Contents
On this page
- The central problem: a trajectory claim from a snapshot
- What a normative reference curve licenses
- Writing about neurodevelopmental change
- Writing about neurodegenerative change
- Biological factor to brain to behaviour: where the language breaks
- Burden figures, and which direction the story runs
- Where this science is genuinely unsettled
- Reporting the study so it can be appraised
- Take-home message
- How to apply this to your study
- Common questions
Academy · Research Writing · Brain change across the lifespan
Writing about biological factors and brain change across the lifespan
Lifespan claims about the brain are claims about change over time, and most of the data behind them is cross-sectional. That gap is where the writing in this area usually fails: a difference between age groups is reported as a trajectory, and a difference between a diagnosed group and controls is reported as a mechanism. Both are fixable in the wording, and neither is fixable in the analysis.
Neurodevelopmental and neurodegenerative conditions are the standard illustrations for how biology and brain relate across a life, and they are useful precisely because they sit at opposite ends of it. They are also where overclaiming is easiest, because the causal story is intuitive and the evidence is usually correlational. This page is about how to write up such work so the claims match the design: what a normative reference curve licenses, what a group difference does not, and where the science is genuinely unsettled rather than merely summarised briefly.
What you will learn
- Why an age difference in a cross-sectional sample is not a developmental or degenerative trajectory
- What normative lifespan reference curves do and do not license, following Bethlehem and colleagues in Nature
- How to write about a biological correlate without implying it is a cause, a consequence or a biomarker
- The specific confounders that make brain–behaviour group differences hard to interpret, and how to report them
- Why burden figures for nervous-system disorders move in opposite directions depending on whether you quote counts or rates
- How to state an unsettled mechanism without either overclaiming or saying nothing useful
What to unlearn
- “Older participants had smaller volumes, so volume declines with age.” In a cross-sectional sample that is an age difference, confounded with cohort, survival and measurement changes. A decline is a within-person observation and needs repeated measurement.
- “Normative curves tell me whether this brain is abnormal.” They give a position in a reference distribution. Bethlehem and colleagues published lifespan reference curves for brain structure in Nature in 2022; a centile is a descriptive statistic, not a diagnosis, and no clinical threshold follows from it automatically.
- “The group difference is a feature of the disorder.” It may equally be a feature of its treatment, of comorbidity, of the circumstances that led to diagnosis, or of who is able to tolerate a scanner. Name the competing explanations rather than letting a reviewer supply them.
- “Burden from these conditions is rising sharply.” Between 1990 and 2021, global DALY counts for nervous-system disorders rose 18.2% while age-standardised DALY rates fell 27.0% and age-standardised death rates fell 33.6%. Which figure you quote determines the story you tell.
- “Saying the mechanism is unclear weakens the paper.” It strengthens it, provided you say which parts are unclear and what would resolve them. A vague mechanistic sentence is what gets challenged.
The central problem: a trajectory claim from a snapshot
Nearly everything interesting about the brain across the lifespan is a claim about change: this structure develops along this course, this function declines from this point, this condition alters that trajectory. Nearly all the available data is a set of measurements taken once, from people of different ages. Reading the second as evidence for the first is the defining error in this literature, and it survives in papers that are otherwise careful.
The reason it survives is that the inference feels harmless. If 70-year-olds have smaller mean volumes than 30-year-olds, surely volume declines. But the two groups differ in more than age: they were born decades apart, into different nutritional, educational, medical and environmental circumstances; the older group has survived to be measured, and the people who did not survive are not in it; and the measurement itself may have changed if the data were collected over a long period on upgraded equipment. Any of these can produce a cross-sectional age difference without any individual changing.
The repair is at sentence level. Write “mean volume was lower in older participants” rather than “volume declined with age”, and if your discussion is about decline, say explicitly that the inference requires longitudinal data that this study does not provide. That is one clause, and it is the difference between a finding a reviewer accepts and one they have to correct.
The same imaging result, written as a trajectory and as an observation
Claims a trajectory and a cause
In this cross-sectional sample of 240 adults aged 25 to 80, hippocampal volume decreased with age and was lower in participants with higher inflammatory marker levels. These findings demonstrate that chronic inflammation accelerates age-related hippocampal atrophy, contributing to the cognitive decline observed in later life and identifying inflammation as a modifiable target for preserving brain health.
“Decreased with age” from a single timepoint states within-person change that was never measured; what was observed is a difference between age groups.
“Accelerates” is a rate claim, which requires at least two measurements per person.
Cohort and survivor effects are unaddressed. The 80-year-olds were born fifty-five years before the 25-year-olds and survived to attend.
