Why English Matters in STROME-ID Reporting

Why English? • Molecular evidence

Genetic similarity is evidence, not a complete transmission story.

STROME-ID shows how careful language connects specimens, laboratory methods, pathogen diversity, time, epidemiology and ethical limits.

Why does English matter when molecular data are used to study infectious diseases? Because a sequence does not arrive with its own interpretation. Researchers must explain how samples were selected, how laboratory results were produced, what counts as similar, how quickly the organism changes and whether place, time and contact information support the same story.

STROME-ID—Strengthening the Reporting of Molecular Epidemiology for Infectious Diseases—is an extension of STROBE for observational research that uses infectious-disease molecular data. The EQUATOR Network entry identifies its scope, while the open statement and PubMed record document the 2014 publication. It adds 20 items to the 22-item STROBE checklist.

The guidance addresses terminology, pathogen diversity, laboratory methods, sample collection, molecular markers, molecular clocks, time frames, multiple-strain infection, dependence between observations, missing data, ascertainment bias, consistency between molecular and epidemiological evidence, and ethics. Those are not merely technical headings. Each is a place where imprecise English can turn a qualified finding into an overconfident narrative.

For students searching how to read pathogen sequencing studies, what molecular epidemiology can prove or why sampling bias matters in infectious disease research, the central lesson is wonderfully transferable: follow the evidence chain, and never let one attractive result erase the conditions that produced it.

Three languages in one study: molecular, epidemiological and inferential

Molecular language describes sequences, markers, alleles, variants, genetic distances and phylogenetic relationships. Epidemiological language describes people, exposure, place, time, symptoms, diagnosis and contact. Inferential language connects the two: consistent with, suggests, supports, cannot exclude or, much more rarely, demonstrates. Confusion often begins when a term from one layer silently changes meaning in another.

For example, two samples may be genetically indistinguishable under the chosen method. That does not automatically show direct transmission from person A to person B. There may be an unsampled intermediary, a shared source, limited diversity in the wider population or insufficient resolution in the marker. The accurate sentence states the molecular result first, then explains which transmission scenarios remain compatible with it.

Time also has several meanings. Specimen collection date is not necessarily infection date. Symptom onset may be recalled imperfectly. Diagnosis depends on testing access. A molecular clock estimates evolutionary time using assumptions and uncertainty. Strong reporting labels each time point so the reader does not mistake one for another.

This disciplined separation connects with eduKate’s guide to scientific evidence-state reporting. Verbs should match the evidential step. “Samples clustered” is a molecular result. “People were linked” is an epidemiological result. “Transmission may have occurred within the setting” is an interpretation that requires both.

Thirteen critical links from specimen to conclusion

The official checklist contains additions positioned across the STROBE structure. The following thirteen-link chain groups related ideas for learning. It does not replace the published checklist; it helps students see how one missing explanation can affect every conclusion downstream.

1. Terminology

This link asks writers to define words such as strain, type, cluster, lineage, relatedness and transmission rather than assuming every field uses them identically. The central reading question is not “Is there molecular data?” but “What information can this molecular evidence resolve under these particular conditions?” That question keeps the interpretation proportional to the method.

If “cluster” means a genetic grouping in one paragraph and a linked outbreak in another, the conclusion may slide from molecular similarity to epidemiological contact without evidence. A polished diagram can make the problem hard to notice because visual order feels like causal order. Readers should return to the prose and locate the sample frame, measurement rule, time relation and uncertainty that support each edge or cluster.

Create a glossary for study-specific terms and state the operational rule, threshold or algorithm behind each label. This reporting move gives another team a fair chance to understand the finding and its limits. It also improves classroom discussion: students can point to a missing link in the evidence chain instead of dismissing an entire study with a vague claim that “correlation is not causation”.

A useful sentence frame is: “Using [method and threshold], the study observed [molecular result] among [sampled population and period]; this is [consistent or inconsistent] with [epidemiological information], but [named limitation] leaves [alternative explanation] possible.” The brackets force important scope back into the sentence.

2. Pathogen diversity

This link asks writers to describe the amount and pattern of genetic or molecular variation in the relevant pathogen population. The central reading question is not “Is there molecular data?” but “What information can this molecular evidence resolve under these particular conditions?” That question keeps the interpretation proportional to the method.

