Misinformation

Read it this way Belief in false COVID-19 or vaccine claims among regular Newsmax viewers, 76 percent, sits 31 percentage points above the 45 percent baseline for all US adults, with OANN and Fox News audiences also well above that line. Because these audience figures come from KFF's narrative summary rather than a full data table, treat the ranking as reliable but the exact percentages as approximate. Use this chart to see whether it shows exposure, belief, spread, or harm, and why the recommendation focuses on targeted response rather than broad undifferentiated messaging.

Caveat The audience figures come from KFF's written summary of the poll (Newsmax 76, OANN 67, Fox 61 versus 45 overall), not a structured data table.

KFF Health Misinformation Tracking Poll Pilot · 2023 · source

Read it this way On Facebook, the twelve Disinformation Dozen accounts alone produced more shares (502,970) than every other tracked anti-vaccine account combined (186,030), the opposite of the Twitter/X pattern, where they contribute only 20,400 of 120,000 tracked shares. Platform concentration of anti-vaccine content is not uniform. It is far worse on Facebook, though the account-level attribution is CCDH's own, not an independently audited count. Use this chart to see whether it shows exposure, belief, spread, or harm, and why the recommendation focuses on targeted response rather than broad undifferentiated messaging.

Caveat The 502,970 and 20,400 figures are CCDH's published shares (689,000 Facebook, 120,000 Twitter/X) split by its stated 73 percent and 17 percent Disinformation Dozen attributions. Other-account values are the remainder.

Center for Countering Digital Hate, The Disinformation Dozen · 2021 · source

Read it this way Follow the Disinformation Dozen node: it feeds 502,970 of the 689,000 Facebook shares but only 20,400 of the 120,000 Twitter/X shares, so the same twelve accounts have very different reach depending on platform. The diagram traces only these two platforms' tracked shares and excludes the broader 425-account, 59.2 million-follower ecosystem CCDH measured on a separate timeframe. Use this chart to see whether it shows exposure, belief, spread, or harm, and why the recommendation focuses on targeted response rather than broad undifferentiated messaging.

Caveat Anti-vaccine content is used here as the one publicly documented case study with real figures. The 425-account ecosystem CCDH tracked reached a combined 59.2 million followers as of December 2020, a follower-reach figure on a different scale and snapshot than these share counts, so it is not drawn as a flow link.

Center for Countering Digital Hate, The Disinformation Dozen · 2021 · source

Read it this way Stated vaccination intent fell by nearly the same amount in both countries, 6.4 points in the US and 6.2 points in the UK, even though the UK started from a higher baseline of 54.1 percent versus 42.5 percent. Because this came from a randomized experiment, the drop can be read as a causal effect of the misinformation shown, within the reported uncertainty range of roughly 4 to 9 points, not just an association. Use this chart to see whether it shows exposure, belief, spread, or harm, and why the recommendation focuses on targeted response rather than broad undifferentiated messaging.

Caveat The baselines (42.5 percent US, 54.1 percent UK) and the modeled declines (6.4 points US, 6.2 points UK) come from Loomba et al. 2021. Post-exposure values are the baseline minus the modeled drop. The 95 percent uncertainty intervals are 8.8 to 4.0 points (US) and 8.5 to 3.9 points (UK).

Nature Human Behaviour, COVID-19 vaccine misinformation and intent · 2021 · source

Read it this way Breast cancer patients who chose alternative medicine faced the highest risk shown here, a hazard ratio of 5.68, more than double the 2.5 overall figure, while the prostate cancer estimate of 1.68 did not reach statistical significance and should not be read as a real effect. These ratios come from one matched cohort of 281 alternative-medicine patients versus 560 conventionally treated patients, so read them as evidence within this study, not a universal harm multiplier for every alternative-medicine choice. Use this chart to see whether it shows exposure, belief, spread, or harm, and why the recommendation focuses on targeted response rather than broad undifferentiated messaging.

Caveat Modeled hazard ratios from one National Cancer Database cohort (281 alternative-medicine patients matched to 560 conventionally treated patients, 2004 to 2013), not a count of deaths. The prostate figure was not statistically significant. Harm estimates across misinformation topics use different study designs and do not compare cleanly.

Journal of the National Cancer Institute, alternative medicine and survival · 2018 · source