“Contributing to the cognitive decline observed in later life” adds an outcome the study did not measure at all.
“Modifiable target” is a translational recommendation resting on an undirected cross-sectional association.
No confounders named. Cardiovascular risk, medication and head motion all plausibly produce this pattern.
Reports what was measured
In this cross-sectional sample of 240 adults aged 25 to 80, mean hippocampal volume was lower in older participants and was inversely associated with inflammatory marker concentration (adjusted for age, sex, intracranial volume and head motion). Because all measurements were taken at one timepoint, these are differences between people of different ages rather than within-person change, and we cannot distinguish an age effect from cohort differences or from selective survival into the older strata. The association with inflammation is similarly undirected: raised markers may precede lower volume, follow it, or share a common cause with it, and participants with higher marker levels also differed in cardiovascular risk and medication use. Establishing whether inflammation precedes volume loss requires repeated imaging in the same individuals, which this design cannot substitute for.
The age finding is stated as a between-person difference, with cohort and survival named as alternatives.
The adjustment set is given, including head motion — which a reader of imaging work will look for specifically.
The inflammation association is explicitly undirected, with all three possibilities stated in one clause.
The competing explanations are reported rather than left for a reviewer to raise.
No outcome is discussed that was not measured.
The design that would settle the question is named, which turns a limitation into a direction for the next study.
The second version reports a weaker result from identical data and is the version that survives review, because it never asks the reader to accept an inference the design cannot carry. Notice that nothing here required extra measurement: every change is a statement about what a single-timepoint design reaches. The one addition that did require foresight was head motion, which has to be retained at processing time — which is the general lesson, that honest wording is cheap and the variables that make it possible are not.
What a normative reference curve licenses
Normative modelling has changed how this field reports individual brain measures. Bethlehem and colleagues published brain charts for the human lifespan in Nature — volume 604, issue 7906, in 2022 — providing reference curves for brain structure across the age range 2. The logic parallels paediatric growth charts: a measurement is expressed as a position in a reference distribution for age and sex rather than as a raw number 2.
That is genuinely useful and it is frequently over-read. Three things to keep straight in your writing. A centile is a descriptive position, not a diagnosis; being at the second centile is unusual, and unusual is not the same as pathological. The reference distribution is a property of the sample it was built from, so if your population differs materially from that sample in scanner, sequence, ancestry or health status, the centile carries that mismatch 2. And a single centile gives no information about change: the clinically interesting question is usually whether someone is deviating from their own prior position, which needs repeat measurement on the same curve.
So the usable sentence is of this form: “expressed against published lifespan reference curves, mean cortical volume in our sample fell at the Nth centile for age and sex; because the reference sample differs from ours in [specific respect], we treat this as descriptive and do not interpret individual centiles clinically.” It is modest and it is what the method supports.
Writing about neurodevelopmental change
Developmental claims have an additional problem: the comparison group is also developing. A difference between a clinical and a comparison group of children can reflect a different endpoint, a different rate, a different timing, or simply the age at which you happened to measure — and these imply quite different things. A group that is “behind” at nine may be indistinguishable at fifteen, which is a delay rather than a deficit, and the two words are not interchangeable.
Four reporting habits follow. State the age range and the spread, not just the mean, because a wide age range in a developing sample mixes several developmental stages. Where you have more than one timepoint, model and report the rate as well as the level, since that is where the developmental claim lives. Avoid “delay” and “deficit” unless your design distinguishes them, and say which one you can support. And be explicit about ascertainment: children referred to a clinical service differ systematically from children identified in a population screen, and a sample of the first cannot describe the second.
Retrospective onset is the other recurring weakness. Age at which criteria were first met is usually established by parental or self-report years afterwards, and should be reported as retrospective report rather than as onset. The classification being applied has its own history too — Kawa and Giordano, in Philosophy, Ethics, and Humanities in Medicine, trace roughly sixty years of DSM revision up to the fifth edition’s release in 2013 — so a developmental category defined under one edition is not identical to one defined under another 11. Our page on writing about psychopathology deals with that in more detail.
Writing about neurodegenerative change
At the other end of the lifespan the specific hazards are different but the structure of the error is the same. Three recur.
Survivor effects. Older samples contain the people who survived to be studied and who were well enough to attend. If the exposure or the pathology you are interested in is associated with mortality or with attrition, your sample is selected on it, and the resulting estimate can be attenuated or even reversed. This is worth a named sentence in the limitations, not a generic nod to selection bias.
Pathology, change and diagnosis are three different things. A biomarker, a rate of change on imaging, and a clinical diagnosis are distinct and only partly aligned, and papers routinely slide between them. Be explicit about which one your outcome is, and do not describe a biomarker finding as a finding about the disease unless the link has been established in a design that could show it.