A marker that separates cases well in a diverse population may add little information where nearly all samples are similar. Readers need the background diversity to interpret apparent matches. A polished diagram can make the problem hard to notice because visual order feels like causal order. Readers should return to the prose and locate the sample frame, measurement rule, time relation and uncertainty that support each edge or cluster.

Report diversity measures, reference populations and the resolution that the chosen method can realistically provide. This reporting move gives another team a fair chance to understand the finding and its limits. It also improves classroom discussion: students can point to a missing link in the evidence chain instead of dismissing an entire study with a vague claim that “correlation is not causation”.

3. Sampling frame and specimen collection

This link asks writers to show who, where and when samples came from, and how clinical or environmental specimens entered the study. The central reading question is not “Is there molecular data?” but “What information can this molecular evidence resolve under these particular conditions?” That question keeps the interpretation proportional to the method.

Convenience samples can over-represent severe cases, tested patients, particular facilities or accessible locations. Missing specimens may follow a pattern rather than occur randomly. A polished diagram can make the problem hard to notice because visual order feels like causal order. Readers should return to the prose and locate the sample frame, measurement rule, time relation and uncertainty that support each edge or cluster.

Map eligible cases to sampled cases, state specimen types and timings, and explain reasons for absent or unusable samples. This reporting move gives another team a fair chance to understand the finding and its limits. It also improves classroom discussion: students can point to a missing link in the evidence chain instead of dismissing an entire study with a vague claim that “correlation is not causation”.

4. Laboratory methods

This link asks writers to describe extraction, amplification, sequencing, typing, quality control and contamination precautions with enough detail to understand comparability. The central reading question is not “Is there molecular data?” but “What information can this molecular evidence resolve under these particular conditions?” That question keeps the interpretation proportional to the method.

A result can change with laboratory platform, protocol, depth, thresholds and quality filters. Naming only the commercial kit or method family leaves important decisions invisible. A polished diagram can make the problem hard to notice because visual order feels like causal order. Readers should return to the prose and locate the sample frame, measurement rule, time relation and uncertainty that support each edge or cluster.

Report protocol versions, controls, failure criteria, repeats, laboratory blinding where relevant and how batches were handled. This reporting move gives another team a fair chance to understand the finding and its limits. It also improves classroom discussion: students can point to a missing link in the evidence chain instead of dismissing an entire study with a vague claim that “correlation is not causation”.

5. Molecular markers

This link asks writers to justify why selected markers can answer the epidemiological question. The central reading question is not “Is there molecular data?” but “What information can this molecular evidence resolve under these particular conditions?” That question keeps the interpretation proportional to the method.

Different markers evolve at different rates and resolve different levels of relatedness. A marker suitable for broad lineage may be too coarse for recent transmission. A polished diagram can make the problem hard to notice because visual order feels like causal order. Readers should return to the prose and locate the sample frame, measurement rule, time relation and uncertainty that support each edge or cluster.

Connect marker choice to the time scale, pathogen biology and inference sought; state resolution and known limitations. This reporting move gives another team a fair chance to understand the finding and its limits. It also improves classroom discussion: students can point to a missing link in the evidence chain instead of dismissing an entire study with a vague claim that “correlation is not causation”.

6. Molecular clocks and evolutionary assumptions

This link asks writers to explain how sequence change is translated into time and which calibration or rate assumptions are used. The central reading question is not “Is there molecular data?” but “What information can this molecular evidence resolve under these particular conditions?” That question keeps the interpretation proportional to the method.

A date inferred from genetic difference depends on an evolutionary model and uncertainty, not on a literal clock. Rate variation or recombination can distort simple timelines. A polished diagram can make the problem hard to notice because visual order feels like causal order. Readers should return to the prose and locate the sample frame, measurement rule, time relation and uncertainty that support each edge or cluster.

State the model, calibration data, priors or constraints, uncertainty intervals and sensitivity to alternative rates. This reporting move gives another team a fair chance to understand the finding and its limits. It also improves classroom discussion: students can point to a missing link in the evidence chain instead of dismissing an entire study with a vague claim that “correlation is not causation”.

7. Epidemiological time frame

This link asks writers to align sampling dates, symptom or diagnosis dates, exposure windows and the period during which transmission could plausibly occur. The central reading question is not “Is there molecular data?” but “What information can this molecular evidence resolve under these particular conditions?” That question keeps the interpretation proportional to the method.

