ISCD CBDT common mistakes
ISCD CBDT · Understand the traps and how to avoid them.
Reporting a T-score computed from a reference database that does not match the patient's sex, ethnicity, or the manufacturer's normative set, rather than verifying which reference population the densitometer applied.
Candidates memorize that T-score compares to a young adult mean and stop there, without learning that the ISCD position is to use the manufacturer's own reference data for the device and to match sex and ethnicity; they assume any young-adult database is interchangeable.
What to do instead
For each scan, state the reference database name, version, and population. Confirm the technologist selected the correct sex/ethnicity option in the software before the report is finalized. Practice reading a printout and locating the reference-set field, then write a one-line justification of why that set is appropriate.
Applying the -2.5 T-score threshold to the wrong skeletal site, for example diagnosing osteoporosis from a forearm or from a single non-hip, non-spine site in a postmenopausal woman, or using the lowest T-score across all sites without regard to which sites are valid for diagnosis.
The rule 'lowest T-score counts' is taught as a shortcut, and candidates lose track of which sites the ISCD considers diagnostic (lumbar spine, total hip, femoral neck, and 33% radius under specified conditions) versus monitoring-only sites.
What to do instead
Build a site-by-site table: which sites are diagnostic, which are monitoring-only, and the conditions for using the radius (hip and spine not usable or not interpretable). Drill cases where the spine and hip disagree and write out which site drives the diagnosis and why.
Treating an invalid or uninterpretable lumbar spine scan as usable data instead of excluding it, for example reporting a spine T-score when fewer than two evaluable vertebrae remain or when a vertebra is obscured by artifact.
Software prints a number regardless of quality, and candidates report what the machine outputs rather than judging whether the minimum number of evaluable vertebrae is met and whether artifacts (surgery, vertebroplasty, scoliosis, degenerative change) invalidate the site.
What to do instead
Learn the evaluability criteria: number of evaluable vertebrae required, exclusion of vertebrae with focal structural change, and the effect of degenerative disease in raising spine BMD. Practice deciding, from a described scan, whether the spine is usable and whether the hip or radius should replace it.
Confusing T-score and Z-score indications, for example using a Z-score to diagnose osteoporosis in an older adult or using a T-score alone in a premenopausal woman, man under 50, or child.
Both scores come from the same printout and differ only in the reference group; candidates under-learn the rule that Z-scores are preferred for premenopausal women, men under 50, and children, and that 'below expected range for age' is defined by Z-score.
What to do instead
Memorize the decision rule by patient category and rehearse it on vignettes: postmenopausal woman and man 50+ use T-score for diagnosis; premenopausal woman, man under 50, and child use Z-score, with the term 'low bone mass for age' or 'below expected range for age' rather than osteoporosis. Write the rule out from memory weekly.
Ignoring the least significant change when comparing two scans, and declaring a real change in BMD when the difference falls within measurement error, or comparing scans done on different devices without cross-calibration.
Candidates focus on the arithmetic of percentage change and forget that change must exceed the LSC at the specified confidence level, and that LSC is site- and technologist-specific and derived from precision assessment.
What to do instead
Practice the calculation: LSC = 1.96 × √2 × precision error (as CV or SD), and apply it to concrete before/after numbers until the threshold is automatic. Learn that serial scans should be done on the same device, and that a change below LSC is reported as no significant change.
Misreading the precision assessment requirement, for example believing a single phantom scan establishes in vivo precision, or that precision is a fixed manufacturer value rather than a facility- and technologist-specific measurement.
Phantom scans and in vivo precision studies look similar on paper, and candidates conflate device calibration with the precision needed to compute LSC; they also overlook that each technologist should be assessed.
What to do instead
Distinguish the two clearly: phantom scanning verifies stability and calibration; in vivo precision requires repeat scans on volunteers with repositioning, analyzed per technologist, and is used to derive LSC. Memorize the recommended number of repeat measurements and the reporting of precision as RMS SD or CV.
Selecting the wrong acquisition mode or scan region for the patient's body habitus or anatomy, for example using the standard spine mode in a patient whose soft tissue thickness exceeds the device's validated range, or failing to adjust the hip region of interest.
Default protocols are accepted without checking the manufacturer's weight and thickness limits, and candidates do not know that mode selection affects accuracy and that an out-of-range patient should be scanned on an alternative device or site.
What to do instead
Learn each device's validated weight and thickness limits and the mode options (for example standard versus thick-patient modes on DXA). Rehearse the decision: if the patient exceeds limits, switch mode only if validated, otherwise use an alternative validated device or a peripheral site, and document it.
Failing to recognize that the report must include the technical quality assessment, the reference data used, and the impression with fracture risk context, producing a report that lists only BMD and T-score numbers.
Candidates study the numbers and treat report writing as clerical, so they omit the components the interpreting physician needs, such as scan quality, artifact, comparison to prior, and whether the study is adequate for diagnosis.
What to do instead
Build a report checklist from memory: patient identifiers and indication, sites scanned, technical quality and limitations, reference database, BMD and T-score/Z-score per site, comparison with prior and LSC, and impression including fracture risk category. Apply it to three sample scenarios and check each element is present.
Confusing the roles and responsibilities in the facility workflow, for example assuming the technologist may interpret the scan or that a physician need not review the acquisition before the report is issued.
The exam tests scope-of-practice and quality-control responsibilities, and candidates who focus only on physics and anatomy skim over who may do what, who signs the report, and what oversight the interpreting physician provides.
What to do instead
Write out the workflow from scheduling to report: who performs the scan, who reviews image quality, who interprets, who communicates results, and what documentation is required. Learn the distinction between acquisition, analysis, and interpretation, and the requirement for physician review of the study.
Overlooking radiation safety and quality-control routines specific to DXA, such as skipping daily quality-control scans, ignoring scatter radiation precautions, or failing to document QC results and corrective action.
DXA is low dose, so candidates deprioritize safety and QC, and they do not connect a failed QC scan to the required action of stopping clinical use until the problem is resolved.
What to do instead
Memorize the QC schedule (daily, weekly, or as specified by the manufacturer and facility protocol), what a control chart is for, and the rule that a shift or drift beyond limits requires investigation before patient scanning. Rehearse the corrective-action sequence and the documentation required.