FlowsBook a callSign inCreate free accountSign up
Population and demographics

Residents per doctor by area: finding where general practice is thin

How to work out residents per doctor by area: count the people in each district, count the practices, divide, and read the gaps without being misled by the ratio.

Part of the guide Location intelligence automation

By the xMap team · · 4 min read

To find residents per doctor by area, split the region into districts or small cells, count the people living in each, count the general practices in each, and divide the first by the second. Districts with far more residents per practice than the rest are the likely gaps. Then multiply by the doctors each practice really has, from your own records, because a database of places counts practices, not the people working in them.

This is for health authorities, primary care networks, clinic groups looking for under-served areas, recruiters of doctors and researchers. It sets out the steps, a made-up example in Toronto, the ways the ratio misleads, and how an AI employee produces the table every quarter.

What goes on top and what goes underneath

On top: residents. The Demographics block counts the people living in each area you give it, from xMap's population grid, along with age groups. Older and very young residents see a doctor more often, so a plain head count understates need in areas with many of either. Keeping the age groups beside the total lets you weight them if your service does.

Underneath: practices. The Nearby places block counts the places of a type inside each area from xMap's places database: general practices, family doctors, medical clinics. That is a count of sites. A single-doctor surgery and a large group practice each count once.

The steps, in order

  1. Choose the areas: drawn districts, pasted boundaries, or every small hexagon for a fine map.
  2. Count residents in each area, with the age groups.
  3. Count general practices in each area, or bring in your own register.
  4. Divide residents by practices (or doctors) for each area.
  5. Rank the areas and look at the top ones on a map before drawing conclusions.

A made-up example in Toronto

xMap's places database lists 329 medical clinics within 10 km of the centre of Toronto (November 2026). That is a real count; the district figures below are invented to show the method.

Made-up districtResidentsAged 60 and overPractices foundResidents per practice
District A, downtown92,00011,000382,421
District B, inner suburb118,00024,000215,619
District C, outer suburb76,00017,500612,667
District D, waterfront41,0005,800123,417

Invented figures to explain the arithmetic; not data about Toronto.

District C stands out: nearly 12,700 residents per practice and a large older population, against about 2,400 downtown. Downtown's low figure is partly an illusion, since many of its practices serve office workers who live elsewhere. District B is the second gap, and its older residents make it more pressing than the raw ratio suggests.

Districts for the ratio, cells for the picture

Once a district stands out, switch the same area to Every hexagon and put the cells on a map with the practices as points. A thin district often turns out to have one dense pocket, a cluster of towers or a new estate, far from the nearest practice, while the rest of it is adequately served. That pocket, not the district as a whole, is where a new practice or a satellite clinic belongs, and it is the map a planning committee will ask to see.

How the ratio misleads

  • Practices are not doctors. Three practices with one doctor each are thinner than one practice with eight.
  • Borders are open. People cross into the next district for a doctor near work or a station. Look at neighbouring areas together.
  • Small areas swing wildly. A cell with 800 residents and no practice divides by zero; it may be next door to a large clinic. Use cells for the map and districts for the ratio.
  • Residents are counted where they sleep. Business districts look over-served on this measure.
  • Some countries have no resident data. The grid covers about 100 countries, among them Canada, France, India, Thailand, Egypt and Australia, but not the United Kingdom or the United States yet.

A quarterly table from an AI employee

  1. Schedule: the first Monday of each quarter.
  2. Area of interest: the districts, drawn or pasted once.
  3. Demographics: Totals for each area.
  4. Nearby places: general practices and family doctors in each area, counted.
  5. Calculated field: residents per practice, and over-60s per practice.
  6. Sort and limit: the thinnest areas first.
  7. Make a report for the board, with a map.

The board receives the districts ranked from thinnest to best served, each with residents, older residents, practices and both ratios, and a PDF to circulate. The Neighbourhoods far from a hospital, Villages with no health centre nearby and New clinics and hospitals to staff templates take the same idea in other directions. The location intelligence guide shows how population and places data work together.

Questions

Is this the same as the official doctor-to-population ratio?

Not exactly. Official ratios count registered doctors. This counts practices from xMap's places database unless you supply your own register of doctors.

What size of area works best?

Districts or neighbourhoods for the ratio; small hexagons for the map, to see where people cluster inside a thin district.

Can it weight older residents more?

Yes. The age groups come with every area, so a Calculated field can give over-60s more weight.

Does it work in England?

Not yet. Resident counts are not available in the United Kingdom.

Templates that do this

Read next

More on population and demographics · Every article · Templates

Put it to work

An AI employee in xMap Flows does this on live data, on its own timetable. Start from a template or describe it in plain words.