Skip to main content

U.S. Population Medical Radiation Exposure

By Troy Zhou, PhD, DABR, DABSNM
• 15 min read

For roughly a quarter century, the message about medical radiation was that it kept rising. From 1980 to 2006, the per capita dose Americans received from medical imaging grew several-fold, driven mainly by computed tomography and nuclear medicine, until medical exposure rivaled natural background as a source of population dose.12

NCRP Report No. 184, published in 2019, documented that the trend reversed. Between 2006 and 2016, the per capita effective dose from diagnostic and interventional medical procedures in the United States fell from about 2.9 mSv to about 2.3 mSv, and the total collective effective dose dropped from roughly 885,000 to 755,000 person-sieverts — even as the population grew by about 24 million people.13

This article explains what those numbers mean, how collective and per capita dose are defined and calculated, which procedures drive population dose, why the long rise reversed, and how a radiation safety officer or imaging facility should use population-level data to sharpen justification, optimization, and risk communication.1

Introduction

Population dose is not an abstraction. It is the aggregate of every justified — and every unjustified — imaging examination, and it is the quantity that connects an individual facility's protocols to a national radiation-protection picture. The National Council on Radiation Protection and Measurements (NCRP) periodically assesses it, and the two most relevant reports are NCRP Report No. 160 (describing 2006) and NCRP Report No. 184 (describing 2016).12

The headline from Report No. 184 is encouraging but easy to misread. The population-average medical dose fell, but not because imaging became less common. The number of CT examinations actually increased. The per-examination and per-procedure doses fell, the mix of procedures shifted away from the highest-dose nuclear medicine studies, and film-based imaging was essentially eliminated — and those optimization effects outweighed the growth in CT volume.13

For a radiation safety program, the practical value of this data is directional. It tells a radiation safety officer where the dose actually is (CT), where optimization paid off (per-procedure dose and nuclear medicine mix), and how to frame medical dose honestly for patients (comparable, on a population average, to natural background). None of that replaces patient-specific dose management, but all of it informs priorities.1

Topic Explanation

What NCRP Report No. 184 assessed

NCRP Report No. 184 is titled Medical Radiation Exposure of Patients in the United States. It estimated, for 2016, both the number of medical procedures using ionizing radiation and the effective dose associated with each category, then compared the results with the 2006 figures from NCRP Report No. 160. Effective doses for procedures were drawn from the published literature, and procedure frequencies were assembled from commercial, governmental, and professional-society data. Tissue-weighting factors from ICRP Publication 60 were used so the comparison with the 2006 data would be consistent.13

Importantly, the population-dose calculation deliberately excluded dose from radiation therapy itself. Radiotherapy target doses are not meaningfully expressed as a population effective dose, so Report No. 184 treated therapeutic dose separately and focused the population estimate on diagnostic and interventional procedures, while a companion summary addressed the ancillary imaging doses that accompany radiation oncology.14

The headline numbers

The core findings, on a population basis, are worth stating precisely:1

  • In 2006, an estimated 377 million diagnostic and interventional radiologic examinations were performed in the United States; in 2016 the number was essentially the same, despite population growth.
  • The number of CT examinations increased from 67 million (2006) to 84 million (2016).
  • The number of other procedures (such as diagnostic fluoroscopy) and nuclear medicine procedures decreased from 17 million to 13.5 million.
  • Dental radiographic and dental CT examinations in 2016 were estimated at about 320 million (counted separately, at low individual dose).
  • The per capita annual effective dose from diagnostic and interventional medical procedures was about 2.9 mSv in 2006 and 2.3 mSv in 2016.
  • The collective effective dose was about 885,000 person-Sv in 2006 and 755,000 person-Sv in 2016.

