How Doctors Decide Which Imaging Test to Order

Selecting the right imaging test is a clinical judgment process that balances diagnostic yield, radiation exposure, cost, and patient-specific factors. Ordering physicians draw on evidence-based guidelines, specialist consultation with radiologists, and a structured understanding of what each modality can and cannot detect. Errors in this process—ordering CT when ultrasound suffices, or delaying MRI when soft-tissue detail is essential—carry real consequences for patients and healthcare systems. This page explains the framework clinicians use to navigate those decisions.

Definition and scope

Imaging appropriateness refers to the degree to which a specific imaging study is matched to a clinical indication in a way that is expected to improve patient outcomes. The American College of Radiology (ACR) has formalized this concept through its ACR Appropriateness Criteria®, a publicly available, evidence-based decision support system covering more than 240 clinical conditions and variants. Each scenario is rated on a 1–9 scale, where 7–9 indicates "usually appropriate," 4–6 indicates "may be appropriate," and 1–3 indicates "usually not appropriate."

The scope of the decision involves five primary imaging modalities in routine clinical use:

Understanding the full landscape of diagnostic imaging at radiologyauthority.com requires familiarity with how each modality occupies a distinct diagnostic niche.

How it works

The imaging selection process follows a structured, iterative logic. Clinicians typically work through the following sequence:

  1. Define the clinical question. What anatomical region is involved? Is the question structural (mass, fracture, obstruction) or functional (perfusion, metabolic activity)?

  2. Assess the pre-test probability. A patient with a unilateral swollen leg after prolonged immobility has a high pre-test probability of deep vein thrombosis; duplex ultrasound is the first-line study per ACR criteria. The same swelling in a 14-year-old athlete shifts the differential, potentially toward MRI.

  3. Apply radiation safety principles. The FDA's Center for Devices and Radiological Health (CDRH) promotes the ALARA principle ("As Low As Reasonably Achievable"), which requires that ionizing radiation exposure be minimized without compromising diagnostic adequacy. A pediatrician ordering imaging for a child must weigh this standard explicitly — a topic addressed in depth at pediatric radiation safety.

  4. Check for contraindications. MRI is contraindicated in patients with certain implanted metallic devices; contrast agents carry reaction risks in patients with impaired renal function or prior contrast reactions. These are gatekeeping criteria, not preferences.

  5. Consult appropriateness criteria or a radiologist. Many health systems have integrated clinical decision support (CDS) tools at the point of ordering. The Protecting Access to Medicare Act of 2014 (PAMA), Section 218 mandated the use of qualified CDS mechanisms for ordering advanced imaging studies under Medicare, with the CMS Appropriate Use Criteria Program establishing a qualified provider list.

  6. Consider logistics and patient tolerance. A claustrophobic patient may not tolerate a 45-minute MRI. An unstable emergency patient may need CT because it is faster and available 24/7 in most hospitals.

Common scenarios

Three high-frequency clinical situations illustrate how the decision tree operates in practice:

Acute abdominal pain in an adult: ACR Appropriateness Criteria rates CT of the abdomen and pelvis with contrast as "usually appropriate" for undifferentiated acute abdominal pain in adults. Ultrasound is preferred first-line for suspected gallbladder disease or appendicitis in younger patients and pregnant women to avoid ionizing radiation. MRI is reserved for pregnant patients where ultrasound is inconclusive. Full decision details are covered in imaging for abdominal pain.

Chest pain with cardiac concern: Electrocardiography and troponin levels guide the first branch. For patients at intermediate risk of coronary artery disease, CT coronary angiography (CTCA) has demonstrated diagnostic accuracy comparable to invasive catheterization in appropriately selected populations, per the NEJM PROMISE trial data (2015). More detail is available at imaging for chest pain and heart conditions.

Headache with neurological symptoms: ACR criteria rate non-contrast CT of the head as "usually appropriate" for sudden severe headache ("thunderclap") to exclude subarachnoid hemorrhage. MRI without and with contrast is preferred for evaluating new focal deficits, suspected demyelinating disease, or posterior fossa pathology where CT resolution is limited. See the dedicated page on imaging for headaches and neurological symptoms.

Decision boundaries

Not every symptom warrants imaging. The ACR, alongside the Choosing Wisely® initiative (American Board of Internal Medicine Foundation), has identified more than 80 radiology-specific recommendations against imaging studies that provide minimal clinical benefit in defined low-risk scenarios. Examples include: imaging for uncomplicated low back pain within the first 6 weeks of onset in patients without red-flag symptoms, and routine chest X-ray prior to low-risk elective surgery.

The regulatory context for radiology establishes the oversight structure within which these decisions operate, including CMS reimbursement rules, Joint Commission standards, and state radiation control programs.

The key contrasts that define appropriateness boundaries are:

Factor Favors lower-intensity modality Favors higher-intensity modality
Patient age Pediatric (minimize radiation) Adult with complex pathology
Clinical urgency Stable, outpatient Acute, hemodynamically unstable
Soft tissue detail needed No Yes (MRI preferred)
Suspected vascular pathology Doppler ultrasound CT angiography or MR angiography
Prior imaging available Repeat may be deferred Comparison changes management

Radiation dose, renal function, implant status, pregnancy status, and institutional availability all function as hard constraints, not preferences, in this framework. Decisions made outside evidence-based criteria carry documentation and reimbursement consequences under Medicare's AUC Program.

References

The law belongs to the people. Georgia v. Public.Resource.Org, 590 U.S. (2020)