Two genetically similar samples collected months apart may not support direct transmission if timing, infectious periods or intermediate cases are inconsistent. A polished diagram can make the problem hard to notice because visual order feels like causal order. Readers should return to the prose and locate the sample frame, measurement rule, time relation and uncertainty that support each edge or cluster.

Use an explicit timeline and distinguish specimen date from onset, diagnosis, admission and reporting dates. This reporting move gives another team a fair chance to understand the finding and its limits. It also improves classroom discussion: students can point to a missing link in the evidence chain instead of dismissing an entire study with a vague claim that “correlation is not causation”.

8. Multiple-strain infection

This link asks writers to report how the study detects and handles more than one strain or variant within a person or sample. The central reading question is not “Is there molecular data?” but “What information can this molecular evidence resolve under these particular conditions?” That question keeps the interpretation proportional to the method.

A consensus sequence can hide minority variants. Selecting one colony or dominant signal may make transmission links look simpler than the underlying infection. A polished diagram can make the problem hard to notice because visual order feels like causal order. Readers should return to the prose and locate the sample frame, measurement rule, time relation and uncertainty that support each edge or cluster.

Describe detection limits, within-host sampling and analytic rules for mixed signals, and discuss what may remain unseen. This reporting move gives another team a fair chance to understand the finding and its limits. It also improves classroom discussion: students can point to a missing link in the evidence chain instead of dismissing an entire study with a vague claim that “correlation is not causation”.

9. Non-independence

This link asks writers to recognise that infectious-disease observations can be connected through transmission, households, facilities, outbreaks or shared sources. The central reading question is not “Is there molecular data?” but “What information can this molecular evidence resolve under these particular conditions?” That question keeps the interpretation proportional to the method.

Standard analyses may treat each record as independent even though one infection changes the probability of another. Precision can then look stronger than it is. A polished diagram can make the problem hard to notice because visual order feels like causal order. Readers should return to the prose and locate the sample frame, measurement rule, time relation and uncertainty that support each edge or cluster.

Explain clustering structures and use or justify methods that account for dependence. This reporting move gives another team a fair chance to understand the finding and its limits. It also improves classroom discussion: students can point to a missing link in the evidence chain instead of dismissing an entire study with a vague claim that “correlation is not causation”.

10. Missing molecular and epidemiological data

This link asks writers to separate specimens not collected, laboratory failures, incomplete metadata and records excluded by analysis. The central reading question is not “Is there molecular data?” but “What information can this molecular evidence resolve under these particular conditions?” That question keeps the interpretation proportional to the method.

A complete-case analysis may preferentially retain people with easier access, higher pathogen load or better records. The resulting network can omit important bridges. A polished diagram can make the problem hard to notice because visual order feels like causal order. Readers should return to the prose and locate the sample frame, measurement rule, time relation and uncertainty that support each edge or cluster.

Report missingness by stage and group, compare included with excluded observations, and explain any imputation or sensitivity analysis. This reporting move gives another team a fair chance to understand the finding and its limits. It also improves classroom discussion: students can point to a missing link in the evidence chain instead of dismissing an entire study with a vague claim that “correlation is not causation”.

11. Ascertainment bias

This link asks writers to describe how testing, surveillance, healthcare access and case definitions determine which infections become visible. The central reading question is not “Is there molecular data?” but “What information can this molecular evidence resolve under these particular conditions?” That question keeps the interpretation proportional to the method.

An apparent cluster may reflect intensified testing in one setting, while silent transmission elsewhere remains unobserved. Molecular detail cannot repair a biased route into the dataset. A polished diagram can make the problem hard to notice because visual order feels like causal order. Readers should return to the prose and locate the sample frame, measurement rule, time relation and uncertainty that support each edge or cluster.

Trace the path from infection to detection and discuss how surveillance intensity affects observed patterns. This reporting move gives another team a fair chance to understand the finding and its limits. It also improves classroom discussion: students can point to a missing link in the evidence chain instead of dismissing an entire study with a vague claim that “correlation is not causation”.

12. Consistency between molecular and epidemiological evidence

This link asks writers to report where genetic patterns, contact histories, place and time agree—and where they do not. The central reading question is not “Is there molecular data?” but “What information can this molecular evidence resolve under these particular conditions?” That question keeps the interpretation proportional to the method.