Effective dose as a population tool — and its limits

These estimates all rest on effective dose, a protection quantity that weights each organ's absorbed dose by its radiosensitivity and sums them to a single whole-body-equivalent number in sieverts. Effective dose is designed for comparing procedures and tracking population trends, not for estimating the risk to a specific patient — it uses reference-population tissue weighting and reference phantoms, not individual anatomy. Population-dose reporting is exactly the setting effective dose was built for; individual risk estimation is exactly where it should be used with caution.12

The long rise this reversed

The significance of the 2016 numbers is clearest against what came before. NCRP Report No. 160, which described the year 2006, found that medical exposure had grown dramatically over the preceding decades and had become, by 2006, the largest man-made source of radiation exposure to the U.S. population — roughly comparable in magnitude to natural background on a per capita basis. That report placed the 2006 per capita effective dose from all sources at about 6.2 mSv, with natural background contributing roughly half and medical procedures contributing most of the remainder.12

The dominant drivers of that increase were computed tomography and nuclear medicine. CT's per-examination dose and rapidly rising volume, together with higher-dose cardiac nuclear medicine studies, pushed the medical component up several-fold between the early 1980s and 2006. The concern at the time was that the trend would continue indefinitely as imaging technology spread.23

Report No. 184 tested that assumption with 2016 data and found it had broken. The combination of lower per-examination CT doses, a decline in the highest-dose nuclear medicine procedures, the near-complete transition away from film, and reduced noninterventional fluoroscopy produced a net decrease in both per capita and collective medical dose. In other words, the technologies that had driven the rise — CT above all — were also where optimization produced the largest savings, which is precisely why they remain the focus of radiation protection effort.13

Key Technical Principles

Collective and per capita effective dose

Collective effective dose is the sum of the mean effective doses delivered by each procedure category , weighted by the number of procedures :

Per capita effective dose is simply the collective dose spread over the population :

Worked example — recovering the per capita figure

Using the reported 2016 collective dose of person-Sv and a U.S. population of about million:

which reproduces the reported per capita value, confirming the internal consistency of the collective and per capita figures.1

Worked example — why CT dominates

The dominance of CT follows directly from : a category's contribution is its volume times its mean dose. Take the 2016 CT volume of million and a representative adult CT effective dose of about mSv:15

That single illustrative figure is already most of the 755,000 person-Sv total, which is why CT — not the far more numerous dental or chest radiographs — governs collective medical dose. Expressed per capita:

In other words, on the order of three-quarters of the population-average medical dose comes from CT, even though CT is a minority of imaging examinations by count. This is the single most actionable fact in the entire dataset: dose-reduction effort returns the most when it is aimed at CT.15

Representative procedure doses

The spread of effective dose across procedures is enormous — more than a factor of a thousand across standard radiography alone. The following representative adult values illustrate the range; actual dose depends on technique, scanner, and patient size.5

Procedure Representative effective dose (mSv)
Chest radiograph (PA and lateral) 0.1
Dental intraoral radiograph under 0.01
Mammography (screening) 0.4
Lumbar spine radiograph 1.5
Head CT 2
Chest CT 7
Abdomen and pelvis CT 10
Coronary CT angiography up to ~16
Tc-99m MDP bone scan 6.3
F-18 FDG PET ~14

For context, published catalog ranges place standard radiography at roughly 0.01–10 mSv, CT at roughly 2–20 mSv, fluoroscopically guided interventional procedures at roughly 5–70 mSv, and most nuclear medicine procedures at roughly 0.3–20 mSv, against an average annual natural background of about 3 mSv.5

Clinical Impact

The population data reframes what "radiation safety" should prioritize. Because CT dominates collective dose, the highest-yield safety investments are CT protocol optimization, dose monitoring, iterative and deep-learning reconstruction, size-based technique, and appropriateness (justification) programs that avoid unnecessary CT in the first place. A facility that optimizes its CT protocols is moving the needle on the largest single component of medical population dose.15

The reversal documented in Report No. 184 is itself a proof of concept: it shows that coordinated attention to per-procedure dose and to procedure selection can lower population dose even while imaging volume holds steady or grows. The decline was driven by lower per-examination CT doses, a shift away from the highest-dose nuclear medicine cardiac studies, the near-elimination of film, and reduced noninterventional fluoroscopy — all of them optimization and justification effects, not reductions in access to care.3

