It can support a description of within-person change when the same participant is measured repeatedly and the analysis preserves that structure. A change in a group average can also reflect who remained, who was added, missing data or changes in measurement; the report should make these factors visible.
Source basis: 1. BMJ Open · 2. Journal of Clinical and Translational Science · 3. PLoS medicine
Check that the same people are followed
A study comparing children, adults and older adults at one point in time is cross-sectional, even when it covers a wide age range. A longitudinal analysis follows participants across stated waves. Some projects also add new participants or use accelerated designs; reports should explain which observations belong to which people and periods.
Read the follow-up schedule and measures
Look for the time between waves, the period covered, and which outcomes or experiences were measured at each wave. Repeating a questionnaire every two years is a series of observations at those intervals, not continuous monitoring. If the instrument, reporter or question wording changes, ask how the study handled comparability.
Track who remained in each analysis
Participant flow can change between enrollment, each follow-up and analysis. Check how many people were invited, responded, were retained, contributed usable data and were included in each result. Reasons for loss to follow-up and missing data matter because the people still responding may differ from those who left. A change in the analyzed group can complicate interpretation of an apparent change over time.
Separate a cohort profile from a population estimate
Autism cohorts are built for particular questions and may have specific age ranges, locations, recruitment pathways and measures. For example, the Australian Longitudinal Study of Adults with Autism describes questionnaire-based follow-up of autistic adults aged 25 and over at two-year intervals. SPARK is a US online research cohort that has also used clinical-site recruitment. These examples show different cohort structures; neither should be treated as a universal template or as automatically representative of all autistic people.
Use reporting guidance as a reading aid
STROBE asks observational-study reports to describe participant selection, follow-up methods, participant counts at study stages, missing data and follow-up time. It is reporting guidance for cohort, case-control and cross-sectional studies—not a design prescription, quality score or substitute for judging bias and applicability.
| Focus | What to document | Interpretation question or limit |
|---|---|---|
| Reading check | What to locate | Interpretive caution |
| Time structure | Same participants, waves and dates | Broad age coverage alone is not longitudinal follow-up |
| Cohort scope | Eligibility, setting and recruitment routes | A cohort description is not a population census |
| Retention | Numbers responding and analyzed at each wave | Attrition may change who contributes later data |
| Measurement | Outcomes, reporter and instrument at each wave | Changed measures may limit direct comparison |
| Conclusion | Within-person pattern, comparison and assumptions | Change over time alone does not establish its cause |
Limitations and unresolved questions
The ALSAA and SPARK descriptions concern specific Australian adult and US online research cohorts. Their design details cannot be generalized to all autism research. Follow-up loss, changing measures, cohort composition and analytic choices affect interpretation; reporting checklists help readers find methods but do not prove that bias is absent.
Implications for research and institutions
When summarizing a longitudinal result, record the cohort definition, observation period, wave schedule, number retained at each wave, measure and reporter, and the analysis population. State whether the claim concerns within-person change, group-level pattern or a comparison between groups, and avoid implying causation from timing alone.
Research question
Does a longitudinal autism study show that an individual changed over time?
It can support a description of within-person change when the same participant is measured repeatedly and the analysis preserves that structure. A change in a group average can also reflect who remained, who was added, missing data or changes in measurement; the report should make these factors visible.
Verified source bibliography
- Samuel R C Arnold; Kitty-Rose Foley; Ye In Jane Hwang et al. (2019). Cohort profile: the Australian Longitudinal Study of Adults with Autism (ALSAA). BMJ Open · DOI
- Amy M Daniels; J Kiely Law; LeeAnne Green Snyder et al. (2023). Effectiveness of multimodal participant recruitment in SPARK, a large, online longitudinal research study of autism. Journal of Clinical and Translational Science · DOI
- von Elm E; Altman DG; Egger M et al. (2007). The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies.. 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.