The central problem, according to Dadgar and Rostami, is heterogeneity—not merely statistical heterogeneity, but the deeper clinical and methodological diversity baked into the studies being pooled. Homologous and heterologous booster strategies are not single interventions. Each label conceals a matrix of variables: which priming vaccine was used, which booster platform followed, how long the dosing interval was, which age groups were enrolled, and what epidemiological backdrop the trial or observational study operated within. Some studies measured immune responses weeks after the booster; others measured at different time points entirely, and many spanned different periods of variant circulation. When such dissimilar comparisons are folded into a single summary estimate, the result may be a number that is difficult to apply to any specific vaccine sequence or any specific population.

This concern echoes a broader methodological literature. A 2023 review and case study published in Science Advances by Meah and colleagues examined design and analysis heterogeneity in observational studies of COVID-19 booster effectiveness and documented how widely these studies diverge in population definitions, outcome measures, and analytic choices. The Dadgar and Rostami letter leans on such work to make its point: a pooled effect estimate drawn from heterogeneous sources can obscure as much as it reveals. For clinicians deciding whether a particular patient who received an inactivated-virus primary series should receive an mRNA booster, a summary statistic averaged across dozens of different combinations offers little actionable guidance.

The second major critique concerns the evidence base itself. The letter notes that the reviewed studies rely heavily on immunogenicity outcomes—neutralizing antibody titers and cellular immune responses—rather than clinical endpoints such as infection, hospitalization, or death. Immunogenicity measures are indispensable tools. They allow rapid comparison of vaccine strategies without waiting for disease events to accumulate, and they provide mechanistic insight into how different platforms prime and reshape immunity. But antibody titers do not always translate directly into proportional reductions in clinically meaningful outcomes, particularly when the circulating variant changes between the time of measurement and the time of exposure. A higher neutralizing titer against one variant may not predict protection against the next. The authors argue, citing a 2024 systematic review with trial sequential analysis by Asante and colleagues in BMC Medicine, that conclusions about the immunological superiority of heterologous boosting should be carefully distinguished from evidence of improved clinical protection.

Third, the letter highlights prior SARS-CoV-2 infection as a critical and often under-controlled confounder. Previous infection substantially modifies baseline immunity and reshapes the immune response to a booster dose, a phenomenon documented in detail by Wachter and colleagues in a 2025 study in the Journal of Allergy and Clinical Immunology showing that prior infection affects adaptive immune responses to Omicron BA.4/BA.5 mRNA boosters. If included studies differed in how they excluded, measured, or stratified participants by prior infection status, then pooled estimates may partly reflect differences in baseline serostatus rather than the independent effect of homologous versus heterologous boosting. The problem is especially acute in observational studies, where individuals who receive different booster regimens may also differ systematically in age, comorbidity, exposure risk, and access to vaccines—differences that can masquerade as treatment effects in a pooled analysis.

Follow-up duration is the fourth pillar of the critique. Many booster studies report immune responses shortly after vaccination, often within a few weeks, when the post-boost antibody peak is at its highest. The letter questions whether an early immunological advantage necessarily implies sustained superiority over months. A higher peak antibody response may fade faster than a more modest one, and neither pattern fully characterizes protection against severe disease, which depends on memory B cells, T cell responses, and mucosal immunity as much as on circulating antibodies. Recent work on immune durability, including a one-year follow-up study by Awadalla and colleagues in Frontiers in Immunology examining humoral and cellular durability across vaccine platforms after homologous and heterologous boosters, illustrates that the durability question is now being addressed directly—but the letter’s point stands: longer follow-up is needed to determine whether early differences between booster strategies persist and remain clinically meaningful.

Perhaps the most granular objection is that the broad category of heterologous boosting may obscure clinically important differences between specific vaccine combinations. Viral-vector-to-mRNA, inactivated-to-mRNA, mRNA-to-mRNA, and protein-subunit-based booster strategies are immunologically distinct interventions. They engage innate immune pathways differently, present antigen in different formats, and carry different safety profiles and practical implications for cold-chain logistics and programmatic delivery. A single pooled comparison of heterologous versus homologous boosting, the authors argue, loses exactly the granularity that clinicians and immunization program managers need. Subgroup analyses, or more formally network meta-analytic approaches that can rank specific sequences against one another, would better inform decisions about which particular booster combination to deploy in which setting.

None of this amounts to a rejection of the underlying review. Dadgar and Rostami are explicit that Pial and colleagues have addressed an important question and provided a useful overview of the available evidence. Their critique is methodological, not adversarial—a familiar genre in epidemiology, where the gap between a statistically significant pooled estimate and a clinically actionable recommendation is often wide. The letter’s summary is measured: heterogeneity in vaccine platforms, booster combinations, dosing intervals, variant periods, prior infection status, and outcome definitions limits the interpretation of a single pooled estimate. Heterologous boosting may generate higher short-term immune responses in some settings, but its comparative clinical benefit, durability, and applicability to specific vaccine sequences require cautious interpretation.

The practical stakes remain high. Booster policy decisions—whether to offer an mRNA booster after a viral-vector primary series, or a protein-subunit booster after inactivated vaccine—are made under real-world constraints of supply, logistics, and variant evolution. If heterologous strategies do confer durable advantages, mixing platforms could become standard practice; if the apparent advantage is an artifact of pooling heterogeneous studies with unbalanced baseline immunity, resources might be better spent elsewhere. The letter’s recommendations for future evidence syntheses are correspondingly concrete: separate immunogenicity from clinical outcomes, stratify by prior infection status and variant period, and evaluate specific booster combinations rather than treating heterologous boosting as a uniform strategy. As SARS-CoV-2 continues to evolve, the quality of the evidence behind booster sequencing will matter as much as the vaccines themselves.

Subject of Research: Methodological critique of a systematic review and meta-analysis comparing homologous and heterologous COVID-19 vaccine booster strategies

Article Title: Comments on " Homologous and heterologous booster of COVID-19 vaccines: A systematic review and meta-analysis"

Article References: Dadgar, F., & Rostami, M. (2026). Comments on " Homologous and heterologous booster of COVID-19 vaccines: A systematic review and meta-analysis". New Microbes and New Infections, Article 101858. https://doi.org/10.1016/j.nmni.2026.101858

Image Credits: AI Generated

DOI: 10.1016/j.nmni.2026.101858

Keywords: COVID-19 vaccines, heterologous boosting, homologous boosting, systematic review, meta-analysis, immunogenicity, SARS-CoV-2 variants, prior infection, booster durability, vaccine platforms, methodology, Comments

At the heart of the exchange is a question that sounds simple but is anything but: when a device on your wrist reports that you slept six hours and twelve minutes, how much of that number should we believe? Actigraphy, the technique in question, infers sleep from movement. The underlying logic is straightforward—people move less when they are asleep—so algorithms classify each epoch of recorded activity as sleep or wake based on how much motion the accelerometer detects. Modern research-grade devices sample acceleration many times per second and apply validated scoring rules to translate that raw signal into sleep onset, sleep offset, and total sleep time. The approach has become the workhorse of ambulatory sleep research precisely because it can collect data for days or weeks in people’s own homes, something the gold standard of polysomnography, with its electrodes, wires, and laboratory bedrooms, simply cannot do at scale.