At the same time, the data cautions against complacency. CT volume is still rising, newer high-dose interventional and image-guided techniques are expanding, and cone-beam CT has grown rapidly, particularly in dentistry. Population dose is a moving target, and the gains of one decade are not guaranteed in the next.3

It is also worth remembering what a per capita average conceals. Dividing collective dose by the whole population spreads the total across everyone, including the majority who have little or no imaging in a given year, so the average markedly understates the dose received by the smaller group of patients who undergo repeated or high-dose examinations. A patient with a chronic condition requiring serial CT can accumulate far more than the 2.3 mSv population average in a single year. This is why population figures belong in program-level prioritization and public communication, while individual cumulative-dose tracking — flagging patients with high recurrent exposure — belongs in patient-level dose management. The two tools answer different questions and should not be substituted for one another.15

Practical Optimization Tips

Anchor patient conversations in population context

When a patient asks about the risk of a scan, a population-average figure of roughly 2.3 mSv per year from medical imaging, comparable to the roughly 3 mSv of annual natural background, gives an honest frame — while being clear that their individual dose depends on the specific examination. This is risk communication grounded in data rather than reassurance.15

Put the effort where the dose is

Direct optimization resources to CT first, because that is where collective dose concentrates. Protocol review by a qualified medical physicist, dose-index monitoring against diagnostic reference levels, and appropriate-use criteria deliver more population-dose benefit per unit effort than optimizing low-dose modalities.15

Track your own facility's dose indices

Report No. 184 is a national average; your facility has its own distribution. Maintain CT dose-index monitoring and compare against diagnostic reference levels and achievable doses so local outliers are caught and corrected — the same optimization behavior that produced the national decline.5

Use effective dose correctly

Use effective dose and population figures for comparing procedures, trending, and communication — not for computing an individual patient's cancer risk, which requires organ doses, age, sex, and appropriate risk models. Overstating the precision of a population quantity undermines credibility.12

Do not let the good-news trend justify unnecessary imaging

A falling per capita dose is not a license to relax justification. The reversal came partly from avoiding unnecessary procedures; sustaining it depends on continuing to ask whether each examination is justified.13

Regulatory Considerations

Population dose assessment is a radiation-protection science activity, not a licensing requirement, but it sits squarely within the protection framework that governs every facility. The guiding principles — justification, optimization (ALARA), and dose limitation — come from the ICRP system of radiological protection (Publications 60 and 103) and are implemented in the United States through NRC regulations (10 CFR Part 20 for radioactive material), FDA and state programs for radiation-producing machines, and accreditation requirements.12

Patient diagnostic exposures are deliberately not subject to regulatory dose limits, because limiting a justified medical dose could cause more harm than the radiation. Instead, patient protection relies on justification and optimization, supported by diagnostic reference levels, protocol review, and qualified medical physics oversight. Population-dose data such as NCRP Report No. 184 informs that system by showing where collective dose accumulates and whether optimization is working at the national scale.12

DRPS supports imaging facilities across Florida, Maryland, Virginia, Washington DC, California, Nevada, Pennsylvania, New York, New Jersey, and Delaware with the protocol optimization, dose monitoring, and physics oversight that turn population-level insight into local dose reduction. Radiation-producing machines such as CT are FDA- and state-regulated, while radioactive material used in nuclear medicine is regulated by the NRC or the relevant Agreement State. Always confirm requirements with the authority having jurisdiction.

Frequently Asked Questions (FAQs)

Is 2.3 mSv the dose from a single scan?

No. The 2.3 mSv figure is the 2016 per capita annual effective dose — the total collective medical dose divided by the entire U.S. population. It is a population average that includes everyone, including the many people who had no imaging that year. An individual examination can be far lower (a chest radiograph) or far higher (an interventional procedure).15

Why did population dose fall if CT use went up?

Because per-examination CT dose fell, the highest-dose nuclear medicine procedures declined, film imaging was essentially eliminated, and noninterventional fluoroscopy dropped sharply. Those optimization and procedure-mix effects outweighed the increase in CT volume.3

Does the data include radiation therapy?