It is tempting to privilege whichever evidence creates the cleanest story. Yet disagreement may reveal missing intermediates, laboratory error, incorrect recall, convergent change or an unsuitable threshold. A polished diagram can make the problem hard to notice because visual order feels like causal order. Readers should return to the prose and locate the sample frame, measurement rule, time relation and uncertainty that support each edge or cluster.

Use a comparison table and discuss competing explanations rather than forcing every case into one narrative. This reporting move gives another team a fair chance to understand the finding and its limits. It also improves classroom discussion: students can point to a missing link in the evidence chain instead of dismissing an entire study with a vague claim that “correlation is not causation”.

13. Ethics, privacy and identifiability

This link asks writers to address consent, governance, data sharing and the possibility that combined molecular and contextual data identify people or communities. The central reading question is not “Is there molecular data?” but “What information can this molecular evidence resolve under these particular conditions?” That question keeps the interpretation proportional to the method.

Pathogen sequences may seem anonymous, but rare variants, dates, locations and network links can narrow identity. Transmission language can also imply blame. A polished diagram can make the problem hard to notice because visual order feels like causal order. Readers should return to the prose and locate the sample frame, measurement rule, time relation and uncertainty that support each edge or cluster.

Describe approvals, consent or waivers, access controls, de-identification limits, community risks and careful communication choices. This reporting move gives another team a fair chance to understand the finding and its limits. It also improves classroom discussion: students can point to a missing link in the evidence chain instead of dismissing an entire study with a vague claim that “correlation is not causation”.

Worked case: interpreting a cluster without inventing a transmission chain

Imagine that six samples collected from people connected with the same venue are placed in one genetic cluster. A dramatic account might say, “One person infected the other five at the venue.” STROME-ID encourages a slower reconstruction. Were all eligible cases sampled? How much pathogen diversity existed in the wider community? Which marker and threshold defined the cluster? Were collection and onset dates compatible with a single event?

Suppose the laboratory method cannot distinguish samples differing below a certain threshold, and similar samples were circulating widely. The cluster may support a common chain of transmission but cannot locate it. If only people who reported the venue were invited for sequencing, ascertainment can strengthen the appearance of a venue-specific cluster. If people without the venue link were less likely to be tested, the comparison group is incomplete.

Now add contact data. Four people attended on the same evening, one attended two weeks later and one worked nearby but reported no direct contact. Molecular and epidemiological evidence partly agree and partly disagree. The mismatch deserves analysis. It might reflect an unsampled intermediary, inaccurate recall, prolonged circulation or a cluster definition that is too broad.

An honest conclusion could read: “The six analysed samples were genetically similar under the stated threshold. Timing and venue information supported a possible shared transmission network for four cases, while the later and indirect cases indicate that direct venue-based transmission cannot be established from the available evidence.” That sentence is not timid; it is informative.

Did You Know? More detailed data can increase privacy risk

A student evidence lab with eight practical tasks

1. Build a term ledger

Choose five technical words from a molecular-epidemiology article. Copy the author’s definition or operational rule, then write what a casual reader might wrongly assume. Finish with a sentence that uses the term precisely.

Reflection prompt: identify the exact noun, verb or qualifier that most improved the accuracy of your final explanation. This makes language revision visible as evidence work rather than cosmetic editing.

2. Reconstruct the sample funnel

Draw eligible cases, tested cases, specimens obtained, laboratory successes, sequences passing quality control and observations included in analysis. Add numbers and reasons for loss at each stage. Discuss who may be systematically missing.

3. Separate four dates

Create columns for exposure, symptom onset, diagnosis and specimen collection. If a paper does not provide one, mark it absent rather than guessing. Explain how substituting one date for another could change the transmission story.

4. Test the marker claim

Write the epidemiological question, the molecular marker used and the resolution required. Ask whether the paper explains why the marker is suitable. If not, identify the inference that becomes uncertain.

5. Audit a network diagram

For every edge in a published network, label the evidence: genetic similarity, documented contact, shared place, temporal compatibility or model inference. An edge with several labels is not automatically direct transmission; it is a transparent bundle of evidence.

6. Write the disagreement paragraph

Find one place where molecular and epidemiological evidence do not align. Give at least three possible explanations and one additional observation that would help distinguish them.

