# Birmingham Dental Site Screening: Count the Unit Before You Count the Market

Canonical: https://www.locusintel.io/blog/birmingham-dental-ods-screen

A reproducible Birmingham dental-practice screen shows how radius, ODS status and postcode deduplication change the shortlist before demand or availability is assessed.

Published: 2026-10-26
Author: Locus Team

A location brief can ask for “the number of nearby dental practices” as if it were one straightforward measure. It is not. The answer changes when the radius changes, when inactive registry records are removed, and when multiple records sharing a postcode are treated as one site.

We tested that problem using the NHS England Organisation Data Service (ODS) search for **Birmingham** and the primary role **General Dental Practice (RO110)**. The 24 October 2026 table returned **196 records**: 168 active and 28 inactive. After postcode geocoding, 194 records could be placed around a fixed central postcode, B1 2JT.

At 8 km, the screen contained **145 ODS records, 125 active records, 134 unique postcodes and 118 active unique postcodes**. Those are all defensible counts, but they answer different questions. None is a measure of demand, capacity, or whether a practice is accepting new patients.

## The buyer question comes before the count

For a dental group or operator considering a new site, the useful first decision is not “which area wins?” It is: **what should this screen be allowed to tell us, and what must we verify next?**

The ODS Data Search and Export service provides organisation details and CSV/Excel exports. Its records are valuable for building a dated, inspectable registry screen. The ODS catalogue also says the dental-practice role includes NHS and private practices, but private codes are issued only on request. It warns that V codes can be location codes, more than one practice can share a location code, and duplicate codes can occur. That makes this a registry screen, not a complete private-competition census.

The NHS [Find a dentist service](https://www.nhs.uk/Service-Search/find-a-dentist) answers a different, patient-facing question: it searches for NHS dentists accepting new patients or offering urgent or specialist care. NHS England’s [dentistry data](https://digital.nhs.uk/data-and-information/data-tools-and-services/data-services/general-practice-data-hub/dentistry) covers NHS activity and patients seen, while explicitly excluding private services and hospital dental services.

That division matters. A registry record is a useful candidate entity. It is not proof of a current opening, patient acceptance, NHS contract, clinical capacity, or competitive pressure.

## What we tested

The protocol fixed the sample before the distance calculations. We retained every row in the Birmingham ODS export, including both active and inactive statuses. ODS `Code` remained the record unit. `Postcode` supplied the location field; identical postcodes were retained as separate records and also counted as unique postcode sites.

We used B1 2JT as a reproducible centre point, geocoded the centre separately, and geocoded each ODS postcode with Postcodes.io. The analysis then applied haversine distance bands of 1, 2, 3, 5 and 8 km. These are comparison radii, not a Locus catchment or travel-time estimate.

Two records could not be geocoded: both were inactive. We did not impute them or silently drop them from the source total. The retained distance table contains 194 geocoded rows and records the coverage gap.

## Radius changes the shortlist faster than a label suggests

![Active ODS general-dental-practice records within fixed radii of B1 2JT: 7 at 1 km, 19 at 2 km, 33 at 3 km, 69 at 5 km and 125 at 8 km.](https://www.locusintel.io/images/blog/birmingham-dental-ods-screen/birmingham-dental-ods-radius-narrow.webp)

*Source: NHS England ODS Advanced Search, Birmingham, table last updated 24 October 2026; Locus calculations on 26 October 2026. ODS records are not demand, capacity, or accepting-patients evidence.*

| Radius from B1 2JT | All ODS records | Active ODS records | Unique postcodes | Active unique postcodes |
| -----------------: | --------------: | -----------------: | ---------------: | ----------------------: |
|               1 km |               8 |                  7 |                8 |                       7 |
|               2 km |              21 |                 19 |               19 |                      17 |
|               3 km |              38 |                 33 |               35 |                      31 |
|               5 km |              82 |                 69 |               75 |                      65 |
|               8 km |             145 |                125 |              134 |                     118 |

All radius counts above use the **194 geocoded rows**. The two source rows without a returned coordinate remain in the 196-row source total and are excluded from every radius band; neither was imputed.

