Two participants can report similar symptoms and describe different limits in their daily lives. A German analysis followed people who had been hospitalised with COVID-19 and collected both kinds of information. The results provide a concrete example of why the denominator belongs alongside every percentage, especially when some follow-up data are missing.

THE SHORT READ
  • The paper’s Results list 473 people with a calculable symptom score at three months. Of them, 361 met the study’s threshold for at least mild post-COVID condition.
  • The 156 people with both symptoms at that level and participation limits made up 43.2% of the 361 and 33.0% of the 473.
  • Figure 3 shows that later status could be missing. In one severe subgroup of 29 people with early participation limits, 10 lacked a later participation result.
THE STUDY AT A GLANCEPost-COVID symptoms and daily participation in two German cohorts · Nürnberger et al.
Paper date
21 November 2026
Setting
Two German cohorts of hospitalised adults
Results denominator at three months
473 with computable symptom scores
Joint count at three months
156 with both indicators

Original publication: 21 Nov 2026 · The date above refers to this brief.

THE NUMBERS, IN CONTEXT

Later participation status for 29 people with early limits and severe symptoms

people

Documented limits at 12 months16
Documented no limits3
Status missing10
029
Counts from one branch of Figure 3D. All 29 had severe symptoms and participation limits at the three-month assessment.
THE EXPERIMENT, EXPLAINED

Find the people behind the three-month percentages

  1. 473 scoreable assessments

    A symptom score was computable at three months for this group.

  2. 361 at least mild

    This subgroup reached the study’s symptom threshold.

  3. 156 with both measures

    These people also met the participation limitation threshold.

Three-month Results counts. Each lower count is nested within the one above.

The questions covered symptoms and participation

The authors combined follow-up data from two cohorts and selected adults who had been hospitalised during the acute illness. The symptom score draws on the symptoms reported and how long they lasted. A separate set of questions asks about activities with family and friends, leisure and usual work, including work at home. The distinction matters because a symptom count can describe illness burden while leaving the extent of everyday disruption uncertain.

Source 1 ↗Source 2 ↗

Follow the numerator through two denominators

At the three-month assessment, the Results report 473 people with enough information for a symptom score. Of those, 361 scored at least mild on the study scale. Among those 361, 156 also reported limitations that met the study’s participation threshold. The first calculation, 156 divided by 361, asks how often participation limits appeared among people who met the symptom threshold. The second, 156 divided by 473, asks how often both measures were present among people with a calculable symptom score. Both use the same 156 people and the same visit.

Source 1 ↗

Read missing follow-up as an outcome of data collection

Figure 3 follows groups from three to 12 months. The panel for severe symptoms contains a branch for 29 people who had participation limits at three months. Of those, 16 had a recorded limit at 12 months, three recorded no limit and ten had missing data. Among the 19 with a known later answer, 16 of 19 is 84.2%. Across the original branch, 16 of 29 is the 55.2% with documented limitations at the second visit. The ten missing results explain the gap between the two calculations.

Source 1 ↗

What a stronger follow-up could report

A study could list how many participants remained eligible, completed each questionnaire and had an evaluable answer at every time point. It could also compare baseline information for people with and without missing follow-up. Those details would give readers a better basis for judging changes over time. The cohort results apply to a selected group of adults who were hospitalised, and the participation measure describes people’s own reports for the activities asked about.

Source 1 ↗
TRACE THE EVIDENCE

Keep missing data in view when reading percentages

Publisher full article, including the inclusion flowchart, Results and Figure 3. The abstract and body give different headline enrolment counts. The three-month percentages in this article use the explicitly counted Results subgroups.

01What changes between 43.2% and 33.0%?

What was observed
Both use 156 people, divided by 361 or 473.
Where the conclusion stops
The denominators define separate questions at the same visit.

Source 1 · Results, Post-COVID condition and activity and participation impairment

02What happened to the severe subgroup at the later visit?

What was observed
The branch lists 16 with participation limits, 3 without and 10 missing at 12 months.
Where the conclusion stops
The chart can separate documented status from unobserved status, which matters for interpreting apparent improvement.

