Skip to content

Healthcare Access and Deprivation in Lokoja LGA

Which wards are cut off from care, and which are crowded out of it?

Lokoja Local Government Area covers 3,187 km², holds 308,138 people and has 112 active clinics. I mapped where those clinics are, who lives near them and how far everyone else must travel. The work used clinic locations from GRID3, 100-metre population rasters from WorldPop and road data from OpenStreetMap, all processed in ArcGIS Pro. Each of the 10 wards was measured on its own.

The study has four parts. Each can be read alone, but they build on one another.

Part 1: Where the clinics are, and who they serve

National health audits count how many people live within reach of a clinic. They average over large areas, so they hide differences between wards, and they assume that a clinic nearby means care received.

At ward level the gap is plain. Lokoja A, in the administrative core, has 40 clinics in 32.1 km², or 1.25 per km². Four rural wards (Oworo, Kakanda, Kupa North and Kupa South) cover 97.04% of the land and have between 0.010 and 0.021 clinics per km².

Counting clinics is only half the picture, so I then divided each ward's population by its clinics. Lokoja C, a peri-urban ward of 0.7 km², has one clinic for 6,576 residents. That is the heaviest load in the LGA, about 2.4 times the average of 2,751. A distance-only audit would call the ward fully served, since nobody lives far from the clinic. The ratio points to a clinic under strain, although I inferred the strain and did not observe waiting times. I call this the Urban Capacity Paradox, and I explain it, and its limits, in an essay.

Part 2: Who lives too far away

The usual way to ask who is within reach is to draw a 5 km circle around each clinic, the distance the World Health Organization uses as a benchmark. A circle assumes people can walk in a straight line over open ground. In a ward crossed by rivers and hills, they cannot. I therefore measured distance in two ways, as a straight line and along the road network.

With 5 km circles, 38,423 residents, or 12.47% of the LGA, live out of reach. Most are in two wards. In Kakanda, 44.7% of residents are unserved and the ward's centre lies 8.5 km from the nearest clinic. Oworo has the largest number unserved, at 18,398. Straight lines understate real journeys, so the true figure is higher.

The road network gave an unexpected result. Routing to the nearest clinic failed in five wards (ArcGIS warning 030025). In Lokoja A, D and E the cause was the data: clinic points sat on footpaths that the road layer does not include. In Kakanda and Kupa South the roads themselves are cut or missing. The first kind of failure is a fault in the map. The second is a fault on the ground, and it is the one that matters to patients.

Part 3: Ranking the wards

Distance and crowding each tell part of the story, so I combined them with a third measure, the scarcity of clinics, into one score: the Geographic Access Deprivation Index (GADI). I rescaled each measure from 0 (best) to 1 (worst), averaged the three with equal weight and sorted the wards into five tiers at the natural breaks in the scores. Kakanda ranks first, at 0.884. Lokoja B ranks last, at 0.042.

Would a different weighting change the order? I tested two. In the first, distance counted for 50% and the other two measures for 25% each. In the second, crowding took the 50%. Kakanda stayed first, Lokoja A stayed ninth and Lokoja B stayed tenth under both. The middle of the table moved. Lokoja C fell from second to fourth place when distance counted more, and returned to second when crowding did. The weighting is therefore a policy choice, and whoever uses the index should say which one they chose.

The score also shows why a ward ranks where it does. Kakanda is the Rural Physical Archetype: it scores 1.00 on both distance and scarcity. Lokoja C is the Urban Congestion Archetype: it scores 1.00 on crowding but only 0.04 on distance. Both rank high for opposite reasons, so one remedy for both would send money to the wrong place.

Part 4: From ranking to action

Lokoja sits where the Niger and Benue meet, and floods are a regular risk. In October 2022, flooding coincided with more than 6,000 suspected cholera cases in the wider region. I treated risk (R) as a function of hazard (H) and vulnerability (V), written R = f(H, V). Flooding is the hazard, and a ward's GADI score stands for its vulnerability. The reasoning is that a ward that is hard to reach in a normal week will be harder to reach in a flood. This is an inference. The cholera figures were not reported by ward, so they cannot show that deprived wards suffered more.

To turn the ranking into a plan, I grouped wards by the kind of problem they have. This gave three tracks for the Lokoja Local Emergency Management Committee (LEMC), set within Priority 3 of the Sendai Framework, which calls for investment in resilience, including the strength of health systems.

  1. Rural Access Track (Kakanda, Oworo, Kupa North, Kupa South and Eggan). Distance and missing roads are the main problem. The actions are boat ambulances on the riverine routes, new or rebuilt health posts on floodplains above the 2022 flood peak, and grading and drainage on the dirt roads to the nearest post.
  2. Urban Capacity Track (Lokoja C, D and E). Clinics are near but busy, or reached by poor paths. The actions are to upgrade the Lokoja C clinic to a 24-hour comprehensive health centre, add nurses and community health extension workers in Lokoja C and E, and clear drainage and mend footpaths in Lokoja D and E.
  3. Central Surveillance Track (Lokoja A and B). These wards are relatively well served, so they need coordination more than new clinics. The actions are to stock cholera treatment supplies, water purification tablets and oral rehydration salts before each flood season, and to agree a plan with the main hospital for surge referrals from rural and riverine wards.

The index shows where to act, not what to build. Engineering surveys and talks with communities should come before any construction. Each track also has one measure of progress: the share of residents beyond 5 km (rural), residents per clinic in Lokoja C (urban), and a written referral plan with stocked supply points (central).

Data and limits

The data are public and span 2024 to 2026, so they show one moment. Small or unregistered clinics may be missing, and a single missed clinic in Lokoja C would roughly halve its load. The study measured distance and crowding only, with no data on cost, drug stocks or health outcomes. I would check the clinic register and the road layer on the ground before using any ranking to allocate money.

Data: GRID3 Nigeria, WorldPop, OpenStreetMap contributors.