But the convenience of actigraphy comes with well-documented trade-offs. Validation studies comparing wrist actigraphy against polysomnography have repeatedly shown that the technique performs well at identifying when people are asleep overall, yet it systematically misclassifies quiet wakefulness as sleep. A person lying still in bed, awake but motionless, looks identical to a sleeping person through the lens of a movement sensor. The practical consequence is that actigraphy tends to overestimate total sleep time relative to polysomnography, and the discrepancy widens in people with fragmented or disturbed sleep, such as patients with insomnia. The American Academy of Sleep Medicine’s systematic review and meta-analysis of actigraphy, cited in the exchange, codified these performance characteristics and remains the reference point for how clinicians and researchers should interpret actigraphic data. Any study that uses actigraphy to make claims about sleep duration must therefore contend with the possibility that some of what it calls sleep is actually stillness.

The second axis of the debate concerns statistics rather than sensors. Khan and Riaz raised concerns about how the original study handled covariates in its regression models, and Baron and Landvatter’s reply engages directly with that critique by invoking a well-known epidemiological pitfall sometimes called the table 2 fallacy. The term, coined by statisticians Daniel Westreich and Sander Greenland, describes a common mistake in which coefficients for confounder variables displayed alongside the main exposure effect in a regression table are interpreted as if they carried the same causal meaning as the exposure itself. In a model estimating how short sleep relates to some health outcome while adjusting for age, sex, and body mass index, the coefficient for body mass index is not necessarily an estimate of the causal effect of body mass index—it is in the model to soak up confounding, and its interpretation depends on assumptions the authors may never have intended to defend.

Related guidance from clinical psychopharmacology researcher Chittaranjan Andrade, also cited in the reply, walks clinicians through what covariates and confounders actually do in adjusted analyses: they change the contrast being estimated. When you add a covariate to a regression, the coefficient on your variable of interest now represents the association at a fixed level of that covariate, holding everything else constant. Whether that adjustment clarifies or distorts the relationship depends on whether the covariate is a true confounder, a mediator on the causal pathway, a collider, or a precision variable. Adjusting for a mediator can erase a real effect; adjusting for a collider can manufacture a spurious one. The reply’s decision to anchor its defense in this literature signals that the authors view the critique as fundamentally a dispute over causal inference conventions rather than over the raw data itself.

Why does this matter beyond the walls of academic sleep medicine? Because actigraphic sleep duration is increasingly treated as a modifiable risk factor, and public health messaging built on observational associations can shift clinical practice and consumer behavior. If a study finds that people whose trackers say they sleep less than seven hours show worse outcomes on some measure, the strength of that finding depends on both the accuracy of the measurement and the integrity of the statistical adjustment. Overestimated sleep duration in a subset of participants could attenuate a true association; mis-specified covariate adjustment could exaggerate or reverse one. The exchange between the Utah team and their critics is thus a compact case study in how the field polices itself, with letters, replies, and methodological citations serving as the peer-review system’s ongoing correction mechanism after initial publication.

The original Baron study deserves attention in its own right for the population it targeted. People who self-report sleeping less than seven hours are a heterogeneous group: some are genuinely short sleepers with objectively verified restricted sleep, while others spend adequate time in bed but sleep poorly, and still others are natural short sleepers who function well on less sleep than average. Prior comparative work, including the community-sample study by Matthews and colleagues cited in the reply, has shown that self-reported habitual sleep, sleep diary estimates, actigraphy, and polysomnography can diverge substantially within the same individuals, with self-report and objective measures sometimes differing by more than an hour. Studying the gap between perceived short sleep and device-measured short sleep is therefore not a niche exercise—it speaks directly to whether subjective complaints of insufficient sleep should trigger the same clinical workup as objectively confirmed sleep restriction.

The technological landscape is also shifting beneath this debate. A 2024 review in Sleep Health compared EEG-based consumer devices, iteratively improved low-cost multisensor trackers, and actigraphy-only devices, reflecting a market in which millions of people now wear sleep trackers daily and researchers increasingly mine that data. Consumer wearables add heart rate, movement, and sometimes temperature signals, and some incorporate limited electroencephalography, promising better discrimination between sleep and quiet wake than movement alone. Yet the validation standards for consumer devices lag behind those for clinical actigraphy, and accuracy can vary across sleep stages, age groups, and clinical populations. The methodological questions raised in this journal exchange—how to score sleep, how to adjust for confounding, how to interpret adjusted coefficients—apply with equal force to the flood of wearable data now shaping popular beliefs about sleep hygiene.

Funding for the underlying research came from the National Heart, Lung, and Blood Institute and the National Center for Advancing Translational Sciences, and the authors report no competing interests. The reply, accepted in June 2026 and published in August as volume 22, article 136 of the journal, is deliberately narrow: it does not claim to resolve the measurement problem, only to clarify what the original analysis did and did not assert. That restraint is itself informative. In a field where sleep duration is routinely framed as a pillar of health alongside diet and exercise, the most rigorous researchers are increasingly explicit that a wrist-worn estimate processed through a regression model is a statistical construct with documented error structure, not a direct readout of biological sleep. Readers interpreting their own tracker data, and clinicians interpreting actigraphic reports, would do well to keep both the sensor’s blind spots and the statistician’s caveats in mind.

What emerges from the exchange is a portrait of a maturing discipline wrestling with its instruments. Actigraphy earned its place in sleep medicine by making large-scale, real-world sleep measurement feasible, and the American Academy of Sleep Medicine’s GRADE-assessed review confirms its clinical utility for evaluating sleep disorders and circadian rhythm disturbances. At the same time, the technique’s known biases and the subtleties of adjusted regression analyses mean that every actigraphic finding carries an error bar that is methodological as much as statistical. The Baron team’s defense, and the critique that provoked it, illustrate how scientific knowledge advances not only through new discoveries but through adversarial scrutiny of existing ones. For a public increasingly obsessed with quantifying its own sleep, the lesson is clear: the number on your wrist is a model’s opinion, and the studies built on such numbers are only as strong as the assumptions buried in their tables.