The operational implication is simple: an 8 km screen produces almost eighteen times as many active records as the 1 km screen. That is not a market conclusion; it is a reminder that “nearby” is a material modelling choice. A buyer comparing two areas can create a ranking simply by allowing one screen to expand farther than the other.

## One radius still has four valid denominators

![At 8 km, the same screen contains 145 ODS records, 125 active records, 134 unique postcodes and 118 active unique postcodes.](https://www.locusintel.io/images/blog/birmingham-dental-ods-screen/birmingham-dental-ods-unit-sensitivity-narrow.webp)

*Source: NHS England ODS Advanced Search and Locus calculations, 26 October 2026. Same-postcode records may represent distinct organisations; postcode deduplication is a site-screening convention, not an identity resolution.*

The 8 km result is the clearest illustration. Removing inactive records changes 145 to 125. Deduplicating postcodes changes the all-record count to 134 and the active count to 118. The differences are not errors to be averaged away. They are signals about the question being asked:

* **ODS records** are appropriate when the question is how many registry entities appear in scope.
* **Active ODS records** are appropriate when the first screen should exclude records marked inactive.
* **Unique postcodes** are a rough site-location view, but one postcode can still contain more than one organisation.
* **Active unique postcodes** are a conservative starting point for a location list, not a verified competitor count.

The source itself should stay visible in the handoff. If the analyst reports only “118 nearby practices,” a later reader cannot tell whether that means active ODS records, postcode sites, or a different dataset altogether.

## What this does—and does not—say about Birmingham

This screen establishes a reproducible registry-based next step: inspect the 31 active unique postcode sites within 3 km, or the 65 within 5 km, using the buyer’s chosen operating radius. It does not identify the best site, estimate unmet demand, measure patient flow, or prove that a practice is a live competitor.

The next validation should join the specific decision to the right evidence. For patient-access questions, check the current NHS patient-facing search and record the date and result. For commercial questions, verify the premises, current opening status, service mix, catchment, access, and any available operating or market evidence. Keep NHS activity and private-practice questions separate; the official NHS dental dashboard does not cover private services.

This is also why a Locus demographic preview should not be blended into this count as if the geographies were identical. A five-mile demographic radius, a 20-minute drive area, a fixed 5 km circle and a postcode registry screen each have different units and coverage. The right comparison is explicit alignment, not a larger-looking number.

## Scope, sources and reproduction

This is a bounded Locus research analysis of public reference data. It is not a Locus suitability score and does not represent NHS endorsement. The ODS standards document describes ODS data as openly available under the UK Open Government Licence; the ODS Data Search and Export service is the current route used for this extract. Postcodes.io supplied coordinates only; its documentation describes the bulk lookup endpoint and the relevant Great Britain postcode data terms.

Attribution and licence note: ODS data are reused under the UK Open Government Licence with NHS England/ODS attribution. The postcode coordinate layer is based on Postcodes.io documentation and Great Britain OS OpenData terms; the associated notices include OS data © Crown copyright and database right and Royal Mail data © Royal Mail copyright and database right. These sources do not endorse Locus or the conclusions here.

The raw CSV, derived distance table, summary JSON, fixed protocol and PowerShell reproduction script are retained with this study. Reproduction requires the ODS Birmingham/general-dental-practice export, the same B1 2JT centre, the five fixed radii, and the stated record/status/postcode definitions. The two un-geocoded inactive rows are part of the limitation and must not be treated as zero-distance or as evidence of absence.

Sources: [NHS England ODS Data Search and Export](https://digital.nhs.uk/services/organisation-data-service/data-search-and-export); [ODS standards and licensing](https://digital.nhs.uk/binaries/content/assets/website-assets/services/ods/odsstandards.pdf); [NHS Find a dentist](https://www.nhs.uk/Service-Search/find-a-dentist); [NHS England dentistry data](https://digital.nhs.uk/data-and-information/data-tools-and-services/data-services/general-practice-data-hub/dentistry); [Postcodes.io bulk lookup documentation](https://postcodes.io/docs/postcode/bulk/) and [licensing](https://postcodes.io/docs/licences/).