No. It describes cases identified under a specific data system, code rule, coverage and time window. A population estimate requires a defined target population, denominator and ascertainment method, with limitations explained.
Source basis: 1. The Lancet Regional Health – Europe · 2. Autism Research · 3. Health Promotion and Chronic Disease Prevention in Canada · 4. PLOS Medicine
Name the population at each step
RECORD distinguishes the source population a study aims to describe, the database population with records in a data system, and the study population selected by researchers using codes or algorithms. Ask how people move from one group to the next. A clinic record, school classification or insurance claim is created for a purpose; people outside that system may not appear in it.
Inspect the case-finding rule
Find the diagnostic codes, number and timing of records, age limits, exclusions and any linkage rules used to identify cases. Check whether the algorithm was validated against a suitable reference standard and whether that validation matches the population and setting in the paper. A method tested in one jurisdiction or age group does not automatically transfer to another.
Read what the source can capture
Routine records may reflect service access, coding practices, coverage rules and documentation, as well as the underlying condition. A study using English primary-care records, for example, compared recorded diagnoses with community prevalence estimates to examine underdiagnosis; its result is tied to its data, comparison method, population and time period. Treat it as a study-specific analysis, not a universal prevalence figure.
Check the denominator and time window
Look for who could enter the database, which years are covered, how eligibility was defined and whether the denominator changes over time. If records from several systems are linked, check the proportion linked at each step and who could not be matched. Missing linkage or incomplete coverage can alter both the numerator and denominator.
Use RECORD as a reporting aid
The RECORD statement supplements STROBE for observational studies using routinely collected health data. It prompts authors to describe data types, code-based selection, validation and linkage. It helps readers find relevant methodological details; it is not a design prescription, validation result or automatic quality score.
| Focus | What to document | Interpretation question or limit |
|---|---|---|
| Reading check | What to locate | Interpretive limit |
| Population | Source, database and study populations | The study population may be narrower than the intended source population |
| Identification rule | Codes, algorithm, dates and validation | A code-defined count depends on the tested rule and context |
| Coverage | Eligibility, access, geography and period | People not captured in the system may be absent from the count |
| Linkage | Data sources, match rates and unlinked records | Unmatched records can change who is represented |
| Claim | Recorded diagnosis, service count or prevalence estimate | These quantities are not interchangeable |
Limitations and unresolved questions
Administrative algorithms and data coverage vary across jurisdictions, age groups and source systems. Autism-specific validation studies report setting-dependent performance and heterogeneous methods. RECORD is reporting guidance; its use does not by itself establish that a dataset is complete, representative or valid for every intended question.
Implications for research and institutions
When comparing administrative autism indicators, record the data system, population covered, identification algorithm, validation evidence, denominator, years and linkage limitations. Label the result as recorded diagnoses or service-system identification unless the study’s design supports a broader population estimate.
Research question
Does an autism diagnosis count in administrative records equal population prevalence?
No. It describes cases identified under a specific data system, code rule, coverage and time window. A population estimate requires a defined target population, denominator and ascertainment method, with limitations explained.
Verified source bibliography
- Elizabeth O’Nions; Irene Petersen; Joshua E J Buckman et al. (2023). Autism in England: assessing underdiagnosis in a population-based cohort study of prospectively collected primary care data. The Lancet Regional Health – Europe · DOI
- Jennifer D Brooks; Jasleen Arneja; Longdi Fu et al. (2021). Assessing the validity of administrative health data for the identification of children and youth with autism spectrum disorder in Ontario. Autism Research · DOI
- Siobhan O’Donnell; Sarah Palmeter; Meghan Laverty et al. (2022). Accuracy of administrative database algorithms for autism spectrum disorder, attention-deficit/hyperactivity disorder and fetal alcohol spectrum disorder case ascertainment: a systematic review. Health Promotion and Chronic Disease Prevention in Canada · DOI
- Eric I Benchimol; Liam Smeeth; Astrid Guttmann et al. (2015). The REporting of studies Conducted using Observational Routinely-collected health Data (RECORD) Statement. PLOS Medicine · DOI
Methodological guidance is distinguished from empirical autism research. Links and relevant source content were checked on 2026-10-28. Citation is not endorsement or affiliation. Follow the source for its full scope, rights and limitations.