Subject of Research: Methodological debate over actigraphic measurement of sleep duration and covariate adjustment in sleep research

Article Title: Reply to “Limitations in study of actigraphic estimates of sleep”

Article References: Baron, K. G., & Landvatter, J. (2026). Reply to “Limitations in study of actigraphic estimates of sleep”. Journal of Clinical Sleep Medicine, 22(1), Article 136. https://doi.org/10.1007/s44470-026-00136-1

Image Credits: AI Generated

DOI: 10.1007/s44470-026-00136-1

Keywords: actigraphy, sleep duration, polysomnography, sleep trackers, table 2 fallacy, confounding, regression analysis, sleep medicine, wearable devices, sleep measurement, Journal of Clinical Sleep Medicine, methodology

The technical details of the methodological evolution are striking. In 2022, the ARWU subject methodology rested on five indicators applied uniformly across all disciplines: Top Journal Papers (TOP), which counted publications in high-impact journals selected through the Academic Excellence Survey; Category Normalized Citation Impact (CNCI), which adjusted citation counts by discipline to allow fair comparison across fields; International Collaboration (IC), which measured the proportion of publications co-authored with institutions abroad; Number of Papers (PUB), a pure volume measure of output per institution; and Award, which recognized major academic prizes. In 2023 the structure remained largely intact, with the only notable refinement being the inclusion of 31 conferences within the TOP indicator for Computer Science and Engineering. Education saw no modifications at all that year, a silence the researchers interpret as evidence that the methodology’s designers were still calibrating their metrics for technical fields while leaving the social sciences largely untouched.

Then came 2024, and with it a genuine turning point. The methodology was reorganized into five weighted categories: World-Class Faculty (20 percent), incorporating Highly Cited Researchers along with new Leadership and Editor metrics; World-Class Output (15 percent), focused on high-impact publications; High-Quality Research (15 percent), introducing first-quartile journal papers (Q1); Research Impact (20 percent), retaining CNCI; and International Collaboration (30 percent), whose weighting was dramatically increased. Two changes stand out. The PUB indicator, which had rewarded sheer publication volume, was eliminated entirely, a clear signal that the ranking now prioritizes quality over quantity. And the traditional Award indicator was replaced by leadership metrics such as journal editorship and scholarly leadership roles, shifting recognition from historical prizes to present-day institutional influence within global academic networks.

For Spanish universities, the consequences of these shifts were heterogeneous and, in places, sobering. Between 2022 and 2023, the average CNCI score across the Spanish institutions in the Education category fell by one point, but with high dispersion: some universities, notably Santiago de Compostela and the Open University of Catalonia (UOC), improved markedly while others declined. International collaboration dropped by an average of 7.8 points, a decline the authors link plausibly to lingering mobility restrictions from the COVID-19 pandemic and a possible prioritization of national projects. Most dramatic was the TOP indicator, which fell by an average of 11.2 points, with every Spanish university in the category recording a value of zero in 2023, meaning that no institution managed to publish in the high-impact journals selected by the ranking’s survey. Meanwhile, the Q1 indicator, measuring the share of publications in the top quartile of journals, rose by an average of 4.1 percentage points, suggesting that while Spanish education researchers were publishing in respectable venues, the very top tier remained out of reach.

The 2024 data, however, delivered a genuine surprise: small Spanish universities, including Huelva, Alcalá, Alicante and Córdoba, appeared in the Education category for the first time, joining the giants such as the University of Barcelona and the Autonomous University of Madrid. Their scores tell a nuanced story. On the faculty quality indicator, Huelva scored 22.4, Alcalá 12.9 and Alicante 7.9, well below the consolidated universities whose scores exceed 30 points, reflecting staff with less international projection, likely tied to institutional size and a historical focus on teaching rather than high-impact research. Yet in research quality the picture brightens considerably: Huelva scored 20.3, Alcalá 18.4 and Alicante 26.7, moderate values that, while below the leaders, meet acceptable standards. Research impact proved remarkably homogeneous among the smaller institutions, with values near 33.6 for Huelva and Alcalá and 33.3 for Alicante, demonstrating that despite lower output volumes, their published work resonates within the field.

Why does Education fare so poorly under these metrics in the first place? The study digs into the structural idiosyncrasy of the discipline. Educational research is overwhelmingly produced by educators and trainers disseminating classroom innovations, teaching methods and applied projects, often through qualitative methodologies and with a fundamentally local or national audience. An article about how to teach medicine is likely to appear in a medical journal rather than an education journal, scattering the field’s output across countless venues and complicating any attempt at homogeneous classification. Education also lacks the internationally prestigious prizes that fuel the Award and faculty indicators in Medicine or Physics, and its first-quartile journals are concentrated in English-speaking and Chinese-speaking countries, marginalizing research published in other languages. The authors note that many Web of Science journals indexed in Education-related categories are actually oriented toward educational technology and informatics with a techno-pedagogical focus rather than pedagogy in the broader sense, further skewing what counts as visible scholarship.

The 2024 methodology amplifies some of these problems even as it solves others. The elimination of PUB removes the crude volume penalty that hurt small institutions, but the new World-Class Faculty category rewards universities wealthy enough to attract or train renowned academics, hold journal editorships and accumulate highly cited researchers, generating what the authors describe as structural inequalities between institutions. The International Collaboration category, now weighted at a hefty 30 percent, penalizes universities with limited resources, language constraints or little international tradition; the small Spanish universities scored dramatically low here, with Huelva at 6.8, Alcalá at 6.3 and Alicante at 5.4, far below the national leaders. The researchers point to programs such as Erasmus+ and Horizon Europe, along with Latin American thematic networks that share a common cultural space with Spain, as practical avenues for closing this gap. The CNCI indicator, meanwhile, continues to disadvantage contextualized, regionally relevant studies that naturally attract fewer international citations than work in Physics or Health Sciences.

The authors do not simply critique; they propose a strategy. Spanish universities, they argue, should pursue a diversified approach that improves ranking performance without compromising educational mission: publish strategically in first-quartile journals while maintaining outlets for local and qualitative work; strengthen international networks through collaborative projects and global consortia; invest in faculty development and stability, including nominations for international awards and leadership roles in journals and scientific associations; and build institutional visibility programs that showcase educational innovation and contextual impact. For smaller universities, entry into the ranking should be read not as exposure of weakness but as an opportunity to carve out a differentiated niche, attract external academic talent and consolidate local talent-development programs that could, over time, lift the faculty indicators.

The study is candid about its own limitations. It depends on bibliometric data from Web of Science and Scopus, databases that privilege English-language, high-impact publications and under-represent research in other languages and formats such as books, pedagogical reports and local studies, precisely the formats in which Education produces much of its value. The lack of disaggregated sub-area data within Education makes it hard to assess how indicators affect fields like educational technology, inclusive pedagogy or teacher training. And the ranking’s constant methodological churn, eliminating PUB one year and introducing World-Class Faculty the next, complicates longitudinal comparison. The authors call for alternative metrics that capture local relevance, community impact and pedagogical innovation, for critical review of the major bibliometric databases, and for qualitative case studies and expert interviews to understand how institutions actually navigate these systems.