Source 1 · Figure 3D

Numbers you can inspect

MeasureValue & unitOrigin & method
Computable symptom scores at three months473 peopleReported Results count for the scoreable group.
Source 1 · Results, Post-COVID condition and activity and participation impairment
At least mild on the symptom score361 peopleReported Subset of the 473 with a score.
Source 1 · Results, Post-COVID condition and activity and participation impairment
Both symptom and participation thresholds156 peopleReported Joint count at the three-month visit.
Source 1 · Results, Post-COVID condition and activity and participation impairment
Joint count as share of people with at least mild symptoms43.2 percentCalculated Divide 156 by 361 and multiply by 100.
Source 1 · Results, Post-COVID condition and activity and participation impairment
Joint count as share of the scoreable group33 percentCalculated Divide 156 by 473 and multiply by 100.
Source 1 · Results, Post-COVID condition and activity and participation impairment
Prior severe subgroup with a later report of limits16 peopleReported Documented participation limits at 12 months.
Source 1 · Figure 3D
Prior severe subgroup with a later report of no limits3 peopleReported Documented no participation limits at 12 months.
Source 1 · Figure 3D
Prior severe subgroup with later status missing10 peopleReported Missing 12-month participation status.
Source 1 · Figure 3D
Share reporting a limit among those observed later84.2 percentCalculated Divide 16 by the 19 with later status, then multiply by 100.
Source 1 · Figure 3D
Share with a documented later limit among the original branch55.2 percentCalculated Divide 16 by all 29 in the original branch, then multiply by 100.
Source 1 · Figure 3D

Compare the actual experiments

These studies answer different questions. Read the unit and endpoint before comparing results.

StudyUnit & settingReadoutInterpretation boundary
Earlier symptom score study and the new participation measure

Source 2 · Appel et al., Abstract and Results

Adults assessed at follow-up in NAPKON cohortsSymptom severity and difficulty in daily rolesWork, family and leisure questions add detail beyond the number and duration of symptoms.
Known later status and missing later status

Source 1 · Figure 3D

Severe-symptom group with participation limits at three monthsParticipation status at the 12-month visitCarry the ten missing observations when reading any percentage from this subgroup.
Download evidence table (CSV)

The export includes claims, available numbers, methods and source locations. It contains our reading notes and published summaries; it is not raw participant data or an independent reanalysis.

Evidence update · 22 Nov 2026
First publication. Any calculations in this brief use reported figures. The reading notes identify possible follow-up questions. AI source check; no human editorial or clinical review.

CONNECT THE EVIDENCE

Same joint count, different questions

GroupCalculated share or countQuestion answered
Among people meeting the symptom threshold156 ÷ 361 = 43.2%How often did this group also report limits in participation?
Among people with a computable symptom score156 ÷ 473 = 33.0%How often were both criteria documented in the scoreable group?
Later status in one severe subgroup16 with limits, 3 without, 10 missingTwenty-nine had limits at three months. Figure 3D tracks this branch.

Press-News calculations from the Results. The last row refers to one Figure 3D branch and a later visit.

READER QUESTIONS

Your questions, answered

Why are both 43.2% and 33.0% correct?

Both calculations start with 156 people. They divide by different groups, either 361 with at least mild symptoms or 473 with a computable symptom score.

What did the participation questions cover?

They asked about usual work, leisure and activities with family and friends, including work at home.

How did the study record people without a later answer?

The trajectory figure has a separate category for missing later data. The chart below uses that category in the severe subgroup.

Can the percentages describe everyone after COVID-19?

They describe analysed adult hospital survivors in the selected German cohorts. A community estimate would need a sample drawn from that target population.

LIMITATIONS

Limits of this interpretation

  • The inclusion flowchart reports 654 pooled participants, while the abstract gives 655 in its three-month summary. The paired percentages here use the 473, 361 and 156 counts reported together in the Results.
  • The sample comes from two German cohorts and excludes people with outpatient treatment alone.
  • Some patients had missing later participation data. Descriptive results and observed follow-up counts depend on who answered.
SOURCE NOTES

Sources & transparency

  1. Nürnberger, Steinmetz, Appel et al. (2026) — Function, activity and participation limitations in patients with post-COVID condition

    Publisher full article, including the inclusion flowchart, Results and Figure 3. The abstract and body give different headline enrolment counts. The three-month percentages in this article use the explicitly counted Results subgroups. · Accessed 22 Nov 2026

    DOI: 10.1038/s41598-026-81336-w
  2. Appel et al. (2024) — Definition of post-COVID syndrome using a symptom-based post-COVID score

    Author abstract of the earlier cohort study. Checked how the symptom score was applied and related to reported quality of life. · Accessed 22 Nov 2026

    DOI: 10.1007/s15010-024-02226-9

Prepared and source-checked with AI for Press-news Team. The source notes identify the material available during preparation. The article received no human editorial or clinical review. The source list supports the reported results and the calculations shown.

Source check: AI source check — primary research and stated source limits

Clinical review: Not applicable to this educational guide

Suggest a correction