The population-dose estimate deliberately excludes therapeutic radiotherapy dose, which cannot be meaningfully expressed as a population effective dose. Ancillary imaging associated with radiation oncology was addressed separately.14

Which tissue-weighting factors were used?

NCRP Report No. 184 used the ICRP Publication 60 tissue-weighting factors so that the 2016 results could be compared consistently with the 2006 data in NCRP Report No. 160. Newer ICRP Publication 103 factors differ for some tissues.12

How does medical dose compare to natural background?

On a population-average basis, the roughly 2.3 mSv per capita medical dose in 2016 is comparable to the roughly 3 mSv average annual natural background in the United States. Individual exposures vary over several orders of magnitude.15

Key Takeaways

  • NCRP Report No. 184 (2019) assessed U.S. medical radiation exposure for 2016 and found the per capita dose fell to about 2.3 mSv from about 2.9 mSv in 2006.1
  • Collective effective dose dropped from about 885,000 to 755,000 person-Sv despite population growth and rising CT volume.1
  • CT is the dominant source of collective medical dose; on the order of three-quarters of the per capita medical dose traces to CT.15
  • The decline came from lower per-procedure doses, a shift away from high-dose nuclear medicine, film elimination, and less noninterventional fluoroscopy — optimization, not reduced access.3
  • Effective dose and population figures are for comparison, trending, and communication, not individual risk estimation.12
  • The highest-yield radiation safety investments target CT optimization, dose monitoring, and justification.15

Conclusion

NCRP Report No. 184 turned a widely held assumption on its head: U.S. medical radiation dose, after decades of growth, declined between 2006 and 2016. The lesson for radiation safety is not that the problem is solved but that optimization and justification work at national scale. Because collective dose concentrates in CT, that is where continued attention pays off most, and because population figures are averages, they belong in risk communication and program prioritization rather than in individual risk estimates. A radiation safety program that understands where the dose is — and keeps asking whether each examination is justified — is what sustains the gains this report documented.135

How DRPS Can Help

Diagnostic Radiation Physics Services (DRPS) helps imaging facilities translate population-dose insight into local action with medical physicist consulting, CT physics testing and protocol optimization, dose-index monitoring against diagnostic reference levels, and radiation safety training for staff and radiation safety officers — all delivered by board-certified medical physicists.

Understanding where population dose comes from is the first step; a documented optimization and dose-monitoring program is how a facility acts on it.

Related Resources

References

  1. Mettler FA Jr, Mahesh M, Bhargavan-Chatfield M, et al. Patient Exposure from Radiologic and Nuclear Medicine Procedures in the United States: Procedure Volume and Effective Dose for the Period 2006-2016. Radiology. 2020;295(2):418-427. doi:10.1148/radiol.2020192256. doi.org
  2. National Council on Radiation Protection and Measurements. Ionizing Radiation Exposure of the Population of the United States. NCRP Report No. 160. Bethesda, MD: NCRP; 2009. ncrponline.org
  3. Mettler FA Jr. Medical Radiation Exposure in the United States: 2006-2016 Trends. Health Physics. 2019;116(2):126-128. doi:10.1097/HP.0000000000000996. doi.org
  4. Milano MT, Mahesh M, Mettler FA, Elee J, Vetter RJ. Patient Radiation Exposure: Imaging During Radiation Oncology Procedures: Executive Summary of NCRP Report No. 184. Journal of the American College of Radiology. 2020;17(9):1176-1182. doi:10.1016/j.jacr.2020.02.002. doi.org
  5. Mettler FA Jr, Huda W, Yoshizumi TT, Mahesh M. Effective doses in radiology and diagnostic nuclear medicine: a catalog. Radiology. 2008;248(1):254-263. doi:10.1148/radiol.2481071451. doi.org
  6. National Council on Radiation Protection and Measurements. Medical Radiation Exposure of Patients in the United States. NCRP Report No. 184. Bethesda, MD: NCRP; 2019. ncrponline.org
  7. International Commission on Radiological Protection. The 2007 Recommendations of the International Commission on Radiological Protection. ICRP Publication 103. Annals of the ICRP. 2007;37(2-4). icrp.org