The broader message resonates far beyond Spain. Global rankings such as ARWU, Times Higher Education and QS have become the dominant lenses through which university prestige is perceived, yet their indicators were designed mainly for scientific and technical disciplines, biasing evaluation against the social sciences and humanities. The 2024 Shanghai methodology represents real progress, a shift from counting papers to weighing quality, leadership and collaboration, but it still embodies what the researchers call a marked positivist vision, prioritizing theoretical, generalizable research over the applied, locally embedded studies that train teachers and shape educational systems. As the authors conclude, the true value of educational research lies in its capacity to transform lives and build inclusive societies, and evaluation systems must evolve so that global excellence and local relevance can coexist, ensuring that small institutions and applied disciplines are not merely counted, but genuinely valued.

Subject of Research: The impact of Shanghai Ranking methodology changes (2022–2024) on the positioning of Spanish universities in the Education category

Article Title: Spanish universities in the Shanghai ranking in education (2022–2024)

Article References: Estrada-Molina, O., Mora-Jaureguialde, B., & Aguaded, I. (2025). Spanish universities in the Shanghai ranking in education (2022–2024). Journal of New Approaches in Educational Research, 14(1), Article 20. https://doi.org/10.1007/s44322-025-00042-z

Image Credits: AI Generated

DOI: 10.1007/s44322-025-00042-z

Keywords: Shanghai Ranking, ARWU, university rankings, Spanish universities, higher education, Education discipline, bibliometrics, research evaluation, international collaboration, methodology, qualitative research, academic rankings

The research team, led by Rosie F. Jackson and Rachael W. Taylor, drew on data from the DREAM study, a randomized crossover trial that manipulated children’s sleep to see how mild sleep deprivation affected diet, activity, and wellbeing. Participants aged 8 to 12 wore a Brinno TLC130 camera on their chest, capturing a static image every two seconds from waking until bedtime over two days in each sleep condition. After applying strict criteria requiring matching data from both intervention conditions, 51 children, just over half of them girls and 14 percent indigenous Māori, provided usable data. The resulting dataset was enormous: 187 observations spanning more than 1.7 million images, covering time before school, after school, and on weekends, when recreational screen use is most likely to occur.

Every image was coded by hand using the free Timelapse 2 software, with two trained coders reaching better than 90 percent agreement before the full dataset was processed. Images were labeled for screens and screen indicators such as keyboards or remotes, for blocked or irrelevant views, and for whether a device was being actively used or merely sitting in the background. Crucially, the researchers made a methodological decision that sets their work apart: instead of baking rules about ambiguous images into the coding protocol itself, they coded each image individually and applied all decision rules afterward, during data processing. That separation meant they could test different assumptions against the same raw data without recoding anything, a luxury most wearable camera studies have never had.

The first question the team examined was how the interval between images affects screen time estimates. Because the cameras had captured a frame every two seconds, the researchers could simulate sparser sampling by systematically discarding images, recreating what a study would have recorded at intervals of 4, 6, 8, 10, 20, 30, or 60 seconds. The results were striking. At the group level, estimates of average screen time barely moved, no matter how widely the frames were spaced. Median differences compared with the dense two-second data were all under two minutes, meaning a study that sampled once a minute instead of twice a second would still arrive at essentially the same population average.

Individual accuracy told a different story. When the researchers looked at how far each child’s estimate deviated from the two-second benchmark, longer intervals introduced more scatter. A 10-second interval emerged as the sweet spot: half of the children had estimates within a minute of the dense-data value, while the workload dropped dramatically. Moving from two-second to 60-second intervals cut the number of images requiring coding to just 3 percent of the original total, a reduction that could shrink months of manual annotation into days. For studies that need reliable data on individual children, such as randomized crossover designs, the message is clear: sample at 10 seconds. For studies that only need group averages, once a minute is enough.

The second question concerned blocked images, one of the messiest realities of wearable camera research. Children turn sideways to talk, slump at the table so the camera points at the floor, or bury themselves under blankets while watching television, leaving the lens staring at fabric or the ceiling. In this dataset, between 4.6 and 19 percent of valid images were blocked. The team tested two rules for handling these gaps. Rule 1, a conservative approach, reclassified up to 10 consecutive no-screen images, a maximum of 20 seconds, as screen time, provided the same device and activity appeared on both sides of the gap. Rule 2 was a deliberate worst-case scenario, allowing unlimited consecutive blocked images to be counted as screen time under the same condition.

The impact of these rules was substantial. Applying Rule 1 increased weekend estimates of total screen use by a median of 8.8 minutes, roughly a 5 percent bump, while the unlimited Rule 2 added a median of 56.4 minutes, an increase of around 32 percent. Effects were smaller before school, when time pressure naturally limits screen opportunities, and larger during after-school and weekend hours when children had freedom to settle in for long sessions. The authors note that Rule 2 likely overestimates true use, since some blocked episodes genuinely involve no screens, but the exercise demonstrates just how much a seemingly innocuous coding assumption can inflate or deflate headline figures. Previous studies have used tolerances ranging from two to 4.5 minutes for blocked images, and because those rules were embedded in coding schedules, their true effect on estimates has never been quantifiable.

The study also distinguished between total screen time, which includes devices visible in the background, and priority screen time, where the child is actively engaged. Priority estimates ran about 10 percent lower than total estimates, a distinction that matters when comparing objective data with questionnaire responses. Parents and children are unlikely to report a television murmuring in the corner of the room, so classifying engagement levels allows researchers to align wearable camera data with the subjective measures that dominate the existing literature, and to understand exactly where the two approaches diverge.

The implications reach beyond screen time research. Wearable cameras are increasingly used to study food marketing exposure, sedentary behavior, and physical activity, and all of these fields face the same unresolved questions about sampling density and missing data. This study provides the first direct evidence that longer capture intervals are far less damaging than commonly assumed, at least for group-level estimates, and that processing rules for obscured images can shift results by tens of minutes per day. The authors recommend that future researchers capture images at 10-second intervals when individual-level accuracy matters, tolerate intervals up to a minute for group comparisons, and, critically, apply assumptions about blocked images during data processing rather than during coding, preserving transparency and the ability to test alternatives.

Challenges remain. Manual coding of millions of images carries a heavy researcher burden, limiting the method to smaller studies where accurate measurement justifies the cost, and the authors point to machine learning as the eventual path forward, though privacy and reliability concerns still need solving. The strict crossover requirements also halved the eligible sample, and the included children differed somewhat from the full cohort in maternal education and household deprivation, though the authors argue this is unlikely to affect conclusions about processing rules. Ethical safeguards, including participant review of images before researchers saw them and secure storage, were followed throughout. What the study delivers is a rare thing in measurement science: a quantified map of how methodological choices ripple through the final numbers. As debates over children’s screen use continue to shape policy and parenting advice, this work suggests the field can now measure the phenomenon with far more confidence, and far fewer photographs, than anyone assumed.

