U.S. adult survey · 2024 evidence
Who shares health information on social media?
Sharing a health article and sharing personal health information are separate survey activities.
Among adults aged 18–34, 24.1% reported sharing personal health information on social media and 39.2% reported sharing general health-related information at some frequency in the past 12 months. Among adults aged 75+, the estimates were 7.9% and 14.4%, respectively.
United States · Civilian noninstitutionalized adults 18+ · Percent within each recorded age band · Includes less-than-monthly sharing
Social Usage analysis of NCI HINTS 7 public-use data, B12b/c and AgeGrpB. Fieldwork March 25–September 16, 2024. Standard errors for ages 18–34: 2.12 and 2.16 percentage points; ages 75+: 1.86 and 2.27. These activities can overlap; the shares must not be added. NCI does not endorse this presentation.
Personal and general sharing across five age groups
Source: NCI HINTS 7, B12b / SocMed_SharedPers and AgeGrpB. Missing sharing answers remain in each denominator. Both charts use a 0–100% scale. Accessible table and sampling errors.
Source: NCI HINTS 7, B12c / SocMed_SharedGen and AgeGrpB. Missing sharing answers remain in each denominator. Both charts use a 0–100% scale. Accessible table and sampling errors.
General health-information sharing has a higher point estimate than personal sharing in each age band. Both activities have lower point estimates in the older bands. These descriptive patterns do not show that aging causes people to share less, and no significance test was performed.
The survey asks separate questions of the same participants. Someone may do both activities, either one, or neither. Adding the bars could double-count participants; subtracting them would not identify people who share general information only.
View all ten estimates, sampling errors and sample counts
| Activity | Age | Share | SE (pp) | RSE (%) | Sample n | Sharing n |
|---|---|---|---|---|---|---|
| Personal | 18–34 | 24.1% | 2.12 | 8.8 | 1,080 | 293 |
| Personal | 35–49 | 16.1% | 1.50 | 9.3 | 1,355 | 262 |
| Personal | 50–64 | 13.3% | 1.13 | 8.5 | 1,728 | 261 |
| Personal | 65–74 | 10.3% | 1.44 | 14.0 | 1,466 | 151 |
| Personal | 75+ | 7.9% | 1.86 | 23.4 | 1,036 | 82 |
| General | 18–34 | 39.2% | 2.16 | 5.5 | 1,080 | 444 |
| General | 35–49 | 32.7% | 2.10 | 6.4 | 1,355 | 481 |
| General | 50–64 | 27.6% | 1.88 | 6.8 | 1,728 | 501 |
| General | 65–74 | 22.2% | 1.48 | 6.7 | 1,466 | 339 |
| General | 75+ | 14.4% | 2.27 | 15.8 | 1,036 | 156 |
Counts are unweighted respondents, not population totals. Every age-domain denominator includes missing/invalid sharing answers. The personal-sharing estimate for ages 75+ has the largest relative standard error, 23.4%; smaller point estimates can have greater relative uncertainty.
What these questions do—and do not—measure
B12b asks about sharing personal health information. B12c asks about sharing general health-related information, with a news article as an example. Both refer to social media in the past 12 months. We combine “Almost everyday,” “At least once a week,” “A few times a month” and “Less than once a month.” This is any reported frequency, not regular or daily sharing.
The questions do not identify a named platform, a public post versus a private group, whose personal health information was shared, or whether the information was accurate. They do not measure private messaging generally, peer support received, diagnoses or treatment outcomes. Discussion with people who have similar health issues is a separate HINTS item covered in the health-information guide.
These percentages use all adults in each recorded age band, not just social-media users. “Never” is not the exact complement of the displayed estimate because item nonresponse remains in the denominator.
Methods, missingness and original dates
How we calculated age estimates and uncertainty
HINTS 7 used an address-based probability sample, selected one adult using the next-birthday method, and collected paper and web responses. We use the original final person weights and the five published AgeGrpB categories. Each percentage is 100 × weighted respondents selecting codes 1–4 ÷ weighted respondents in that age band.
We recompute both numerator and denominator for each of 50 replicate weights. The JK1 standard error is the square root of 0.98 × the sum of squared differences between each replicate percentage and the full-sample percentage. These age estimates and SEs are Social Usage calculations; no confidence interval or significance test is supplied.
Age was unclassified for 613 participants, representing 7.3% of the weighted sample. They are excluded from age domains. We use no imputation or additional age-nonresponse adjustment, so the subgroup results depend on recorded age. Missing sharing responses stay in each age denominator.
All ten cells meet our safeguards: numerator and complement counts of at least 50, domain counts of at least 100, and relative SE at most 30%. These are product safeguards, not thresholds issued by NCI. Source references: questionnaire PDF p5/printed p4, codebook PDF pp81–82/printed pp80–81 and p524/printed p523, History Document pp2,8,14 in the original package.
Source version, review and reuse
- Data collection
- March 25–September 16, 2024
- Reference window
- Past 12 months at each response
- Public-use package
- HINTS7_STATA_20250731; updated August 2025, exact publication day unknown
- Original source retrieval
- October 1, 2026
- Retained source-rights check
- October 7, 2026
- Assisted analysis review
- 2026-10-08
The national calculations reproduce our previously reviewed sharing estimates; age-domain counts reproduce the codebook. Reusing this release does not refresh the survey. Assisted factual and technical review is not independent human editorial approval.
HINTS public-use guidance and the NCI reuse policy support attributed aggregate analysis with confidentiality and precision safeguards. This report provides no data download, image export or embed. Respondent records, source graphics and logos are not republished.
Cite this finding
Social Usage analysis of National Cancer Institute, HINTS 7 (2024). No NCI endorsement implied. Past-12-month sharing on social media, any frequency (codes 1-4). Personal health information: 18-34 years 24.1% (SE 2.12 percentage points); 35-49 years 16.1% (SE 1.50 percentage points); 50-64 years 13.3% (SE 1.13 percentage points); 65-74 years 10.3% (SE 1.44 percentage points); 75+ years 7.9% (SE 1.86 percentage points). General health-related information: 18-34 years 39.2% (SE 2.16 percentage points); 35-49 years 32.7% (SE 2.10 percentage points); 50-64 years 27.6% (SE 1.88 percentage points); 65-74 years 22.2% (SE 1.48 percentage points); 75+ years 14.4% (SE 2.27 percentage points). U.S. adults in recorded AgeGrpB bands; missing frequency answers retained. Age unclassified for 7.3% of weighted participants, excluded from age domains. Fieldwork March 25–September 16, 2024; package updated August 2025, retrieved October 1, 2026. B12b/c; codebook PDF pp81-82 and524; History Document p2; 50-replicate JK1 SEs, no CI or significance test. No added audiences or inferred joint sharing. https://hints.cancer.gov/dataset/HINTS7_STATA_20250731.zip Assisted review 2026-10-08. https://socialusage.com/reports/us-health-information-sharing/