Understanding CT Radiation in Emergency Medicine
CT scans are among the most valuable diagnostic tools in emergency medicine. They use ionizing X-rays to produce cross-sectional images, delivering higher radiation doses than plain radiographs but providing far more diagnostic information.
For medically appropriate ED scans, clinical benefit usually outweighs the small, uncertain long-term stochastic risk. The key word is "appropriate" — imaging decisions should be driven by clinical indication, not default ordering.
This tool helps trainees understand the relative scale of imaging doses, the strength of evidence at different dose levels, and how to frame these tradeoffs in clinical conversation.
Putting Radiation Dose in Context
The average American receives approximately 3 mSv of background radiation per year from cosmic rays, radon, and terrestrial sources. A single chest X-ray delivers about 0.1 mSv — roughly one day of background exposure.
Most diagnostic CT scans fall in the 2–15 mSv range. A head CT delivers about 2 mSv (8 months of background), while a CT abdomen/pelvis delivers about 8 mSv (2.7 years of background). These comparisons help communicate relative scale without implying precise risk.
Key principle: dose equivalents are for communication and scale, not for calculating individual cancer risk.
Competing Models of Radiation Risk at Low Doses
The relationship between radiation dose and cancer risk at low doses is one of the most debated questions in radiation biology. Above ~100 mSv, epidemiological data (primarily from atomic bomb survivors and nuclear worker cohorts) directly demonstrates increased cancer risk. Below that threshold, four competing models offer very different predictions.
The Linear No-Threshold (LNT) model — the current regulatory standard — assumes risk is proportional to dose with no safe level. The Threshold model argues that cellular repair mechanisms can fully handle low doses, implying zero excess risk below a threshold (perhaps 50–200 mSv). Radiation Hormesis goes further, suggesting low doses may be slightly beneficial by upregulating DNA repair and immune surveillance. The Supralinear model takes the opposite position: low doses may be disproportionately harmful because repair mechanisms respond less efficiently to sparse, low-level damage.
The BEIR VII committee (2006) evaluated all four and retained LNT as the best available model for radiation protection policy, while explicitly noting that the data at low doses are insufficient to distinguish between these models. UNSCEAR (2012) reached a similar conclusion. This unresolved scientific debate is precisely why professional organizations like AAPM and HPS advise against giving patients precise individual risk numbers from diagnostic CT scans.
The practical implication: at diagnostic CT doses (1–15 mSv), science cannot tell you whether the risk is very small, zero, or slightly negative. What it can tell you is that the risk — whatever it is — is far smaller than the clinical consequences of missing a serious diagnosis.
Evidence and Uncertainty at Low Doses
At high doses (above ~100 mSv), cancer risk increases are directly demonstrated in epidemiological studies. At typical diagnostic imaging doses (below 10 mSv), individual risk cannot be reliably separated from statistical noise in population studies.
Risk estimates at low doses rely on mathematical extrapolation — primarily the Linear No-Threshold model. Professional organizations including AAPM, HPS, and ICRP have published position statements cautioning against over-precise individual risk communication at these dose levels.
This does not mean "zero risk." It means the risk is small enough that we cannot measure it directly and should not overstate our certainty about it.
Communicating with Patients About CT Radiation
Effective radiation communication avoids both extremes: dismissing radiation entirely ("there is no risk") and catastrophizing it ("this CT could cause cancer").
A balanced approach acknowledges that CT uses ionizing radiation, notes that the dose is generally small for a single indicated scan, and emphasizes that the immediate diagnostic benefit typically outweighs the uncertain long-term risk.
Recommended framing: "This scan uses radiation, but for a single medically appropriate CT the long-term cancer risk is generally thought to be small and hard to measure precisely. We are recommending it because the immediate benefit is finding or ruling out a dangerous condition now."
