England’s children’s public health nursing data rest on mixed local returns and weighted national estimates
England’s 2025 children’s public health nursing methodology lays out how official statistics were assembled for the health visiting workforce, the school nursing workforce, and service activity tied to the healthy child programme. The publication draws on returns from 145 local councils, while six councils did not submit data. For policymakers, service planners, and clinical leaders, the structure matters because these figures are not simple headcounts pulled from a single national register. They are constructed from local submissions collected over different reporting windows, then standardized through reproducible processing and national scaling.
The result is a statistical picture built from operational reality rather than from a single uniform dataset. In workforce surveillance, that distinction affects how much confidence can be placed in regional comparisons, staffing benchmarks, and commissioning decisions.
Workforce counts were converted to whole-time equivalent posts
The methodology uses total established whole-time equivalent WTE figures for both health visiting and school nursing. WTE converts full-time and part-time hours into the equivalent number of full-time posts, which allows local councils with different staffing mixes to be compared on a like-for-like basis. The publication groups roles by professional registration or function, and those categories were tested with local authorities to confirm that they were understandable and aligned with service structures.
In health visiting, the published categories include specialist community public health nurse, other registered, other clinical, non-registered, other support, management, administration, and other management and administration. School nursing uses a parallel structure, but the job roles within each category differ because the two services are organized differently. Those differences are not cosmetic. They shape how workforce pressure appears on paper and how easily a service can be scaled when demand increases.
A team leader with caseload, for example, is counted differently from a team leader without caseload. So is a school nurse with specialist community public health nurse registration versus a non-SCPHN school nurse. The grouping scheme preserves the clinical meaning of the roles while still allowing the data to be summarized at council, regional, and England level.
Remote contacts and review counts can distort delivery settings
Service activity covers statutory and non-statutory healthy child programme reviews, but only statutory review statistics are presented in the commentary. Not every area delivers every type of review, so the raw data for other reviews remain in the published tables rather than in the narrative summary. That choice avoids comparing areas on measures that are not uniformly offered.
The handling of delivery settings is more revealing than it first appears. Percentages for settings may not sum neatly to 100% because councils may overcount reviews, especially remote contacts. In practical terms, a child can be logged in more than one component if the local reporting system captures the encounter in overlapping ways. The methodology resolves this by using each council’s reported total when totals and subcomponents do not match. That preserves internal consistency, but it also signals that service activity data carry a reporting artefact that readers should not confuse with clinical duplication.
Remote reviews have become a meaningful feature of service delivery, especially where access, geography, or workforce constraints shape contact patterns. A data model that cannot accommodate overlapping reporting would exaggerate the precision of the underlying service counts.
National estimates were scaled for missing council returns
Six local councils did not submit workforce data, so national workforce totals were not simply summed from participating areas. Instead, the publication applied a population-based correction factor. England’s population aged 0 to 5 was 3,072,243, while the population covered by responding councils was 2,956,757, using the most up-to-date Office for National Statistics population data for 2025 accessed through the Nomis platform. Dividing those figures produced a correction factor of 1.0391.
Observed totals from responding councils were multiplied by that factor to estimate the national workforce for health visitors and school nurses. This is a defensible statistical solution when the relationship between workforce numbers and the under-5 population is assumed to be similar in responding and non-responding councils. It is also an assumption worth keeping in view. If the missing councils differ systematically in staffing model, service pressure, or commissioning arrangement, the national estimate will still be useful, but it will inevitably be an approximation rather than a census.
The report does not present that limitation as a flaw; it presents it as a transparent feature of official estimation. Transparency is the clinical virtue here. Public health nursing data are only as credible as the methods used to handle incomplete returns.
Aggregation rules keep larger councils from being drowned out
Percentages at local, regional, and national level are calculated from aggregated counts rather than by averaging local percentages. Numerators and denominators are summed separately before the percentage is calculated. This approach ensures that larger councils contribute proportionately to regional and national estimates and avoids giving equal weight to small and large areas simply because they are both councils.
The same rule applies across England and the Office for National Statistics 1TL regions. Regional totals are generated by summing local council WTE values and recalculating percentages from the aggregated counts. The published values are then rounded to the nearest whole number.
This method sounds technical, but the bedside consequence is real. If workforce density or review uptake is miscalculated, planning for child health surveillance, early identification of developmental concerns, or referral pathways can be distorted long before any deficit appears in frontline care. Data architecture influences service visibility.
A reproducible R workflow underpins the official tables
Data extraction was done directly from submitted returns using a reproducible process written in R. The workflow checked for completeness, expected structure, and duplicate records, and it standardised local council identifiers through string cleaning. Blank, non-numeric, or invalid entries were treated as missing values and excluded from calculations where appropriate.
That level of documentation is important because the publication sits within official statistics in development. Readers need to know not only what was counted, but how the counts were cleaned, aggregated, and converted into publication tables. The reproducible pipeline means the outputs can be regenerated consistently from the same source returns, applying the same rules each time. In a system spanning 145 councils and multiple staffing structures, reproducibility is not administrative polish. It is what keeps the dataset audit-ready.
These figures describe service capacity, not just staffing numbers
The publication’s value lies in how it links workforce structure to service delivery. Health visiting and school nursing are preventive services, and the workforce behind them shapes whether families receive timely reviews, whether statutory contacts are completed, and how consistently local programmes are delivered. A staffing dataset that distinguishes registered from non-registered roles, or management from clinical posts, can reveal where service capacity is concentrated and where it may be thin.
Because the publication aggregates local returns and uses a correction factor for missing councils, its estimates should be read as national planning figures rather than absolute clinical census counts. The methodology is careful, explicit, and operationally grounded. What it cannot do is remove variation in local reporting practice, missing submissions, or the structural differences between councils that commission from NHS providers, charities, private providers, or deliver services themselves.
Those are the constraints that shape every number in the report. They also define the next question for workforce intelligence: how to make child public health data both comparable and sensitive enough to reflect the services children actually receive.