Subject of Research: Objective measurement of children's screen time using wearable cameras and the effect of image capture intervals and data processing rules

Article Title: The influence of different processing rules on wearable camera data estimates of habitual screen time in children

Article References: Jackson, R. F., Meredith-Jones, K. A., Haszard, J. J., Galland, B. C., Morrison, S., Jaques, M., & Taylor, R. W. (2026). The influence of different processing rules on wearable camera data estimates of habitual screen time in children. Journal of Activity, Sedentary and Sleep Behaviors, 5(1), Article 3. https://doi.org/10.1186/s44167-026-00095-1

Image Credits: AI Generated

DOI: 10.1186/s44167-026-00095-1

Keywords: screen time, wearable cameras, children, objective measurement, data processing, image coding, sedentary behavior, sleep deprivation, methodology, digital devices, child health, New Zealand

Free water imaging is an advanced diffusion magnetic resonance imaging approach that attempts to separate the diffusion signal arising from brain tissue from the signal contributed by freely diffusing water in the extracellular space. In principle, the method quantifies a ‘free water fraction,’ a scalar measure that rises when extracellular fluid accumulates. Because neuroinflammation, cell loss, and tissue degeneration are all thought to expand the extracellular space, an elevated free water fraction has been interpreted by many groups as a proxy for neurodegenerative change. The nigrostriatal system, the midbrain circuitry that deteriorates in Parkinson’s disease, has been the principal target of these measurements, and numerous studies have reported elevated free water in the substantia nigra of patients compared with healthy controls.

The appeal of the technique is easy to understand. Parkinson’s disease remains a clinical diagnosis, supported by dopamine transporter imaging and response to levodopa, yet the field has long lacked a biomarker that tracks the underlying biology over time. Clinical rating scales are influenced by medication, symptom fluctuation, and rater variability. Structural atrophy measures change slowly and nonspecifically. Against this backdrop, a diffusion metric that appears sensitive to microstructural change in the substantia nigra, potentially within a single scanning session and without ionizing radiation or contrast agents, has generated considerable enthusiasm, including as a candidate progression biomarker in therapeutic trials.

That enthusiasm, however, has collided with a persistent methodological problem: the free water signal is extraordinarily sensitive to how the data are acquired and processed. The model underlying free water imaging fits a two-compartment representation to diffusion-weighted signals, and this fitting problem is ill-conditioned, meaning that small perturbations in image quality, noise, motion, or gradient performance can shift the estimated free water fraction by amounts comparable to the group differences reported in disease studies. Echo-planar imaging distortions, eddy currents, subject head motion, and even the choice of preprocessing pipeline can leave systematic fingerprints on the resulting maps. Critics have argued that some reported patient-control differences may reflect these technical confounds rather than genuine biology.

The critique that prompted the reply appears to press exactly this point, arguing that free water findings in Parkinson’s disease should not be treated as a binary verdict, for or against the method, but that the method itself matters decisively. The phrase ‘not a binary verdict’ suggests a call for nuance: the question is not simply whether free water imaging works, but under which acquisition protocols, preprocessing choices, and analysis pipelines it can be trusted, and where its limits lie. The reply, published in the same journal, represents the original authors’ defense of their approach and their response to the methodological objections raised.

Debates of this kind are not academic quibbles. If free water imaging is adopted as a secondary or exploratory endpoint in clinical trials, systematic measurement error could obscure true disease slowing, inflate apparent effect sizes, or generate spurious signals that misdirect therapeutic programs. Conversely, if genuine biological signal exists and is dismissed because of technical skepticism, the field may abandon a useful window into neuroinflammation and tissue integrity. The stakes are amplified by the broader push toward biomarker-based staging of Parkinson’s disease, exemplified by recent biological definitions of the disease that incorporate alpha-synuclein seed amplification assays and other molecular measures. Imaging markers that complement these fluid biomarkers would be valuable, but only if their measurement properties are rigorously characterized.

Methodological scrutiny of free water imaging has intensified in recent years. Studies have examined the test-retest reliability of the measure, the influence of scanner vendor and field strength, and the reproducibility of findings across independent cohorts. Some analyses have found that free water elevations in the substantia nigra are robust and correlate with clinical severity, while others have reported that apparent effects diminish or change direction when alternative preprocessing pipelines are applied. Multi-site harmonization efforts have highlighted the difficulty of pooling free water estimates across scanners, and work in other neurological conditions has shown that the metric can be confounded by factors as mundane as ventricular proximity and as consequential as prior imaging artifacts.

Within this contested landscape, the exchange in npj Parkinson’s Disease illustrates how the field is negotiating standards. Replies and counter-replies of this sort serve a function beyond the immediate dispute: they force researchers to articulate the assumptions of their models, the sensitivity analyses they performed, and the conditions under which their conclusions hold. For readers and clinicians, the practical takeaway is that a free water fraction reported in a paper is not a universal constant but the output of a specific acquisition and analysis chain. Comparing values across studies without accounting for those chains risks comparing apples to oranges, a caution that applies to many advanced diffusion techniques, including neurite orientation dispersion and density imaging and related microstructural models.

For patients and families, the debate may seem remote, but its consequences are concrete. Biomarkers determine who is enrolled in trials, when treatments are judged to work, and how quickly disease-modifying therapies reach the clinic. A reliable imaging marker of nigral degeneration could shrink trial sizes, shorten durations, and enable earlier intervention, which is why the National Institutes of Health and the Parkinson’s community have invested heavily in biomarker validation programs. The current exchange should be read as part of that validation process: an insistence that before free water imaging is elevated to a verdict on disease progression, the method must demonstrate that its signal is separable from its noise.

The publication of the reply, alongside the critique it addresses, gives the research community and interested readers an unusually transparent view of a scientific disagreement in progress. Both documents are openly accessible through the journal, allowing independent readers to weigh the arguments for themselves. Whatever the resolution, the episode underscores a principle that extends well beyond this single technique: in neuroimaging, the method matters, and the credibility of any biomarker rests on the reproducibility of the pipeline that produces it. As free water imaging continues to be tested in longitudinal cohorts and interventional studies, exchanges like this one will help determine whether it earns a durable place in the Parkinson’s disease toolkit or remains a promising but contested research measure.

