CRT (IJCAHPO) common mistakes
CRT · Understand the traps and how to avoid them.
Reading the question stem once and answering from the first familiar keyword (e.g., seeing 'peak flow' and jumping to a flow-volume answer) without identifying what is actually being asked.
High-stakes timed exams reward speed, and candidates over-learn pattern-matching during practice, so they stop parsing the stem's key qualifier (MOST likely, FIRST action, EXCEPT).
What to do instead
Adopt a two-pass stem habit: on pass one underline the task word and the key clinical qualifier; on pass two predict the answer before looking at options. Practice with stems that contain a 'MOST likely' vs 'FIRST' contrast until the habit is automatic.
Choosing a physiologically correct but non-prioritized action, such as answering 'obtain an ABG' when the patient is in visible distress and the correct answer is 'begin bag-mask ventilation.'
Candidates reason from textbook completeness (gather all data) rather than from resuscitation sequencing, and content review rarely drills priority ordering.
What to do instead
Rehearse a fixed priority hierarchy for every scenario: safety, airway/breathing/circulation stabilization, then diagnostics, then definitive therapy. Before selecting, ask 'does this option stabilize or does it merely inform?' and eliminate inform-only options when instability is present.
Applying adult normal values or adult ventilator settings to neonatal/pediatric items, or vice versa, because the age qualifier in the stem was skimmed.
Most study time is spent on adult critical care, and adult parameters become the default mental template; pediatric/neonatal specifics are memorized late and weakly.
What to do instead
Build a one-page age-stratified reference (normal respiratory rate, heart rate, blood pressure, tidal volume, and key device settings by age band) and force yourself to state the patient's age band aloud before answering any item involving numbers.
Selecting a device or modality the patient cannot tolerate or that is contraindicated for the stated condition (e.g., recommending noninvasive ventilation for a patient with an unprotected airway or copious emesis).
Candidates memorize device indications as a list but not the exclusion criteria, so the indication cue dominates and contraindications are never checked.
What to do instead
For each major modality, maintain a paired 'indications / absolute and relative contraindications' card and run a two-second contraindication screen on the chosen option before committing. Include at least one contraindication-focused practice item per modality.
Miscalculating derived values (minute ventilation, tidal volume from IBW, oxygen cylinder duration, or expected PaO2 from FiO2) and then picking the option closest to the flawed arithmetic.
Calculations are done mentally under time pressure, unit conversions are skipped, and candidates do not sanity-check whether the result is physiologically plausible.
What to do instead
Write every calculation out with units carried through, and finish with a plausibility check (e.g., 'is 12 L/min minute ventilation plausible for an adult?'). Practice the same five calculation families until each takes under 45 seconds with written work.
Treating an arterial blood gas as a single-step interpretation (naming the primary disorder and stopping) instead of completing compensation and oxygenation assessment.
Candidates learn acid-base as disorder recognition and never build the habit of checking expected compensation, anion gap, or the oxygenation axis.
What to do instead
Use a rigid, written ABG algorithm every time: pH, primary disorder, expected compensation and whether it matches, anion gap if metabolic acidosis, then oxygenation. Apply the identical sequence in practice so it transfers under exam pressure.
Anchoring on the first plausible diagnosis and dismissing later options that better explain the full data set, especially when the stem includes a conflicting value.
Confirmation bias plus time pressure makes candidates stop evaluating once an option feels familiar; the discriminating detail is usually the one lab or vital sign that does not fit.
What to do instead
Before finalizing, explicitly test the leading option against every abnormal data point in the stem and ask 'which finding does this option fail to explain?' Switch only when another option explains more of the data.
Confusing similar-sounding concepts or abbreviations (e.g., CPAP vs. bilevel support, compliance vs. resistance, shunt vs. dead space, FiO2 vs. flow) and answering on the basis of the label rather than the definition.
These term pairs are learned together in review, so they are stored as a fuzzy cluster rather than as distinct, contrastable definitions.
What to do instead
Create explicit contrast tables for each confusable pair: definition, formula or mechanism, clinical example, and one distinguishing question. Self-test by generating the definition from the term and the term from the definition in both directions.
Spending disproportionate time on a small number of hard items, leaving later items rushed or unanswered.
Candidates treat every item as equally worth solving and lack a rule for when to abandon an item, so a single calculation can consume several minutes.
What to do instead
Set a per-item time budget from the total item count and time allowed, and adopt a hard cutoff: if no answer is reachable within the budget, mark a best-guess answer, flag the item, and move on. Rehearse this in full-length timed practice so the cutoff is practiced, not aspirational.
Reviewing missed practice items by reading the correct answer only, without identifying why the chosen distractor was attractive.
Answer-key review feels productive and is faster than error analysis, so the underlying reasoning gap survives and the same distractor is chosen again.
What to do instead
Log every miss with three fields: the reasoning that produced the wrong choice, the specific fact or rule that was missing, and the stem cue that should have triggered the right rule. Re-test those logged items after a delay and track whether the same error class recurs.