7. Run a privacy thought experiment

List the variables a public figure contains. Imagine combining them with local knowledge. Remove or aggregate details until the communication retains scientific value without making identification or blame unnecessarily likely.

8. Scale the headline

Write a sensational headline, a technically accurate headline and a family-friendly accurate headline for the same result. Compare which words changed and whether any version exceeded the evidence.

Advice for students, parents and teachers

Students do not need advanced genetics to practise the reasoning. Begin with a simplified tree or cluster diagram and a short table of dates and locations. Ask what the diagram directly shows, what the table directly shows and what conclusion needs both. The key is to keep layers separate before combining them.

Parents can ask gentle questions: “Could there be an unsampled person?”, “Does similar mean identical?”, “Which date is this?”, and “What else could produce the same pattern?” These questions model curiosity, not distrust. A student who can name an alternative explanation is ready to ask what evidence would make one explanation more plausible.

Teachers can use role cards: laboratory scientist, epidemiologist, statistician, community representative and journalist. Each role notices different reporting needs. The exercise shows why interdisciplinary work requires shared English. Precision allows experts to disagree productively without speaking past one another.

Other languages remain essential. Community terms, local histories and non-English sources may clarify exposure and meaning that a standardised report overlooks. English can support international exchange, but responsible research should not treat it as the only place where relevant knowledge lives.

Common overclaims and how to repair them

“The sequence proves who infected whom.” Repair: state similarity and explain what direction-of-transmission evidence is missing.
“All outbreak cases were sequenced.” Repair: show the eligible-to-analysed funnel and define “case”.
“The strains were identical.” Repair: name the method, genome coverage and threshold.
“The molecular clock dated the infection.” Repair: report the modelled time estimate and uncertainty, and distinguish infection from lineage divergence.
“Missing samples were random.” Repair: describe reasons for missingness and test plausible patterns.
“The cluster was caused by the venue.” Repair: separate shared place, temporal overlap, molecular relatedness and alternative sources.
“Anonymous data created no ethical risk.” Repair: discuss re-identification, group harms, consent, governance and communication.

Frequently asked questions

What does STROME-ID stand for?

It stands for Strengthening the Reporting of Molecular Epidemiology for Infectious Diseases. It extends STROBE for observational studies that use infectious-disease molecular data.

How many items did it add?

The 2014 statement reports that 20 items were added to the 22-item STROBE checklist, focusing on threats and reporting needs specific to molecular infectious-disease epidemiology.

Does genetic similarity prove direct transmission?

Usually not by itself. Similarity may be compatible with direct transmission, an unsampled intermediary, a shared source or limited background diversity. Time, place, sampling and epidemiological links matter.

Why report pathogen diversity?

The meaning of a match depends partly on how common similar molecular profiles are in the wider population. Low diversity can make many unrelated chains appear similar under a coarse method.

What is a molecular clock?

It is a model that relates molecular change to time using evolutionary assumptions and calibration. Its output has uncertainty and may be affected by rate variation, recombination and model choice.

Why are infectious-disease observations non-independent?

One case can change the risk of another through transmission, and cases may share households, facilities or sources. Analyses and uncertainty estimates need to recognise these connections.

What is ascertainment bias?

It occurs when the route by which infections are detected and included differs across people, places or times. Testing access, surveillance intensity and case definitions shape the visible dataset.

Why discuss multiple-strain infection?

A person or specimen can contain more than one strain or variant. Methods that record only a dominant signal may hide diversity and simplify transmission patterns.

Does the checklist decide whether a study is ethically acceptable?

No. It prompts transparent ethical reporting, but formal ethical review, consent, governance, privacy protection and community engagement require their own processes.

Can secondary students read these papers?

Yes, selectively. Start with the abstract, sampling description, definitions, main figure, limitations and ethics. Use the evidence-chain questions rather than trying to master every laboratory detail at once.

Useful next reading

Use the EQUATOR STROME-ID page, the open guideline text and the PubMed record as primary starting points. For broader research habits, read sources, attribution and verification, checking source credibility before using evidence, and the related discussion of STROBE-Vet, clustering and generalisability.

The happy conclusion is that complexity can be made navigable. When English keeps samples, methods, time, diversity, bias and ethics visible, molecular evidence becomes easier to question and safer to use. That is not a barrier to discovery; it is how discovery becomes shareable knowledge.