Reverse causation over long intervals. Preclinical disease can precede diagnosis by years, so an exposure measured a few years before diagnosis may be a consequence of early pathology rather than a cause of it — reduced physical activity, weight change, social withdrawal are the standard examples. The usual response is a sensitivity analysis excluding the years closest to diagnosis, and if you did not run one, the limitation should say so specifically.
Biological factor to brain to behaviour: where the language breaks
The characteristic sentence in this field has three links: a biological factor relates to a brain measure, which relates to a behavioural or clinical outcome. Each link is usually established separately and correlationally, and the chain is then written as though it were a pathway. It is worth separating the claims explicitly in your own text, because reviewers do.
For each link, the honest statement names the design. A cross-sectional correlation between a genetic or metabolic variable and a brain measure supports association only. A correlation between the brain measure and the outcome, measured at the same time, supports association only. Putting them together does not produce a mediation claim; mediation requires a design and an analysis built for it, with temporal ordering and sensitivity to unmeasured confounding of the mediator–outcome path. A three-link chain written in causal verbs from two cross-sectional correlations is the most common form of overclaiming in this literature.
The additional problem specific to brain measures is that they sit downstream of a great deal. Medication, substance use, sleep, cardiovascular health, head motion in the scanner, and the stress of being ill all affect imaging measures, and several of them differ systematically between clinical and comparison groups. Reporting head motion, medication exposure and comorbidity by group is not a courtesy; without them, a reader cannot evaluate the finding at all. The same applies to multiple comparisons across many regions or voxels: state the correction and whether the result survives it.
What each design supports about change across the lifespan
| Data you have | Supports | Does not support | The wording that overreaches |
|---|---|---|---|
| One measurement, many ages | An age difference in the sample measured | Change within any individual; a trajectory | “Volume declines from the third decade.” |
| Two or more measurements, same people | Within-person change, and its rate | Cause of that change | “The exposure accelerated atrophy.” |
| Clinical group versus comparison group, once | A group difference at the time of measurement | That the difference is a feature of the disorder rather than of treatment, comorbidity or selection | “The disorder is characterised by reduced volume.” |
| Biological factor correlated with a brain measure | Association between the two | A pathway to any clinical outcome | “The variant affects behaviour via this structure.” |
| A measurement expressed on a normative curve | A position in a reference distribution for age and sex | A diagnosis, or deviation from the person’s own prior course | “The scan shows abnormal development.” |
| Exposure measured a few years before diagnosis | An association with subsequent diagnosis | Causal priority, because preclinical disease can precede diagnosis by years | “Reduced activity contributes to onset.” |
Design capabilities as set out in the STROBE statement and its explanatory material (von Elm and colleagues, BMJ and International Journal of Surgery); normative-curve interpretation follows the reference-distribution logic of Bethlehem and colleagues, Nature. The overreaching phrasings are illustrative examples written for this page.
Burden figures, and which direction the story runs
If your introduction opens with a burden claim, choose the figure deliberately, because the available figures point in different directions. The GBD 2021 Nervous System Disorders Collaborators, in The Lancet Neurology, reported that in 2021 disorders affecting the nervous system were the leading grouped cause of disability-adjusted life-years, accounting for 443 million DALYs (95% uncertainty interval 378–521) and affecting 3.40 billion people, or 43.1% of the world’s population, across 37 conditions in 204 countries and territories 24.
The trend figures from the same analysis are the instructive part. Between 1990 and 2021, DALY counts rose 18.2%, while age-standardised death rates fell 33.6%, age-standardised DALY rates fell 27.0%, and age-standardised prevalence changed by only 1.5% 24. More people are affected because there are more people and they are older; the risk at a given age has fallen substantially 24. An introduction quoting only the count asserts a worsening crisis; one quoting only the rate asserts progress. Both are selective readings of the same analysis.
Two further cautions. The grouping covers 37 conditions affecting the nervous system, so it is not a figure for any one disorder and not a figure for mental disorders, which GBD treats separately 24. And the uncertainty interval is wide — 378 to 521 million — so quote it. Our page on writing clinical research for a public-health audience covers counts and rates in more depth.
Counts and age-standardised rates for the same grouping of conditions over the same period, moving in opposite directions. Figures are for disorders affecting the nervous system as a grouping, not for any single disorder. Source: GBD 2021 Nervous System Disorders Collaborators. Global, regional, and national burden of disorders affecting the nervous system, 1990–2021: a systematic analysis for the Global Burden of Disease Study 2021. The Lancet Neurology.
