Source. Every figure on this page is computed from the study records in our library when the page loads. It uses the number of participants each record states, and the study design the score reads from the record and from PubMed's publication types. 9,360 of the 27,380 records (34.2%) carry a participant count; the rest are not counted here.
The count is whatever the summary pipeline extracted from the paper. For a meta-analysis it is usually the pooled total; for a registry study it may be the whole registry. It is a size, not a quality.
- 9,360studies with a participant count, of 27,380
- 254median participants
- 66–1,324.8interquartile range
- 3,031studies with 100 or fewer people (32.4%)
How studies are distributed by size
Number of studies in each size band, out of the 9,360 with a recorded count. The bands are unequal on purpose: they follow the orders of magnitude the field actually works at.
Size by study design
Median participant count for each design group with at least 20 studies, with the interquartile range under each bar. A study is grouped by the higher-ranked of its recorded design and its PubMed publication type, the same rule the score uses.
The order is the point. Reviews and meta-analyses pool other studies, so their counts are the largest. Cohort studies follow thousands of people for years. Randomised trials, the design that can show cause and effect, are in the middle of the range: a trial has to recruit, treat and follow every participant, which is expensive, so most are far smaller than the observational studies around them. The smallest medians belong to cross-sectional and case-control designs and to studies with no design recorded, a group that includes much of the laboratory and imaging work.
Interpretation: a large sample is not the same as strong evidence. The library's largest counts come from pooled and registry studies, which can only show association. Its most decisive single results, the randomised trials, are mid-sized by this table.
| Design group | Studies with a count | Median | Lower quartile | Upper quartile |
|---|---|---|---|---|
| Meta-analysis or systematic review | 631 | 1,687 | 193 | 11,134 |
| Cohort or longitudinal study | 1,952 | 619.5 | 154.2 | 3,371.2 |
| Cross-sectional study | 820 | 294 | 97.25 | 1,095 |
| Case-control study | 168 | 258.5 | 92 | 1,348.8 |
| No design recorded | 4,926 | 167 | 47 | 819.5 |
| Randomised or phase 2/3 trial | 861 | 147 | 53.5 | 462.5 |
Size by subject
Subjects with at least 20 studies carrying a count, ordered by median. The second column shows how many of the subject's studies record a count at all.
| Subject | With a count / in subject | Median | Lower quartile | Upper quartile |
|---|---|---|---|---|
| Prevention and risk | 366 / 1,004 | 2,006 | 362.5 | 26,698.2 |
| Costs and economics | 75 / 199 | 1,620 | 417 | 25,109 |
| Long-term cohort studies | 366 / 542 | 1,559 | 276.5 | 8,312.2 |
| How common it is | 273 / 522 | 1,278 | 214 | 7,907.5 |
| Disparities | 384 / 833 | 1,165 | 262.2 | 5,706 |
| Health systems and care delivery | 230 / 654 | 956.5 | 99.25 | 12,509.5 |
| Risk factors | 733 / 2,174 | 645 | 147 | 5,461.5 |
| Drug development | 106 / 873 | 570 | 59 | 12,701.2 |
| Genetics | 514 / 1,615 | 497 | 125 | 2,311.5 |
| Vascular contributions | 190 / 546 | 427.5 | 121.2 | 1,349.2 |
| Laboratory science | 20 / 46 | 403 | 167.2 | 748.5 |
| TREM2 and immune genetics | 65 / 701 | 303 | 42 | 1,003.5 |
| Biomarkers | 871 / 1,798 | 233 | 84 | 645 |
| Getting a diagnosis | 388 / 815 | 204 | 66 | 1,049.8 |
| Digital health tools | 400 / 1,094 | 196.5 | 48.25 | 1,953.8 |
| Prevention trials | 322 / 835 | 192 | 52.5 | 1,201 |
| Caregiving | 311 / 833 | 191 | 54 | 727 |
| Treatments | 583 / 1,489 | 189 | 63 | 726 |
| Clinical safety | 270 / 630 | 185.5 | 72.25 | 821 |
| Brain tissue and pathology | 152 / 612 | 169 | 49 | 1,099.2 |
| End-of-life care | 315 / 855 | 164 | 43 | 1,350 |
| Mild cognitive impairment | 297 / 701 | 163 | 48.5 | 512.5 |
| Cognitive tests and assessment | 481 / 881 | 130 | 53 | 495 |
| Clinical trials | 87 / 162 | 119 | 48 | 309 |
| Tau biomarkers | 345 / 690 | 111 | 44 | 355 |
| Lewy body dementia | 115 / 566 | 107 | 40 | 394 |
| Disease mechanisms | 100 / 1,380 | 101.5 | 27.5 | 418.2 |
| Brain scans and imaging | 543 / 1,459 | 94 | 35 | 300 |
| Down syndrome and Alzheimer's | 137 / 664 | 72 | 26 | 227 |
| Risk genes beyond APOE | 69 / 847 | 67 | 30 | 241 |
| Inflammation and the immune system | 70 / 868 | 60 | 25.75 | 418.5 |
| Hippocampus size on brain scans | 182 / 492 | 52.5 | 23 | 187.2 |
Scroll the table sideways to see every column.
Size by evidence rating
Our summaries rate each study's evidence. The rating is made from design and reporting, not from size, so this table asks whether size tracks it anyway.
| Rated in our summary | Studies with a count | Median | Lower quartile | Upper quartile |
|---|---|---|---|---|
| high | 1,482 | 272 | 67 | 1,640.5 |
| moderate | 3,332 | 486 | 125 | 2,220 |
| low | 4,486 | 156 | 45 | 769 |
| null-result | 23 | 598 | 204 | 1,021 |
| early | 30 | 270 | 75.75 | 1,959.2 |
| preliminary | 7 | 68 | 19 | 7,496 |
The ten largest counts in the library
As recorded in the file. Several are pooled analyses or national registries, where the count is the number of records analysed rather than people recruited.
- Association between vaccinations and risk of dementia: a systematic review and meta-analysis
- Long-term air pollution exposure and incident dementia: a systematic review and meta-analysis
- Risk of Dementia in Individuals With Emergency Department Visits or Hospitalizations Due to Cannabis
- Time to nursing home admission and death in people with dementia: systematic review and meta-analysis
- Domain-specific physical activity levels and risk of dementia: a systematic review and meta-analysis
- Changes in Alcohol Consumption and Risk of Dementia in a Nationwide Cohort in South Korea
- Association of Metformin use with risk of dementia in patients with type 2 diabetes: A systematic review and meta-analysis
- Anti-diabetic agents and the risks of dementia in patients with type 2 diabetes: a systematic review and network meta-analysis of observational studies and randomized controlled trials
- Association of Medicare bundled payment model with joint replacement care for people with dementia
- Medicare's Annual Wellness Visit and diagnoses of dementias and cognitive impairment
Browse the library with filters →
Related: how much of the library is free to read · APOE4 by the numbers · how to read a study