Diagnostic Radiographer common mistakes
Diagnostic Radiographer · Understand the traps and how to avoid them.
Treating the annual occupational dose limit (20 mSv/year averaged over 5 years, 50 mSv in any single year) as interchangeable with the public limit (1 mSv/year) or with a single-exam fetal limit, and applying the wrong number when a stem describes a worker, a member of the public, or an embryo/fetus.
Candidates memorize a single 'dose limit' figure and then pattern-match on the word 'limit' without first classifying the exposed person. The ICRP three-tier structure (occupational, public, conceptus) is learned as a list rather than as a decision tree, so under time pressure the first recalled number wins.
What to do instead
Before answering any dose-limit item, write one word identifying the exposed category: WORKER, PUBLIC, or FETUS. Then apply the tier: worker 20 mSv/year averaged over 5 consecutive years with no year exceeding 50 mSv; public 1 mSv/year; conceptus 1 mSv total after declaration of pregnancy. Practise by re-sorting a mixed set of stems into those three buckets before looking at the answer options.
Confusing the three radiation risk models (linear no-threshold, linear-threshold, hormesis) and misassigning which model underpins radiation protection practice, or attributing deterministic effects to stochastic mechanisms and vice versa.
The models are taught together in one radiobiology block, so the labels blur. Deterministic effects (threshold, severity rises with dose) and stochastic effects (no threshold, probability rises with dose) get merged because both are described as 'dose-related'.
What to do instead
Anchor each model to one sentence you can recall cold: LNT = risk proportional to dose with no safe threshold, the basis of current protection; threshold = a dose below which no effect is assumed; hormesis = low doses supposedly beneficial, not used for protection. For effects, use the two-question test: does severity increase with dose (deterministic) or only probability (stochastic)? Link deterministic examples to explicit thresholds (e.g. transient erythema around 2 Gy, cataract around 0.5 Gy) so the category has concrete hooks.
Answering quality-control items with a remembered pass/fail number but no reference to the test method, so the candidate cannot tell whether a stem is describing a kVp accuracy check, a timer accuracy check, a half-value layer measurement, or an output reproducibility test.
QC standards are studied as tables of tolerances. Candidates memorize '±5%' or '±10%' without binding each tolerance to its instrument and procedure, so a stem that describes the action but not the name of the test is unrecognizable.
What to do instead
Build a two-column table: left column the physical quantity being verified (tube voltage, exposure time, radiation output, beam quality/HVL, collimation, AEC performance), right column the method and its tolerance. For each row, state the measuring device (non-invasive kVp meter, oscilloscope or digital timer, dosimeter, aluminium attenuators, test tool). When a stem describes an action, name the test first, then recall the tolerance.
Assuming a passing AEC or output measurement means the unit is performing correctly, without checking that the measurement conditions match the clinical ones (detector selected, kVp used, phantom thickness, field size, grid present).
Candidates treat QC as a checklist of 'was it done' rather than 'was it done under the specified conditions'. AEC tests are especially vulnerable because chamber selection, kVp, and phantom thickness all change the result, and a number in tolerance under the wrong conditions is meaningless.
What to do instead
For every QC item, state the conditions before judging the number: which AEC chamber, what kVp, what attenuator thickness, what collimation, grid in or out, and what the baseline was. If a stem omits a condition that would change the result, treat that omission as the point of the question. Practise writing the full test protocol for AEC reproducibility and for output reproducibility from memory, including setup, then compare against a reference.
Adjusting exposure factors by intuition ('increase mAs for a thicker part') instead of applying the 15% rule for kVp change and the inverse square law and grid conversion factors quantitatively.
Candidates know the rules exist but have not practised applying them to numbers, so under exam pressure they fall back on qualitative reasoning. Density and contrast are also confused because both change when kVp changes.
What to do instead
Drill the arithmetic until it is automatic: a 15% increase in kVp approximately doubles receptor exposure (so halve mAs to maintain it); doubling SID quarters exposure (inverse square); changing from a 5:1 to an 8:1 grid needs roughly a 1.5x mAs increase and 8:1 to 12:1 roughly 2x. Work 20 numeric problems where you must state the direction and magnitude of the mAs change, and separately state the effect on contrast (kVp up = lower contrast, longer scale).
Evaluating an image for 'good quality' by whether it looks pleasing rather than by the four named criteria (density/receptor exposure, contrast, recorded detail/spatial resolution, distortion), and failing to attribute an artefact to its source.
Clinical training rewards images that look acceptable. Candidates therefore answer from aesthetic impression and cannot separate a contrast problem (kVp, windowing) from a detail problem (focal spot, motion, OID) from a distortion problem (part angulation, SID).