Where this science is genuinely unsettled
It is worth stating plainly which parts of the lifespan picture are not resolved, because a paper that writes around them reads as either naive or evasive. Four areas, and this is not an exhaustive list.
Normative variation versus pathology. Reference curves describe where a measurement sits in a distribution; what makes a position clinically meaningful is a separate empirical question, and for most brain measures the thresholds are not established. Bethlehem and colleagues’ charts are a description of normative variation, not a diagnostic instrument 2.
Direction of effect for most brain–behaviour associations. Whether a structural or functional difference precedes, accompanies or follows the clinical state is unresolved for most conditions, because the longitudinal datasets starting before onset are few and small.
Whether developmental and degenerative processes share mechanisms. The two ends of the lifespan are often discussed under one framework, and whether that framework reflects shared biology or convenient narrative is open.
Generalisability of imaging normative data. Reference samples are not demographically representative of the world’s populations, and the degree to which centiles transfer across ancestry, scanner and health context is not fully characterised.
Write these as what they are: open questions with specified consequences for your interpretation. “Because the direction of this association is unresolved in the existing longitudinal literature, we report it as a correlate and do not infer that it precedes onset” is a sentence that costs nothing and forecloses an objection.
Reporting the study so it can be appraised
Most lifespan brain work is observational, so STROBE applies: von Elm and colleagues’ statement, in BMJ, sets out 22 items, 18 of which apply to cohort, case-control and cross-sectional designs alike, with 4 design-specific 9,20. The items that bite hardest here are eligibility and selection, variable definition, confounder handling, missing data and participant flow — and in imaging work, exclusions for data quality belong in that flow, reported with reasons and by group, because quality-based exclusion is rarely independent of the clinical status being studied 20.
Three additions specific to this field. Report the acquisition and processing pipeline in enough detail to be reproduced, including software versions, because processing choices change results. Report the multiple-comparison approach and whether findings survive it. And where you use published reference curves, state which, and whether your acquisition is compatible with the one they were built on.
Where your paper is a review rather than a primary study, PRISMA 2020, described by Page and colleagues in BMJ, governs the reporting, and GRADE, described by Guyatt and colleagues in BMJ, gives you a way to say that a body of evidence is low-certainty without dismissing it 4,12,19. Choosing the right framework is usually straightforward; our reporting guidelines page covers the awkward cases.
Take-home message
Take-home message
A lifespan claim is a claim about change, and cross-sectional data cannot support one: write age differences as differences, and name cohort and survivor effects rather than gesturing at selection bias. Normative reference curves, such as the lifespan brain charts Bethlehem and colleagues published in Nature in 2022, give a position in a reference distribution — not a diagnosis, and not deviation from a person’s own prior course. Keep the three links of a biology–brain–behaviour chain separate, and report head motion, medication and comorbidity by group, because without them the finding cannot be assessed. When quoting burden, choose deliberately: nervous-system DALY counts rose 18.2% from 1990 to 2021 while age-standardised rates fell 27.0%.
How to apply this to your study
How to apply this to your own study
- Search your draft for “declines”, “decreases with age”, “accelerates” and “progresses”. If your design has one timepoint per person, every one of these needs rewriting as a difference.
- Name cohort effects and selective survival explicitly in any sample spanning a wide age range, instead of a generic sentence about selection.
- Report your adjustment set in full, including head motion, and state whether findings survive the multiple-comparison correction you used.
- Report quality-based exclusions by group, with reasons. Scan quality is rarely independent of the clinical status under study, so these belong in the participant flow.
- Report medication, comorbidity and substance use by group, or state plainly that they were not assessed.
- Separate the three links of any biology–brain–behaviour chain and state the design behind each. Do not assemble a mediation claim from two cross-sectional correlations.
- If you use published reference curves, name them and state acquisition compatibility, and present centiles as descriptive rather than diagnostic.
- In any pre-diagnosis exposure analysis, run or acknowledge a lag. Excluding the years closest to diagnosis addresses reverse causation from preclinical disease.
- Pick your burden figure deliberately — count, crude rate or age-standardised rate — quote the uncertainty interval, and say which grouping it covers.
- Write one sentence naming the design that would settle your main question. It converts a limitation into something a reader can use.
Common questions
No. A cross-sectional sample gives you a difference between people of different ages, which is confounded with at least three things: cohort effects, since the older participants were born into different nutritional, educational and medical circumstances; selective survival, since only those who lived and were well enough to attend are in the sample; and measurement change, if data were collected over years on upgraded equipment. Any of these can produce an age gradient with no individual changing. Write “mean volume was lower in older participants”, and if your discussion concerns decline, state that the inference requires repeated measurement in the same people. That is one clause and it removes a correction a reviewer would otherwise make.
