If you wanted to know who can find reliable health information online, where would you put the questionnaire? In a Bulgarian study, 546 of the 933 analysable responses came in on paper. The findings offer a portrait of people’s confidence and an illustration of how the route into a survey shapes what we can learn.
- Half of respondents fell in the high category on a scale of perceived digital health skills. The threshold was 30 out of 40 points.
- The researchers recruited through participant networks and offered identical questions online and on paper. Paper supplied 58.5% of the final sample.
- Respondents with lower educational attainment or lower income generally reported lower scores. Recruitment and self-report limit generalisation from this group.
- Publication date
- 22 November 2026
- Respondents analysed
- 933 adults aged 20–79
- Survey formats
- 387 online and 546 on paper
- Score range
- 8 to 40 points
Original publication: 22 Nov 2026 · The date above refers to this brief.
How respondents rated their digital health skills
percent of respondents
From an answer to a confidence category
Choose a survey format
Adults could complete the same questions online or on paper.
Add eight ratings
Each answer contributes between one and five points.
Read the threshold
A total of 30 or more enters the high self-rating category.
What counted as a high score?
The Bulgarian questionnaire asks people how well they believe they can find, evaluate and use electronic health information. Eight items each contribute one to five points. A total of 30 enters the high category, equivalent to an average of 3.75 points per item. The researchers report rounded shares of 14% low, 36% moderate and 50% high. A high score records confidence across those items. The study did not ask respondents to open two websites and identify which had the more reliable evidence.
Source 1 ↗A paper form changed who could answer
The team collected 959 questionnaires through a network-based recruitment process. After excluding 26 without valid consent, it analysed 387 online and 546 paper responses. Paper therefore accounted for 58.5% of the final sample and online for 41.5%. Someone could join without completing a web form, an important feature in a study about using the internet. Because people chose their format, the totals alone cannot tell us whether answering on paper changed anyone’s score.
Source 1 ↗Who is represented in the result?
The study found differences across age, education and income groups. People with lower education or lower income generally reported less confidence. People with higher education made up 57% of respondents, and women made up 64%. Network recruitment does not give every Bulgarian adult a known chance of selection. Even an accurately calculated percentage for this sample therefore has limits as a nationwide estimate.
Source 1 ↗How to test the skill behind the answer
An earlier Dutch study paired the eHealth Literacy Scale with nine observed internet assignments. In its performance sample of 88 adults, the correlation between the rating and successful completion was 0.18. Different language, tasks and internet tools prevent transferring that value directly to Bulgaria. The design does suggest a concrete follow-up. Ask participants to find the source behind a medical claim, record whether they succeed and compare that result with the confidence they reported beforehand.
Source 1 ↗Source 2 ↗Track the survey route and the kind of skill measured
Publisher article. Checked recruitment and exclusions, the eight-item score and thresholds, Figures 1 and 2, and the socioeconomic comparisons. No individual survey responses were available for this check.
01What does the high category measure?
- What was observed
- A total of 30–40 points on eight self-ratings of digital health skills.
- Where the conclusion stops
- No directly observed search task was part of the Bulgarian outcome.
Source 1 · Methods on assessment and Results, Figure 2
02Who supplied the survey responses?
- What was observed
- The analysis contains 387 online and 546 paper questionnaires.
- Where the conclusion stops
- Participants selected their preferred format and entered through participant networks.
Source 1 · Methods on recruitment and inclusion
Numbers you can inspect
| Measure | Value & unit | Origin & method |
|---|---|---|
| Questionnaires received | 959 questionnaires | Reported All submitted questionnaires before the consent check. Source 1 · Methods, recruitment |
| Analysed responses | 933 respondents | Reported Final group with valid informed consent. Source 1 · Methods, recruitment |
| Paper forms included | 546 responses | Reported Final analysis by format. Source 1 · Methods, recruitment |
| Online forms included | 387 responses | Reported Final analysis by format. Source 1 · Methods, recruitment |
| Share of final sample on paper | 58.5 percent | Calculated 546 divided by 933 times 100, rounded to one decimal. Source 1 · Methods, recruitment |
| Share of final sample online | 41.5 percent | Calculated 387 divided by 933 times 100, rounded to one decimal. Source 1 · Methods, recruitment |
| Low category | 14 percent | Reported Rounded proportion reported for 8–18 points. Source 1 · Results, Figure 2 |
| Moderate category | 36 percent | Reported Rounded proportion reported for 19–29 points. Source 1 · Results, Figure 2 |
| High category | 50 percent | Reported Rounded proportion reported for 30–40 points. Source 1 · Results, Figure 2 |
| Average item score at the high cut point | 3.75 points per item | Calculated The cut point of 30 divided by eight items. Source 1 · Methods, assessment |
| Rating and task completion in the Dutch study | 0.18 correlation | Reported Separate Dutch performance sample. The reported p-value was 0.09. Source 2 · Dutch validation, Results |
Compare the actual experiments
These studies answer different questions. Read the unit and endpoint before comparing results.
| Study | Unit & setting | Readout | Interpretation boundary |
|---|---|---|---|
| Dutch eHEALS validation Source 2 · Methods, Study 2 and Results | Adults who rated skills and completed observed tasks | Association between a scale and task completion | The older Dutch estimate cannot identify performance in the Bulgarian respondents. |
| Bulgarian mixed-format survey Source 1 · Methods, recruitment and assessment | Adults choosing a paper or online questionnaire | A common self-report score in a volunteer sample | Access to a form and success on a search task are distinct measurements. |
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.
What would count as evidence for each question?
| Question | Available measurement | What it can show |
|---|---|---|
| How confident did respondents feel? | Eight self-report items, summed to 8–40 points | The share in the high category was 50% in this sample. |
| How did people take part? | 546 paper and 387 online responses in the analysis | Both routes contributed substantially to the final group. |
| Can participants identify a reliable source? | Observed assignments in an earlier Dutch study | A task can directly measure success. A local comparison would require Bulgarian participants and relevant tasks. |
The first two rows use the Bulgarian study. The last describes the kind of evidence the Dutch team collected.
Your questions, answered
Did half the participants pass a test of online searching?
Half reached the high category on a self-rating scale. An observed search task would provide a separate measure of performance.
Why offer paper in a study of online health information?
It allows participation without submitting an online form. Paper responses supplied 546 of the 933 questionnaires analysed.
Can the study estimate the score of the average Bulgarian adult?
It describes the volunteers who took part. Recruitment through networks leaves uncertainty about how closely the sample resembles all Bulgarian adults.
What does the Dutch correlation add?
It shows how one study compared self-report with directly observed assignments. The strength of that relationship in Bulgaria remains unknown.
Limits of this interpretation
- Participants were recruited through networks and chose their survey format.
- The instrument captures perceived ability, and the score bands depend on the selected thresholds.
- The Dutch performance study used a different population and period. Its correlation provides context for the measurement question.
Sources & transparency
- Pesheva and Grigorov (2026) — Digital Health Literacy Among the Bulgarian Population: Socioeconomic Factors Associated with Its Level
Publisher article. Checked recruitment and exclusions, the eight-item score and thresholds, Figures 1 and 2, and the socioeconomic comparisons. No individual survey responses were available for this check. · Accessed 22 Nov 2026
DOI: 10.3390/healthcare14234016 - van der Vaart et al. (2011) — Validation of a Dutch version of the eHealth Literacy Scale in two adult populations
Publisher article. Checked the second sample, its observed internet assignments and the association between task completion and self-rated eHealth literacy. · Accessed 22 Nov 2026
DOI: 10.2196/jmir.1840
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