Subject of Research: Free water diffusion imaging as a biomarker of neurodegeneration and disease progression in Parkinson's disease

Article Title: Reply to ‘The method matters: free water imaging in Parkinson’s disease is not a binary verdict’

Article References: Roh, Y. H., Youn, J., Kim, S.-Y., Heo, H., Song, S., & Sohn, B. (2026). Reply to ‘The method matters: free water imaging in Parkinson’s disease is not a binary verdict’. npj Parkinson's Disease, 12(1), Article 218. https://doi.org/10.1038/s41531-026-01490-w

Image Credits: AI Generated

DOI: 10.1038/s41531-026-01490-w

Keywords: Parkinson's disease, free water imaging, diffusion MRI, biomarkers, substantia nigra, neurodegeneration, npj Parkinson's Disease, neuroimaging, clinical trials, methodology, test-retest reliability, nigrostriatal degeneration

The technique rests on diffusion-weighted MRI, which sensitizes the MR signal to the random Brownian motion of water molecules. In a typical acquisition, the signal is measured along many diffusion-encoding directions, and a model is fitted to describe how the apparent diffusion coefficient varies with direction. In most brain tissue, diffusion is restricted and anisotropic, meaning water moves more easily along axonal bundles than across them. Free water imaging extends the standard diffusion tensor model by adding an isotropic compartment: a fraction of the voxel’s water is assumed to diffuse freely and equally in all directions, unconstrained by tissue microstructure. The estimated volume fraction of this compartment, often called the free water fraction, is the quantity that studies have linked to Parkinson’s disease.

What the commentary emphasizes is that this seemingly single number is, in practice, the product of a long chain of decisions. Every stage of the pipeline matters: the strength and number of diffusion-encoding gradients, the number of directions acquired, the echo time and voxel size, the correction for head motion and eddy currents, the approach to removing non-brain tissue, the handling of signal dropout, the fitting algorithm used to estimate the free water fraction, and the way regions of interest are defined in the midbrain. Each of these choices can shift the estimated values, and because different studies make different choices, their results are not always directly comparable.

This matters acutely in Parkinson’s disease research because the effect sizes involved are modest. The changes in free water fraction reported between people with Parkinson’s disease and healthy controls are typically small in absolute terms, often on the order of a few tenths of a percent to a few percent of the signal fraction. When the biological signal is that subtle, even small methodological differences can rival or exceed the effect being sought. A pipeline that smooths data aggressively, or that defines the substantia nigra generously, may report group differences where a more conservative pipeline finds none. Conversely, an underpowered or noisy acquisition may obscure real biology. The commentary’s central claim is that free water imaging findings in Parkinson’s disease should therefore be read as conditional statements, valid for a particular acquisition, preprocessing stream, and region-of-interest strategy, rather than as universal truths about the diseased brain.

The stakes are high because free water imaging has been proposed as a candidate imaging biomarker for disease progression and for use in clinical trials. Several longitudinal studies have suggested that free water fraction in the substantia nigra increases over time in people with Parkinson’s disease, raising hopes that the measure could serve as a sensitive endpoint for disease-modifying therapies. If those hopes are to be realized, the field needs to know how much of the measured change reflects biology and how much reflects the measurement apparatus. A biomarker that drifts with scanner software updates, or that responds more strongly to a change in preprocessing than to a change in the disease, cannot support the weight of a multi-center trial.

The commentary also addresses a conceptual trap: the tendency to interpret an elevated free water fraction as a direct, one-to-one readout of neuroinflammation. The biological rationale is plausible, because inflammatory processes such as astrocytic activation and microglial responses can expand the extracellular space and increase the mobility of water. But elevated free water is not specific to inflammation. Edema, enlarged perivascular spaces, tissue atrophy with partial volume effects from cerebrospinal fluid, and even residual artifacts from motion or susceptibility gradients can all inflate the estimate. Treating free water fraction as a binary indicator of an active inflammatory process, present or absent, oversimplifies what is in fact a composite measurement influenced by multiple tissue properties and multiple sources of error.

Partial volume contamination deserves particular attention in the midbrain, where the structures of interest are small and intimately surrounded by cerebrospinal fluid spaces. The substantia nigra lies adjacent to the interpeduncular cistern, and even with careful region-of-interest placement, signal from free cerebrospinal fluid can leak into the measured voxels, especially at the resolutions commonly used in research scanning. Some pipelines attempt to correct for this, while others rely on conservative masking. The commentary suggests that differences in how this problem is handled may explain a substantial portion of the variability in the literature, with some studies reporting robust group differences and others reporting null results for ostensibly similar comparisons.

None of this, the authors are careful to note, amounts to a dismissal of free water imaging. On the contrary, the technique remains one of the most promising MRI-based approaches to the nigral pathology that defines Parkinson’s disease, precisely because it targets a biologically meaningful property of tissue rather than a gross structural change that appears only late in the disease course. The argument is for methodological transparency and rigor: studies should report their acquisition parameters and preprocessing steps in full, share their analysis code where possible, and validate their pipelines against phantom data or across independent datasets. Harmonization efforts across scanning sites, and sensitivity analyses that show how results change under alternative processing choices, would allow the field to distinguish findings that are robust from those that are artifacts of a particular workflow.

For clinicians and trial designers, the practical message is one of calibrated expectations. Free water imaging is not yet a diagnostic test, and a single elevated value in an individual patient should not be read as a verdict on their disease state. The technique’s near-term value lies in group-level comparisons and longitudinal tracking within carefully controlled studies, where its sensitivity to change can be exploited while its methodological dependencies are held constant. As the field moves toward standardization, the commentary argues, the goal should be pipelines whose outputs are stable across sites and scanners, so that the biological signal of neurodegeneration can finally be separated from the technical noise of measurement. In free water imaging, the method is not a mere technicality; it is part of the result itself, and recognizing that is the first step toward turning an intriguing research measurement into a dependable clinical tool.

Subject of Research: The influence of image processing methodology on free water imaging measurements in Parkinson's disease

Article Title: The method matters: free water imaging in Parkinson’s disease is not a binary verdict

Article References: The method matters: free water imaging in Parkinson’s disease is not a binary verdict. (n.d.). https://doi.org/10.1038/s41531-026-01492-8

Image Credits: AI Generated

DOI: 10.1038/s41531-026-01492-8

Keywords: Parkinson's disease, free water imaging, diffusion MRI, neuroinflammation, biomarkers, image processing, substantia nigra, magnetic resonance imaging, neurodegeneration, methodology, method, matters

The letter arrives as a direct response to an updated meta-analysis of randomized controlled trials of acceptance and commitment therapy for insomnia, published earlier in the same journal by Barroso and colleagues. That meta-analysis pooled trial data to estimate how well ACT performs against control conditions and other active treatments, and its findings have been interpreted by some as support for expanding ACT’s footprint in behavioral sleep medicine. Saripudin and Saputri do not dispute the underlying data. Instead, they use the meta-analysis as a springboard to ask a more fundamental question: are current trials of ACT for insomnia designed well enough to tell us where the therapy genuinely helps, for whom, and at what point in a treatment pathway it should be deployed?