Where their measurement sits in a reference distribution for age and sex, and nothing more without further evidence. Bethlehem and colleagues published lifespan reference curves for brain structure in Nature, volume 604, issue 7906, in 2022, and the logic is that of a paediatric growth chart. Three limits follow. A low centile is unusual, not pathological; clinical thresholds for most brain measures are not established. The distribution belongs to the sample it was built from, so differences in scanner, sequence, ancestry or health status transfer into the centile. And a single centile says nothing about change — the clinically interesting question is usually deviation from a person’s own earlier position, which needs repeat measurement on the same curve.
As a difference observed at the time of measurement, with the competing explanations reported rather than left implicit. A brain measure sits downstream of medication, substance use, sleep, cardiovascular health, head motion in the scanner and the experience of being ill — and several of these differ systematically between clinical and comparison groups. So report medication exposure, comorbidity and head motion by group, and report quality-based exclusions by group with reasons, since scan quality is rarely independent of clinical status. Then describe what you found without making the disorder the agent: “participants meeting criteria for X had smaller mean volume than comparison participants”, not “X is characterised by reduced volume”.
Only if a design capable of showing a pathway was used. The common pattern is a correlation between a biological variable and a brain measure, a separate correlation between that brain measure and an outcome, and a discussion written as though the chain were established. It is not: mediation requires temporal ordering and an analysis sensitive to unmeasured confounding of the mediator–outcome path, neither of which two cross-sectional correlations provide. Write each link with its own design named and its own undirected phrasing, then say what a mediation design would require. This is the most frequent overclaim in lifespan imaging papers and the easiest for a methodologically literate reviewer to find.
Whichever matches your claim, labelled as what it is, with its uncertainty interval. The GBD 2021 Nervous System Disorders Collaborators reported 443 million DALYs in 2021 (95% UI 378–521) for disorders affecting the nervous system, affecting 3.40 billion people or 43.1% of the world’s population, across 37 conditions in 204 countries. Over 1990 to 2021 the DALY count rose 18.2%, while age-standardised death rates fell 33.6%, age-standardised DALY rates fell 27.0% and age-standardised prevalence changed 1.5%. Counts reflect population growth and ageing; age-standardised rates reflect risk. Quoting only one direction is selective. Note too that this is a grouping of 37 conditions, not a figure for any single disorder.
Treat it as a design problem with a specific remedy, not a limitation to mention. Preclinical disease can precede diagnosis by years, so an exposure measured shortly before diagnosis — reduced physical activity, weight change, social withdrawal — may be an early consequence of pathology rather than a cause. The standard response is a lag analysis: repeat the analysis excluding the years closest to diagnosis and report whether the association persists. If it attenuates, say so. If you could not run such an analysis, state that specifically and name the interval you were unable to exclude, rather than writing that reverse causation cannot be ruled out, which conveys nothing a reader can use.
Name them and state the consequence for your interpretation. Four are worth knowing. The relationship between normative variation and pathology is unresolved for most brain measures, so centiles are descriptive. The direction of most brain–behaviour associations is unresolved, because longitudinal datasets starting before clinical onset are few and small. Whether neurodevelopmental and neurodegenerative processes share mechanisms is open, despite often being discussed under one framework. And imaging normative data are not demographically representative worldwide, so centile transfer across ancestry, scanner and health context is incompletely characterised. A sentence of the form “because X is unresolved, we report this as a correlate and do not infer Y” forecloses an objection at no cost.
STROBE, in most cases. Von Elm and colleagues, in BMJ, set out 22 items, of which 18 apply to cohort, case-control and cross-sectional designs and 4 are design-specific — so almost all of it applies whichever you ran. The items that matter most here are eligibility and selection, variable definition, confounder handling, missing data and participant flow, with quality-based imaging exclusions reported inside that flow, by group and with reasons. Add three field-specific requirements: acquisition and processing pipeline with software versions, since processing choices change results; the multiple-comparison approach and whether findings survive it; and, if you used published reference curves, which ones and whether your acquisition is compatible with theirs.
