The questionnaire, known as the GLAM Lifestyle Questionnaire, was developed by the Gut Microbiota Study Group of the Italian Association of Clinical Microbiologists. Its candidate items cover dietary patterns, exercise and sleep, three domains associated in previous human research with variation in the gut microbiota. The underlying objective is methodological rather than therapeutic: microbial sequencing can describe communities, but standardized information about the people carrying them may help researchers interpret differences between individuals and compare results across cohorts.

Candidate content was generated through a focused narrative review of human observational and intervention studies, together with established lifestyle-assessment frameworks including the Food Frequency Questionnaire, the International Physical Activity Questionnaire and the Pittsburgh Sleep Quality Index. This literature work provided the rationale for asking candidate questions, but it was not a systematic review. Complete database exports and record-level selection decisions were unavailable, no formal risk-of-bias assessment or certainty grading was performed, and the researchers did not pool the evidence into a common effect estimate.

The developmental process was initially described as a modified Delphi exercise, but the researchers characterized what they actually conducted as an initial expert appraisal. Seventeen experts from the GLAM network were invited, and nine completed the assessment, giving a participation rate of 53%. The respondents represented expertise in clinical microbiology, microbiota research, infectious diseases, gastroenterology and microbial genomics.

Each candidate item was scored on a five-point scale ranging from “not relevant” to “highly relevant,” with ratings of 4 or 5 counted as positive judgments of relevance. An item met the study’s prespecified agreement criterion when at least 70% of respondents assigned it a 4 or 5. With nine panelists, that threshold required at least seven positive ratings, or 77.8%; six positive ratings represented 66.7% and fell below the cutoff.

The criterion was not changed after results were seen. Among the 25 items classified as supported, 16 received seven positive ratings, eight received eight and one received nine. Six of the 12 items below the threshold received six positive ratings, so nearby classifications should be interpreted cautiously because each expert represented 11.1 percentage points.

Diet received the strongest panel support, with 18 of 23 candidate items meeting the threshold. Dietary fiber received nine positive ratings, the only item in the entire questionnaire to receive a rating of 4 or 5 from every expert. The panel also judged numerous questions about fruit, vegetables, whole grains, fermented foods, meat, fish, beverages, ultra-processed foods, fast food, water intake, probiotic and prebiotic foods, and plant-based consumption to be relevant.

Five dietary questions remained exploratory: type of drinking water, organic food consumption, cooking methods, type of sweeteners and food purchasing location. Physical activity also fared relatively well, with six of eight candidate items meeting the threshold. The supported questions concerned frequency and duration of vigorous activity, total weekly activity time, sedentary behavior, participation in structured exercise and occupational physical activity.

Moderate-activity frequency and walking frequency each received six positive ratings and were retained as exploratory rather than rejected outright. Only total sleep duration crossed the sleep threshold. Questions about sleep-onset latency, wake-up time, subjective sleep quality, night awakenings and sleep-medication use received fewer than seven positive ratings.

The contrast among domains reflects the judgments of this particular panel; it does not prove that diet has stronger biological effects or that sleep and activity are unimportant. Published findings for all three domains are heterogeneous, and human evidence linking specific sleep and activity measures to microbial outcomes remains limited. Likewise, the literature appraisal does not establish that every supported question has an independent microbial association, a causal effect or a validated response threshold.

Agreement here is an expert relevance judgment, not a grade of evidence certainty or a clinical recommendation. All 37 questions remain candidate content rather than a finalized scale. The experts rated their potential relevance to microbiome research, not whether respondents would understand them, whether the questions captured lifestyle comprehensively, or whether the answers were accurate.

No participant-level lifestyle or microbiome data were collected, and no composite lifestyle score was developed. The measurement properties of the source questionnaires also cannot simply be transferred to the GLAM items because their wording, response categories and intended context differ. Several technical limitations constrain the conclusions.

The exercise contained only one rating round, without controlled feedback or repeated voting, so the stability of the apparent agreement is unknown. It should not be treated as a stable conventional Delphi consensus. The original individual responses were unavailable for independent reanalysis, and five reported medians were mathematically incompatible with their positive-rating counts, so the paper omits those medians and related quartile summaries.

The small panel, incomplete participation and exclusively Italian setting also make the findings preliminary and potentially sensitive to non-response and cultural bias. Before research deployment, the authors call for further expert rounds, operational refinement, cognitive interviews, feasibility testing and assessment of repeatability in an Italian population. Terms such as “regularly,” “frequently,” “often” and “adequate” need explicit recall periods and response options.

Self-reports may also introduce recall and social-desirability errors. The module does not cover the full exposome: antibiotics and other medications, psychological and socioeconomic conditions, travel, household and pet exposures, and occupational or environmental factors would need to be collected separately within a broader host-exposure framework. One proposed future application is the characterization of fecal microbiota transplantation donors.

Systematic lifestyle metadata might add context to donor microbiota, but the study provides no evidence that questionnaire responses predict donor eligibility, microbiota quality, treatment response or the proposed “super-donor” phenomenon. Its narrower claim is that a modular lifestyle survey could standardize information often missing from sequencing studies. The next empirical test is whether collecting these variables makes microbiome analyses more reproducible or improves interpretation beyond clinical, pharmacological and demographic covariates.

Subject of Research: Development and initial Italian expert appraisal of a lifestyle questionnaire for microbiome host metadata

Article Title: Development and expert appraisal of a lifestyle questionnaire for microbiome host metadata

Article References: Rocchetti, A., Gotta, F., Bottino, P., Vay, D., Pizzo, V., Gugliotta, V., Grossi, A. A., Besutti, V., Bianchetti, M., Bona, E., Castagliuolo, I., Corbella, M., Gasbarrini, A., Marangoni, A., Masucci, L., Mauro, M. V., Putignani, L., Quaranta, G., Sanguinetti, M., & Drago, L. (2026). Development and expert appraisal of a lifestyle questionnaire for microbiome host metadata. Frontiers in Cellular and Infection Microbiology, 16, Article 1932122. https://doi.org/10.3389/fcimb.2026.1932122

Image Credits: AI Generated

DOI: 10.3389/fcimb.2026.1932122

Keywords: gut microbiome, host metadata, questionnaire development, expert appraisal, diet, physical activity, sleep, fecal microbiota transplantation, methodological research, Italian panel, Development, expert