The honest answer to "is radiation harmful?" is not one word: it depends. What matters is not whether radiation is present, but its type, dose and where it is delivered. Very high doses are genuinely harmful; at the low doses of diagnostic imaging the risk is very small but not taken as zero. Science clearly separates these two situations — and understanding the distinction prevents both needless fear and needless indifference. (This article is only about ionizing radiation.)
Two kinds of effect
Radiation's health effects split into two basic groups:1
- Deterministic (tissue reactions): appear above a threshold dose; severity rises with dose. Below the threshold they do not occur.
- Stochastic: have no threshold; with dose it is the probability, not the severity, that is assumed to rise. Cancer is in this group.
Deterministic (threshold)
These appear only at high doses — far above diagnostic imaging. Examples are skin reddening (erythema), hair loss and cataract of the eye lens. Each has a threshold: for example the lens cataract threshold is now taken as ~0.5 Gy, and early transient skin erythema is around ~2 Gy.3 A diagnostic radiograph or CT stays far below these thresholds; in practice these effects only arise in high-dose situations like long interventional fluoroscopy or radiotherapy. The good news: kept below the threshold, these effects are preventable.
Stochastic (no threshold)
This is the group that gets the most attention, because it concerns low doses too. Foremost among stochastic effects is cancer. To stay on the safe side, the protection approach uses the linear no-threshold (LNT) model: every dose, however small, is assumed to carry a small added risk.1 A key point here: LNT is used not to precisely compute individual risk, but to set protection cautiously. At very low doses the true risk is scientifically hard to measure directly; so we assume the unknown on the safe side.
Putting risk in context
Without numbers, risk stays abstract. According to the BEIR VII report, the population-average fatal cancer risk from radiation is roughly ~5% per 1,000 mSv (1 Sv).2 Let's make that concrete: for a chest radiograph (~0.1 mSv) the added risk is on the order of a few in a million — far below many everyday risks. Even for an abdomen CT (~8 mSv) this is a small addition for an individual, and in a justified exam the diagnostic benefit easily outweighs it.
The real issue is not in the individual but in the total: when millions of exams are done per year, these small individual risks become meaningful at the population level. That is exactly why every exposure should be justified and its dose optimized — i.e. ALARA. The goal is not to be afraid, but to avoid the unnecessary.
References
- ICRP Publication 103. The 2007 Recommendations of the ICRP — doku reaksiyonları (deterministik) ve stokastik etkiler; korunmada eşiksiz doğrusal (LNT) model. icrp.org
- BEIR VII Phase 2. Health Risks from Exposure to Low Levels of Ionizing Radiation, National Research Council, 2006 — düşük doz kanser riski tahminleri (~%5/Sv mertebesinde, popülasyon ortalaması).
- ICRP Publication 118 — göz merceği katarakt eşiği ~0,5 Gy. Bushberg, The Essential Physics of Medical Imaging, 3rd ed., Bölüm 20 (Radiation Biology) — cilt ve doku reaksiyonu eşik dozları.
- İlişkili: Radyasyon Nedir? · Doz Neden Önemli? · ALARA Prensibi · Radyasyondan Korunma