References and sources
- International Committee of Medical Journal Editors (ICMJE). Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals. 2026. read 9 October 2026. icmje.org
- Bethlehem RAI, Seidlitz J, White SR, et al. Brain charts for the human lifespan. Nature. 2022;604(7906). doi:10.1038/s41586-022-04554-y read 9 October 2026. doi.org/10.1038/s41586-022-04554-y
- Churruca K, Pomare C, Ellis LA, Long JC, Henderson SB, Murphy LED, Leahy CJ, Braithwaite J. Patient-reported outcome measures (PROMs): a review of generic and condition-specific measures and a discussion of trends and issues. Health Expectations. 2021;24(4):1015-1024. doi:10.1111/hex.13254 read 9 October 2026. doi.org/10.1111/hex.13254
- Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021:n71. doi:10.1136/bmj.n71 read 9 October 2026. doi.org/10.1136/bmj.n71
- Jawaid SA, Jawaid M. Common reasons for not accepting manuscripts for further processing after editor's triage and initial screening. Pakistan Journal of Medical Sciences. 2019;35(1):1-3. doi:10.12669/pjms.35.1.28 read 9 October 2026. doi.org/10.12669/pjms.35.1.28
- Sterne JAC, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ. 2019;366:l4898. read 9 October 2026. frontiersin.org
- Fujita MSL, Agustin-Lacruz MdC, Terra AL. Journals' guidelines about title, abstract and keywords: an overview of Information Science and Communication Science areas. European Science Editing. 2018;44(4):76-79. read 9 October 2026. europeanscienceediting.org.uk
- Bannas P, Reeder SB. How to write an original radiological research manuscript. European Radiology. 2017;27(11):4455-4460. doi:10.1007/s00330-017-4879-8 read 9 October 2026. doi.org/10.1007/s00330-017-4879-8
- von Elm E, Altman D, Egger M, Pocock S, Gotzsche P, Vandenbroucke J. The STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) statement: guidelines for reporting observational studies. International Journal of Surgery. 2014. doi:10.1016/j.ijsu.2014.07.013 read 9 October 2026. doi.org/10.1016/j.ijsu.2014.07.013
- Reeve BB, Wyrwich KW, Wu AW, et al. ISOQOL recommends minimum standards for patient-reported outcome measures used in patient-centered outcomes and comparative effectiveness research. Quality of Life Research. 2013;22(8):1889-1905. read 9 October 2026. link.springer.com
- Kawa S, Giordano J. A brief historicity of the Diagnostic and Statistical Manual of Mental Disorders: issues and implications for the future of psychiatric canon and practice. Philosophy, Ethics, and Humanities in Medicine. 2012;7:2. doi:10.1186/1747-5341-7-2 read 9 October 2026. doi.org/10.1186/1747-5341-7-2
- Guyatt G, Oxman AD, Akl EA, et al. GRADE guidelines: 1. Introduction - GRADE evidence profiles and summary of findings tables. Journal of Clinical Epidemiology. 2011;64(4):383-394. read 9 October 2026. link.springer.com
- de Vet HCW, Terwee CB, Mokkink LB, Knol DL. Measurement in medicine. Cambridge University Press. 2011. read 9 October 2026. link.springer.com
- Schulz KF, Altman DG, Moher D. CONSORT 2010 Statement: updated guidelines for reporting parallel group randomised trials. Trials. 2010;11(1). doi:10.1186/1745-6215-11-32 read 9 October 2026. doi.org/10.1186/1745-6215-11-32
- Mokkink LB, Terwee CB, Patrick DL, et al. The COSMIN study reached international consensus on taxonomy, terminology, and definitions of measurement properties for health-related patient-reported outcomes. Journal of Clinical Epidemiology. 2010;63(7):737-745. doi:10.1016/j.jclinepi.2010.02.006 read 9 October 2026. doi.org/10.1016/j.jclinepi.2010.02.006
- Cella D, Riley W, Stone A, Rothrock N, Reeve B, Yount S, et al. The Patient-Reported Outcomes Measurement Information System (PROMIS) developed and tested its first wave of adult self-reported health outcome item banks: 2005-2008. Journal of Clinical Epidemiology. 2010;63(11):1179-1194. doi:10.1016/j.jclinepi.2010.04.011 read 9 October 2026. doi.org/10.1016/j.jclinepi.2010.04.011
- Sijtsma K. On the use, the misuse, and the very limited usefulness of Cronbach's alpha. Psychometrika. 2009;74(1):107-120. read 9 October 2026. link.springer.com
- Kuper A, Reeves S, Levinson W. An introduction to reading and appraising qualitative research. BMJ. 2008;337:a288. read 9 October 2026. link.springer.com
- Guyatt GH, Oxman AD, Vist GE, et al. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. BMJ. 2008;336(7650):924-926. read 9 October 2026. link.springer.com
- von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP, et al. Strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. BMJ. 2007;335(7624):806-808. doi:10.1136/bmj.39335.541782.AD read 9 October 2026. doi.org/10.1136/bmj.39335.541782.AD