What to do instead
For any image-quality stem, run a fixed four-step checklist in order: receptor exposure, contrast, spatial resolution, distortion. Then ask which technical variable controls the criterion named in the stem (focal spot size and OID for resolution; kVp and subject contrast for contrast; SID and part-film distance for magnification). For artefacts, classify by stage: exposure (motion, double exposure), processing (roller marks, chemical staining), receptor (dead pixels, ghosting), or handling (scratches, dust).
Positioning by approximate body landmarks without verifying the specific bony alignment criteria for the projection, so the answer names the right projection but the described centring point, patient rotation, or beam angulation is wrong.
Candidates learn projection names and 'where to point' but skip the evaluation criteria. Exam stems often describe a subtle error (rotation, wrong centring, insufficient inspiration) and ask what is wrong; without the criteria memorized, the candidate cannot spot it.
What to do instead
For each commonly examined projection, memorize three things as a triplet: patient position and part alignment, central ray entry point and angulation, and the evaluation criteria (which structures must be shown, how much of them, and what indicates no rotation). Rehearse by describing a chest PA (scapulae clear of lung fields, clavicles symmetric, 9-10 posterior ribs visible), a lateral knee (condyles superimposed, open joint space), and an AP pelvis (symmetric obturator foramina, coccyx in midline) from memory.
Treating patient identification, informed consent, and pregnancy screening as paperwork to be completed after the patient is positioned, rather than as steps that determine whether the examination proceeds.
Candidates study these as administrative topics in isolation and do not integrate them into a procedural timeline. Stems that place the error mid-procedure (e.g. a patient mentions a possible pregnancy after positioning) then defeat a candidate who only knows the rules abstractly.
What to do instead
Learn the sequence as a timeline you can recite: verify identity with two identifiers, confirm the request and clinical question, screen for pregnancy and contrast allergy/renal function, obtain and document consent where required, then position. For any scenario stem, locate the event on that timeline and ask which step should already have occurred. Practise with stems that introduce late information and decide whether the exam continues, is modified, or is deferred and escalated.
Naming a structure or plane from a cross-sectional image by its shape alone, without orienting the image (which side is left, which is anterior, is it axial, coronal, or sagittal) and without using the surrounding landmarks to confirm the level.
Candidates memorize isolated 'signs' (e.g. a structure that looks like a certain letter) but not the relational anatomy. On axial images, left and right are reversed relative to the viewer, and level changes alter which organs appear, so shape matching alone produces confident wrong answers.
What to do instead
For every cross-sectional item, first state the plane and the orientation convention (axial viewed from below: patient's left on the viewer's right; anterior at the top). Then identify two or three landmarks that fix the level (vertebral body shape, aortic bifurcation, renal hilum, diaphragm, pubic symphysis). Only after that name the structure. Practise by covering labels on images and writing the plane, orientation, and level before the structure name.
Mixing up the physics quantities and units (kVp vs keV, mAs vs mGy, exposure vs absorbed dose vs equivalent dose vs effective dose) and using the wrong quantity in a calculation or a protection argument.
The vocabulary is visually and phonetically similar, and candidates learn definitions without the unit attached. Without units, the quantities collapse into 'dose' and the reasoning that follows is unsound.
What to do instead
Build a quantity-unit-what-it-measures table and drill it: exposure in coulombs per kilogram (air ionization), absorbed dose in gray (energy per mass), equivalent dose in sievert (absorbed dose weighted by radiation weighting factor), effective dose in sievert (tissue-weighted sum). Attach the conversion: 1 Gy = 1 J/kg; for diagnostic X-rays the weighting factor is 1, so Gy and Sv are numerically equal but not conceptually interchangeable. Answer every calculation by writing the unit first, then the number.
Answering ethics and regulatory items from personal opinion about what seems fair, instead of citing the governing principle (beneficence, non-maleficence, autonomy, justice) or the specific regulatory obligation (justification, ALARA, audit, incident reporting, scope of practice).
These items read like opinion questions, so candidates respond as members of the public rather than as regulated professionals. They also do not distinguish legal duties from good practice, so they cannot tell which answer is mandatory.
What to do instead
For each scenario, name the principle or duty before choosing: autonomy (patient's informed choice, including refusal), beneficence/non-maleficence (net benefit, minimizing harm), justice (fair allocation, no discrimination), and the regulatory duties of justification, optimization/ALARA, dose limitation, and documentation. Ask whether the stem describes something legally required, professionally expected, or merely desirable, and choose the answer that matches the strongest applicable obligation. Where a stem involves a conflict, state which duty takes precedence and why.