Acceptance and commitment therapy differs from standard CBT-I in a philosophically and mechanically important way. Rather than targeting sleep directly through stimulus control, sleep restriction, and cognitive restructuring of dysfunctional beliefs about sleep, ACT draws on relational frame theory and psychological flexibility research, a framework codified by Steven Hayes and colleagues in their foundational 2006 model of the approach. In the insomnia context, this means helping patients observe their sleep-related thoughts and anxieties without struggling against them, defusing from catastrophic predictions about the consequences of a bad night, and redirecting energy toward valued activities regardless of how sleep unfolds. The metacognitive model of insomnia developed by Ong, Ulmer, and Manber in 2012 provides the theoretical scaffolding here: insomnia is maintained not simply by poor sleep but by excessive monitoring of sleep, worry about sleep, and futile efforts to control a largely involuntary process.

The first methodological challenge highlighted by the letter concerns the heterogeneity of what actually gets called ACT in insomnia trials. Component network meta-analysis of CBT for chronic insomnia, such as the 2024 systematic review by Furukawa and colleagues in JAMA Psychiatry, has shown that the specific ingredients bundled inside a treatment package matter, and that different delivery formats produce different effect sizes. When one trial delivers pure acceptance and defusion exercises while another folds in sleep restriction, relaxation training, and psychoeducation under the ACT banner, the pooled estimate from a meta-analysis becomes difficult to interpret. Saripudin and Saputri argue that future trials must specify and measure the treatment components with the same rigor applied to pharmacological dosing, otherwise clinicians cannot know whether it is the acceptance processes or the borrowed behavioral elements driving observed improvements.

A second challenge involves the measurement of the very mechanisms ACT claims to target. If psychological flexibility, acceptance, and defusion are the active ingredients, then trials should track changes in these processes and demonstrate that they mediate sleep outcomes. Many existing studies measure only symptom endpoints such as insomnia severity or sleep-onset latency, leaving a black box between intervention and effect. Without mediation analyses and process measures, the field cannot distinguish a genuine mechanism-driven therapy from a repackaged relaxation intervention. The letter’s authors contend that this measurement gap is not a technical detail but the central scientific question, because it determines whether ACT is a distinct therapeutic modality for insomnia or simply a sympathetic framing of existing techniques.

The third pillar of the letter is the proposal to integrate ACT into a stepped-care architecture for insomnia. Stepped-care models allocate treatment intensity according to severity and prior response, beginning with low-intensity, widely accessible interventions and escalating only when needed. In such a system, ACT could occupy a strategically valuable position: for patients who refuse or fail sleep restriction, for those whose insomnia is entangled with chronic pain, anxiety, or depression, and for individuals whose primary problem is the struggle against sleep rather than sleep itself, acceptance-based approaches may be more acceptable and more effective than classic CBT-I protocols. Positioning ACT as either a first-line option for struggle-dominant presentations or a second step after CBT-I nonresponse is, in the authors’ view, a testable and clinically meaningful hypothesis that current trial designs have not yet addressed.

Underpinning this stepped-care argument is the often-overlooked issue of acceptability. Research on adherence to CBT-I, including the systematic review by Matthews and colleagues published in Sleep Medicine Reviews, has documented that many patients abandon sleep restriction partway through, and studies of patient perceptions toward insomnia treatments by Cheung and colleagues have shown that people frequently hold reservations about behavioral prescriptions that initially worsen sleep. Acceptability, the letter argues, should be treated as a formal outcome metric rather than an afterthought. Trials should systematically quantify tolerability, dropout attributable to treatment burden, patient-rated willingness to continue, and the match between treatment demands and patient preferences. If ACT’s chief comparative advantage is that patients will actually do it, then the field needs standardized instruments to demonstrate that advantage rather than relying on anecdote.

The implications of the letter extend beyond behavioral sleep medicine into health policy and service design. Insomnia affects a substantial share of adults, contributes to accidents, cardiovascular and metabolic disease, and psychiatric morbidity, and consumes enormous healthcare resources, yet access to trained CBT-I therapists remains scarce in most health systems. Digital and guided self-help formats have been proposed to close the access gap, and the component questions raised by Saripudin and Saputri apply with equal force here: if ACT is delivered through an app or a brief primary-care consultation, which elements survive the compression, and do the acceptance processes still engage? The authors suggest that acceptability metrics become a gating criterion in stepped-care algorithms, allowing services to route patients toward the least intensive treatment they will actually complete, which maximizes population-level benefit from limited clinician time.

None of this diminishes the significance of the updated meta-analysis that prompted the correspondence. Pooling randomized controlled trials is exactly the right starting point, and the finding that ACT produces measurable benefits for insomnia patients is noteworthy on its own terms. The letter’s contribution is to insist on intellectual honesty about what the pooled estimate can and cannot support. Effect sizes derived from heterogeneous packages, without mechanism data, without acceptability measurement, and without head-to-head comparisons at defined steps of a care pathway, justify curiosity rather than clinical rollout. The distinction matters because premature scaling of an incompletely characterized therapy risks repeating the access-versus-evidence tensions that have complicated other behavioral health rollouts.

What happens next will depend on whether trialists take up the design challenges laid out in the letter. The prescriptions are concrete: pre-register component definitions, include validated process measures of psychological flexibility and acceptance, conduct formal mediation analyses, report standardized acceptability outcomes, and embed ACT within stepped-care trials that compare explicit treatment sequences rather than isolated packages. If the field responds, the result could be a genuinely personalized behavioral sleep medicine, in which a patient whose insomnia is driven by nocturnal struggle is steered toward acceptance-based work while a patient with entrenched sleep-incompatible habits receives restriction-based CBT-I first. If the field does not respond, acceptance and commitment therapy risks remaining an attractive but under-specified option, championed by enthusiasts and underused by systems. Saripudin and Saputri’s letter is, in effect, a roadmap for turning promising pooled statistics into a defensible, patient-centered treatment strategy, and it arrives at a moment when demand for better insomnia care has never been higher.

Subject of Research: Optimizing acceptance and commitment therapy for insomnia through improved trial methodology, stepped-care integration, and acceptability measurement

Article Title: Optimizing the role of acceptance and commitment therapy for insomnia: methodological challenges, stepped-care integration, and acceptability metrics

Article References: Saripudin, M., & Saputri, H. (2026). Optimizing the role of acceptance and commitment therapy for insomnia: methodological challenges, stepped-care integration, and acceptability metrics. Journal of Clinical Sleep Medicine, 22(1), Article 160. https://doi.org/10.1007/s44470-026-00184-7

Image Credits: AI Generated

DOI: 10.1007/s44470-026-00184-7

Keywords: acceptance and commitment therapy, insomnia, cognitive behavioral therapy for insomnia, stepped care, psychological flexibility, sleep medicine, meta-analysis, treatment acceptability, randomized controlled trials, behavioral sleep medicine, adherence, methodology

The letter is a critical appraisal of a real-world study by Ahn and colleagues that examined whether GLP-1RAs act as cardiovascular risk modifiers in people with obstructive sleep apnea and obesity. Real-world studies of this kind have become enormously influential because they mine large clinical databases to answer questions that randomized trials either have not yet addressed or cannot practically address. With millions of patients now prescribed GLP-1RAs for type 2 diabetes, obesity, and increasingly for their demonstrated cardiovascular benefits, the stakes for getting these observational analyses right could hardly be higher. Policy decisions, prescribing patterns, and reimbursement frameworks increasingly rest on the kind of database evidence that Sarıoğlu’s letter scrutinizes.