- Kliewer MA. Writing it up: a step-by-step guide to publication for beginning investigators. American Journal of Roentgenology. 2005;185:591-596. read 9 October 2026. tech.snmjournals.org
- Sollaci LB, Pereira MG. The Introduction, Methods, Results, and Discussion (IMRAD) structure: a fifty-year survey. Journal of the Medical Library Association. 2004;92(3):364-371. read 9 October 2026. genreacrossborders.uwaterloo.ca
- Huth EJ. Structured abstracts for papers reporting clinical trials. Annals of Internal Medicine. 1987;106(4):626-627. read 9 October 2026. stat.cmu.edu
- GBD 2021 Nervous System Disorders Collaborators. Global, regional, and national burden of disorders affecting the nervous system, 1990-2021: a systematic analysis for the Global Burden of Disease Study 2021. The Lancet Neurology. 23(4):344-381. read 9 October 2026. ro.uow.edu.au
- Radiology (Radiological Society of North America). Radiology scientific style guide and author instructions: scientific editing. Radiology. read 9 October 2026. pubs.rsna.org
- Fernandez E, Garcia AM, Seres E, Bosch F. Students' satisfaction and perceived impact on knowledge, attitudes and skills after a 2-day course in scientific writing: a prospective longitudinal study in Spain. BMJ Open. 8(1):e018657. read 9 October 2026. bmjopen.bmj.com
Every source above is a peer-reviewed article, a publisher’s record of one, or a standard issued by the body that owns it. Each carries the date it was last read. Where a standard moves, this article is revised rather than left up with a note.
How to cite this article
Pick your referencing style and copy it. Your university or target journal may vary the punctuation — check its guide if it has one.
Pubrica Academy. (2026, October 9). Writing about biological factors and brain change across the lifespan. Pubrica. https://pubrica.com/academy/artical/examine-the-relationship-between-biological-factors-and-the-brain-across-the-life-span-as-illustrated-by-neurodevelopmental-and-neurodegenerative-disorders/
Pubrica Academy. “Writing About Biological Factors and Brain Change Across the Lifespan.” Pubrica, 9 Oct. 2026, pubrica.com/academy/artical/examine-the-relationship-between-biological-factors-and-the-brain-across-the-life-span-as-illustrated-by-neurodevelopmental-and-neurodegenerative-disorders/.
Pubrica Academy (2026) Writing about biological factors and brain change across the lifespan. Pubrica. Available at: https://pubrica.com/academy/artical/examine-the-relationship-between-biological-factors-and-the-brain-across-the-life-span-as-illustrated-by-neurodevelopmental-and-neurodegenerative-disorders/ (Accessed: 15 October 2026).
Pubrica Academy. “Writing About Biological Factors and Brain Change Across the Lifespan.” Pubrica. October 9, 2026. https://pubrica.com/academy/artical/examine-the-relationship-between-biological-factors-and-the-brain-across-the-life-span-as-illustrated-by-neurodevelopmental-and-neurodegenerative-disorders/
Pubrica Academy, ‘Writing about biological factors and brain change across the lifespan’ (Pubrica, 9 October 2026) <https://pubrica.com/academy/artical/examine-the-relationship-between-biological-factors-and-the-brain-across-the-life-span-as-illustrated-by-neurodevelopmental-and-neurodegenerative-disorders/> accessed 15 October 2026.
Pubrica Academy. Writing about biological factors and brain change across the lifespan. [Internet]. Pubrica; 2026 Oct 9 [cited 2026 Oct 15]. Available from: https://pubrica.com/academy/artical/examine-the-relationship-between-biological-factors-and-the-brain-across-the-life-span-as-illustrated-by-neurodevelopmental-and-neurodegenerative-disorders/
Pubrica Academy, “Writing about biological factors and brain change across the lifespan,” Pubrica, Oct. 9, 2026. [Online]. Available: https://pubrica.com/academy/artical/examine-the-relationship-between-biological-factors-and-the-brain-across-the-life-span-as-illustrated-by-neurodevelopmental-and-neurodegenerative-disorders/
@misc{examine_the_relationship_between_biolo_2026,
author = {{Pubrica Academy}},
title = {Writing about biological factors and brain change across the lifespan},
howpublished = {Pubrica},
year = {2026},
month = {oct},
url = {https://pubrica.com/academy/artical/examine-the-relationship-between-biological-factors-and-the-brain-across-the-life-span-as-illustrated-by-neurodevelopmental-and-neurodegenerative-disorders/},
note = {Accessed: <span data-cite-acc="iso">2026-10-15</span>}
}This article is written to be learned from. Cite it if you are referring to what it says; if you are using the evidence behind it, cite the sources listed above instead.
