across four distinct systems simultaneously — and the coordination between these systems is weak.

System 1: Mission Vatsalya / DCPU. The NGO's relationship with the DCPU is the foundation. DCPU can authorise open shelter operations, refer children to CCIs when needed, provide sponsorship assessments, and connect families with MGNREGS or PMAY-G for stabilisation. The NGO should have an MoU with the DCPU and clear referral protocols.

System 2: Childline. If the NGO is not itself a Childline partner, it needs a functional working relationship with the Childline partner organisation in its district. All night-time emergency responses go through Childline 1098.

System 3: CWC. For children who need formal care orders, the CWC pathway is mandatory. NGO caseworkers need to understand how to prepare Social Investigation Reports (SIRs) and Individual Care Plans (ICPs), which the NIPFP study found were frequently inadequate in Odisha CCIs due to insufficient consultation with children.

System 4: Education system. The National Child Labour Project (NCLP) schools, bridge course centres, and RTE Section 12 seats are the pathways back into education. No-document enrollment policies exist on paper in Odisha; implementation varies by school and by the persistence of the NGO caseworker.

The convergence problem is a coordination problem. A child who arrives at an open shelter at night has health needs (Ayushman Bharat, NHM referral), a potential CWC case, a family stabilisation need (MGNREGS or NFBS), and an education need (bridge course or NCLP) — all managed by different departments that do not share information or have joint review mechanisms.

Why 70 Percent Go Back to the Street

The most damning statistic in this space comes from a 2021 Delhi study, but it reflects a structural reality that Odisha shares: approximately 70 percent of rescued street children return to the street within six months of placement in a shelter or CCI. The reasons are not primarily about the quality of institutional care, though that matters. They are systemic:

Families are not stabilised. If a child's parent is a daily wage labourer sleeping on a construction site, placing the child in a CCI does not solve the family's housing or income problem. When the CCI's maximum stay period approaches or when the child runs, the family situation is exactly as it was.

Institutional care is not designed for non-linear reintegration. The JJ Act envisages a CCI as a temporary measure pending reintegration or adoption. But reintegration requires a home to return to, a school that will accept the child without documents, and a neighbourhood where the child is not immediately drawn back into street survival routines. None of these are automatic.

The street offers peer networks and autonomy. Particularly for older adolescents (14–17) who have been on the street for extended periods, the peer group on the street is their community. The CCI or open shelter can feel like prison by comparison, regardless of quality of care. A 15-year-old who has been sorting scrap in Cuttack's Bidanasi market for two years knows that environment, has relationships there, and has income. What does reintegration offer that is comparable?

Substance dependence is underaddressed. Solvent sniffing (whitener, correction fluid, glue) is documented among street-connected adolescent boys across Indian cities, including Bhubaneswar. It is a coping mechanism for cold, hunger, and trauma, and it creates dependency that standard child protection responses are not equipped to address.

What NGOs in This Space Actually Do (and Should Do Differently)

The most effective NGOs working with street-connected children in Indian cities have learned that the traditional rescue-and-place model is inadequate. A more effective model has five elements:

1. Presence before crisis. Outreach workers who are physically present in the children's environments — platforms, markets, ghats, construction sites — at the times and in the patterns that children are present. Not visiting. Present. This requires a different staffing model from AWC-style or office-based programming.

2. Demand-led services. Children engage with services that meet their felt needs, not services that meet programme design assumptions. Hot food at night, a safe place to sleep, access to a charging point, a trusted adult who is not going to report them to police — these are the access points. Education and documentation come later, when trust is established.

3. Family tracing and stabilisation, not just child placement. For children with identifiable families, the parallel workstream is stabilisation of the family — ration card, job card, PMAY application, ESIC if the parent is in construction. This requires NGO staff who can navigate welfare portals and who have relationships with block-level officials.

4. Street-based education, not just school enrollment. For children who cannot attend regular school, NCLP bridge courses and open schooling (NIOS) are available options. The key is that the educational pathway meets the child where they are, not the other way around.

5. Documented case management. Every child should have a named caseworker, a documented situation, a care plan, and a review schedule. This sounds basic; in practice, most open shelters in India function as drop-in centres with no individual case tracking. The SIR and ICP documentation the JJ Act mandates is the right framework — the problem is it is treated as a compliance document rather than a living tool.

For CSR Managers: What Funding in This Space Should Look Like

Street-connected children programmes suffer from a specific CSR funding problem: the metrics don't look clean. You cannot easily count "lives transformed" in the way a school-enrollment programme can count "children enrolled." A child who came to an open shelter 40 times over eight months, went back to the street, came back again, got her ration card processed, enrolled in a NIOS course, dropped out, came back, and is now working at a regulated market stall — has that life been improved? Yes. Measurably? Not in a way that looks good on an annual CSR report.

This means CSR investment in this space requires funders who are willing to track process indicators (number of outreach contacts, number of children accessing meals, number of DCPU referrals, number of documents processed, number of families stabilised) rather than outcome indicators in the short term, and who understand that impact in child protection accumulates over years, not quarters.

The most fundable models are those that sit between purely reactive (rescue-and-CCI) and purely preventive (unverifiable community awareness). Open shelter operations with documented outreach, DCPU partnerships that demonstrate institutional linkage, and education-via-NCLP or NIOS programmes that show school progression — these are defensible to CSR compliance teams while still being honest about the complexity of the problem.

Odisha's DCPU structure, despite its staffing gaps, is an underused partnership opportunity for CSR-funded NGOs. An NGO that has a formal MoU with the DCPU, contributes to its case management capacity (even by sharing a trained social worker), and co-ordinates referrals through the formal child protection system is both more effective and more credible than one operating in parallel to the state system.

Districts Where Need Is Concentrated

While street-connected children exist in all of Odisha's urban centres, NGO capacity is most acutely needed in:

• Bhubaneswar (Khurda district): Capital city dynamics, Smart City displacement of informal settlements, Railway Station platform concentration, temple city migrant pull.
• Cuttack (Cuttack district): Oldest city in Odisha, dense market areas, river ghat population, historic informal economy.
• Rourkela (Sundargarh district): Steel-economy scrap trade child labour, tribal community migration from adjoining districts.
• Berhampur (Ganjam district): Migration marshalling point, high Ganjam outmigration creating family care gaps.
• Sambalpur (Sambalpur district): Western Odisha migration destination, gateway to eastern Odisha for Bolangir/Kalahandi families.

Schemes at a Glance

What This Convergence Note Is For

This note is written for three audiences. For NGO practitioners, it is a map of the terrain — the statutory framework, the scheme architecture, and the practical realities of what effective programmes in this space require. For CSR managers, it is an honest account of why this work is hard to fund and why it is worth funding anyway. And for district officials — DCPUs, CWC members, municipal social welfare officers — it is a reminder that the children on the platforms and under the bridges of Odisha's cities are not an administrative inconvenience. They are children. The system was built for them. The work is to make it actually reach them.

Sources: UNICEF India, Study of Children in Street Situations in India (multi-city); NIPFP Working Paper 418, Mission Vatsalya Public Expenditure and Institutional Analysis (2024); Odisha State Child Protection Society Wikipedia entry; Prarambha — Odisha State Policy for Children (2022); Ministry of Women and Child Development, Childline 1098 Annual Reports; JJ Act 2015 and 2021 Amendment; NCPCR SOP for Children in Street Situations (2020).

This article is part of the JaBaSu Convergence Notes series — practitioner-depth analysis of how multiple schemes, departments, and sectors intersect around specific communities in Odisha. For scheme primers, compliance guides, and sector spotlights, visit jabasu.org/knowledge-commons.