At the heart of the critique lies a technical concept that has shaped pharmaco-epidemiology for nearly two decades. Immortal time bias arises when a period of time during which the outcome of interest cannot occur is improperly included in one group’s follow-up, typically the treated group. The classic formulation comes from epidemiologist Samy Suissa, whose 2008 paper in the American Journal of Epidemiology laid out how this bias operates: if researchers define the exposed group by a prescription that occurs sometime after cohort entry, but count that patient’s follow-up from the moment of entry, the patient must survive long enough to receive the prescription. That guaranteed survival window, the ‘immortal time,’ makes the treated group appear artificially protected, generating spuriously favorable results for the drug.

Sarıoğlu points out that this structure is particularly easy to fall into when studying GLP-1RAs, because these drugs are often initiated months or even years after a patient first enters the health system with obesity, sleep apnea, or diabetes. When investigators anchor their analysis at the date of an obstructive sleep apnea diagnosis or at a baseline clinic visit, but classify patients as GLP-1RA users only once a prescription appears later in their record, the exposed group has, by construction, accumulated event-free time before treatment ever began. Unless the analysis explicitly accounts for that window—through techniques such as time-dependent exposure modeling, matching on the time to treatment, or active-comparator new-user designs—the resulting hazard ratios can suggest cardiovascular protection that reflects study design rather than pharmacology.

The letter also situates its argument within a broader and ongoing refinement of how epidemiologists understand these biases. A 2025 paper by Miguel Hernán and colleagues in the journal Epidemiology provided a structural description of the family of biases that generate immortal time, framing them through the lens of causal diagrams and target trial emulation. That work emphasized that immortal time bias is not a single mistake but a constellation of design choices—how cohorts are defined, how exposure is classified, how follow-up begins and ends—that collectively manufacture a comparison between people who could not yet have experienced an event and those who could. Sarıoğlu’s letter applies this modern framework to the specific case of GLP-1RAs in sleep apnea populations, effectively asking whether the original study emulated the randomized trial it intended to mimic.

The target trial framework, as it is known, asks investigators to specify, before touching the data, the randomized trial they would ideally conduct: who would be eligible, how treatment would be assigned, when follow-up would start, and what outcome would be measured. In a well-executed emulation, the moment of cohort entry and the moment treatment is assigned coincide, or the analysis explicitly handles the gap between them. When they diverge—as they do whenever a prescription recorded at an arbitrary later date defines the exposed group—the emulation drifts away from the trial it was meant to mirror, and the divergence is precisely where bias enters. Sarıoğlu’s central claim is that the timing of exposure classification in real-world cardiovascular analyses of GLP-1RAs represents exactly such a divergence, and that readers should interpret effect estimates from such studies with corresponding caution.

Why does this matter so much for this particular drug class and this particular patient population? Obstructive sleep apnea affects roughly a billion people worldwide and is strongly associated with obesity, hypertension, arrhythmias, and increased cardiovascular mortality. GLP-1RAs have generated intense excitement because randomized trials in other populations, notably patients with type 2 diabetes and established cardiovascular disease, showed meaningful reductions in major adverse cardiovascular events. Translating those findings to sleep apnea populations through observational data is an attractive and legitimate research strategy. But it is also a strategy in which the exposure is highly patterned by the very health trajectories under study: patients who remain well enough, engaged enough with care, and clinically stable enough to receive a GLP-1RA prescription are systematically different from those who deteriorate, drop out, or die before such a prescription is written. Any analysis that does not neutralize this selection can convert healthier-patient dynamics into apparent drug benefit.

The letter does not claim that GLP-1RAs lack cardiovascular benefits, nor does it assert that the original study’s conclusions are necessarily wrong. Its point is narrower and, in a sense, more important: the direction and magnitude of any bias introduced by exposure timing cannot be determined from the published results alone, and the credibility of real-world evidence for this drug class depends on design features that must be transparently reported. Sarıoğlu, writing as the sole author of the letter, conceived the commentary, reviewed the relevant literature, and drafted and revised the manuscript, drawing on no external funding and declaring no competing interests. The letter is based exclusively on critical appraisal of previously published work and involves no new data collection, which means its contribution is methodological rather than empirical—it is a lens, not a dataset.

The wider lesson extends well beyond sleep medicine. As GLP-1RAs are studied for an ever-expanding list of outcomes—from kidney disease and heart failure to dementia and addiction—real-world database studies will continue to proliferate, and each carries the same vulnerability if exposure timing is mishandled. The epidemiological community has developed reliable remedies: defining cohort entry at the moment of treatment eligibility, modeling exposure as a time-varying covariate, using new-user designs that exclude prevalent users, and emulating target trials with explicit cloning, censoring, and weighting strategies. Sarıoğlu’s letter serves as a reminder that applying these tools is not pedantic hair-splitting but the difference between evidence that can guide patient care and evidence that merely reflects who managed to stay alive and in care long enough to fill a prescription.

Published on 9 September 2026 as a letter to the editor in the Journal of Clinical Sleep Medicine, the commentary adds a careful methodological voice to one of the most consequential drug-evidence debates of the decade. Whether future real-world analyses of GLP-1RAs in obstructive sleep apnea and obesity will confirm, revise, or overturn the cardiovascular signals reported to date remains an open question. What the letter makes clear is that answering it responsibly requires paying close attention not just to whether patients took these drugs, but to precisely when the clock on their follow-up started—and whether that clock was fair to both the treated and the untreated.

Subject of Research: Methodological bias in real-world observational studies of GLP-1 receptor agonist exposure timing and cardiovascular risk in obstructive sleep apnea and obesity

Article Title: Timing of GLP-1RA exposure in real-world cardiovascular risk analyses

Article References: Sarıoğlu, G. (2026). Timing of GLP-1RA exposure in real-world cardiovascular risk analyses. Journal of Clinical Sleep Medicine, 22(1), Article 161. https://doi.org/10.1007/s44470-026-00188-3

Image Credits: AI Generated

DOI: 10.1007/s44470-026-00188-3

Keywords: GLP-1 receptor agonists, immortal time bias, cardiovascular risk, obstructive sleep apnea, obesity, pharmaco-epidemiology, real-world evidence, observational studies, target trial emulation, type 2 diabetes, methodology, Journal of Clinical Sleep Medicine