Lifespan and brain-change claim audit
Run part A over the whole manuscript, then part B on each sentence that makes a claim about change, cause or clinical meaning. Most papers have fewer than six such sentences, and they are where the review problems come from.
LIFESPAN BRAIN-CHANGE CLAIM AUDIT
Manuscript: ______________________ Design: ______________________
Timepoints per participant: ______ Age range: ______ to ______
PART A - ONCE PER MANUSCRIPT
A1. WHAT MY DESIGN CAN REACH
Circle one:
ONE measurement per person -> age DIFFERENCES only
TWO OR MORE per person -> within-person CHANGE and its rate
[ ] No sentence in the manuscript claims more than the circled option
A2. WORD SEARCH - rewrite every hit if one timepoint per person
[ ] "declines" / "decreases with age" / "increases with age"
[ ] "accelerates" / "slows" / "progresses" / "trajectory"
[ ] "develops" / "matures" (in a cross-sectional sample)
[ ] "atrophy" used where "lower volume" is what was observed
A3. ALTERNATIVES TO AN AGE EFFECT - named explicitly?
[ ] Cohort differences (participants born decades apart)
[ ] Selective survival / selective attendance
[ ] Scanner, sequence or software change over the collection period
[ ] Differential consent or tolerance of the procedure by age
A4. GROUP COMPARISONS - reported BY GROUP?
[ ] Medication exposure .................... ___ or declared unassessed
[ ] Comorbidity ........................... ___ or declared unassessed
[ ] Substance use ......................... ___ or declared unassessed
[ ] Head motion ........................... ___
[ ] Quality-based exclusions, with reasons . ___
[ ] Recruitment route (clinic / community / registry): ____________
A5. IMAGING METHODS A READER NEEDS
[ ] Scanner, field strength, sequence parameters
[ ] Processing pipeline AND software versions
[ ] Intracranial volume or equivalent adjustment stated
[ ] Multiple-comparison approach stated
[ ] Whether each finding survives that correction
[ ] Full adjustment set listed for every model
A6. IF USING NORMATIVE REFERENCE CURVES
Curves used: ______________________________________
[ ] Acquisition compatibility with the reference sample addressed
[ ] Ways my population differs from the reference sample, stated:
______________________________________________________
[ ] Centiles presented as DESCRIPTIVE, not diagnostic
[ ] No claim about deviation from a person's own prior course unless
repeat measurement on the same curve exists
A7. IF EXPOSURES PRECEDE A DIAGNOSIS
[ ] Lag analysis run, excluding the ______ years closest to diagnosis
[ ] If not run, the interval that could not be excluded is named
A8. BURDEN AND CONTEXT FIGURES
For each one cited:
Figure: __________ Type: count / crude rate / age-standardised rate
Grouping it covers: ______________________
Is that grouping what my paper is about? Y / N
Uncertainty interval quoted? Y / N
[ ] If counts and rates point different ways, both are reported
PART B - RUN ON EACH CLAIM SENTENCE
Find them: title; last line of the abstract; first line of the
discussion; the mechanism paragraph; the conclusion.
Sentence: ____________________________________________________
B1. Is this a claim about DIFFERENCE, CHANGE, CAUSE, or CLINICAL
MEANING? __________
B2. Does my design support that category (see A1)? Y / N
B3. If it is causal, which direction is excluded, and by what?
____________________________________________________________
(If nothing excludes the reverse, the verb must change.)
B4. If it chains biology -> brain -> behaviour, is each link stated
separately with its own design? Y / N
B5. Named competing explanations present in the same paragraph?
medication ___ comorbidity ___ motion ___ cohort ___ survival ___
B6. Rewritten sentence:
____________________________________________________________
PART C - THE UNSETTLED SCIENCE, STATED
Normative variation vs pathology: ____________________________
Direction of effect: _________________________________________
Shared vs distinct developmental / degenerative mechanisms: ___
Generalisability of the reference data: ______________________
The design that would settle my main question: _______________ Pubrica AcademyWritten and maintained by the subject matter experts of Pubrica who do this work. Articles are revised when the standard underneath them moves, and each one carries the date it was last checked.
How Pubrica helps
Pubrica’s medical writing and biostatistics teams read lifespan and neuroimaging manuscripts for the mismatch between design and claim: trajectory language on cross-sectional data, mediation claims assembled from separate correlations, and centiles read as diagnoses. We check that the adjustment set, head motion, quality-based exclusions and multiple-comparison handling are all reported, and we rewrite the limitation so it names the design that would resolve the question rather than conceding generically.