HKCFP Conjoint MCQ common mistakes
HKCFP Conjoint MCQ · Understand the traps and how to avoid them.
Choosing the answer that reflects what the candidate personally does in their own clinic rather than what the question's stem describes, then moving on without checking the other options.
Experienced family physicians answer from habit and pattern recognition; the Conjoint MCQ rewards the best answer for the scenario given, which often differs from routine local practice, especially where guidelines have moved on.
What to do instead
Before committing, re-read the stem and underline the defining features (age, duration, red flags, setting, resources stated). Ask 'does this option fit every stated feature?' and eliminate options that fit only some. Practise justifying the two strongest options in one sentence each before selecting.
Stopping at the first plausible option and not reading all four or five choices to the end.
Time pressure and the single-best-answer format create a satisficing habit; many stems are constructed so that an early option is a common trap and a later option is the intended answer.
What to do instead
Adopt a fixed routine: read all options first, mark each as true/false/unsure against the stem, then choose. If two remain, compare them directly on the single discriminating feature rather than re-reading the whole stem.
Missing the negative or 'EXCEPT'/'LEAST likely' framing and answering the positive version of the question.
Stems are long and the qualifier is often a single word buried mid-sentence; under time pressure the brain reads the topic and skips the instruction.
What to do instead
Circle or mentally capitalise the instruction word in every stem before looking at options. If you catch yourself having answered the positive version, re-answer from scratch rather than trying to convert your choice.
Failing to act on red flags because the stem's presentation is common and reassuring (for example, treating new rectal bleeding in a 55-year-old as haemorrhoids, or exertional chest tightness as musculoskeletal).
Family medicine training emphasises high-prevalence benign explanations; exam writers deliberately use common presentations with one or two sinister features to test safety-netting and referral thresholds.
What to do instead
Build a personal red-flag checklist per symptom (bleeding, weight loss, night pain, age over threshold, family history, immunosuppression) and scan the stem for it before deciding. When a red flag is present, the answer is almost always investigation or referral, not reassurance.
Managing the presenting complaint while ignoring the preventive or chronic-disease item embedded in the stem (for example, missing that the patient is overdue for cervical screening or that a new drug interacts with existing therapy).
Candidates read the stem as a single-problem consultation; Conjoint-style items often test whole-person care, and the correct option addresses the comorbidity, medication, or screening gap.
What to do instead
After identifying the presenting problem, list the patient's other active issues and medications from the stem and check whether the chosen option is safe and complete in that context. Practise 'one problem, whole patient' summaries during revision.
Selecting a management option that is unavailable or inappropriate in the stated setting (for example, ordering an MRI first-line in a resource-limited clinic, or admitting a patient who could be safely managed in the community).
Candidates answer from a hospital or tertiary perspective, or from an idealised guideline pathway, ignoring the setting cues the stem provides.
What to do instead
Note the stated setting and resources explicitly. Rank options by the accepted stepwise pathway (clinical assessment, then basic investigations, then specialist referral or admission) and prefer the earliest appropriate step unless the stem shows it has already been done.
Over-investigating: choosing a battery of tests or imaging when the stem describes a self-limiting condition with no red flags.
Anxiety about missing serious disease, and familiarity with defensive practice, push candidates toward maximal investigation; examiners test whether you can justify conservative management.
What to do instead
For each investigation option ask 'will the result change management, and is the pretest probability high enough to justify it?' If not, choose the conservative option with explicit safety-netting and follow-up.
Misreading the epidemiological or study-design option (for example, confusing sensitivity with positive predictive value, or absolute with relative risk reduction) and choosing the numerically impressive answer.
These items are short and look like recall questions, so candidates answer quickly from memory of definitions rather than reasoning from the numbers given in the stem.
What to do instead
Write the 2x2 table or the risk difference out from the stem numbers before choosing. Remember that predictive values depend on prevalence while sensitivity and specificity do not, and that relative risk reduction always looks larger than absolute risk reduction.
Failing to recognise the ethical or communication dimension of a scenario, choosing a clinically correct but ethically wrong action (for example, disclosing information to a relative without consent, or continuing treatment against a competent patient's refusal).
Candidates focus on diagnosis and treatment and treat consent, confidentiality, and capacity as background; examiners weight these heavily in family medicine.
What to do instead
When a stem involves a relative, an adolescent, an incompetent patient, or a refusal, apply a fixed sequence: assess capacity, obtain consent, protect confidentiality, act in the patient's best interest, and document. Choose the option that preserves the patient's autonomy and the doctor-patient relationship.
Leaving no time to review flagged items and never checking that the answer sheet matches the intended answers, especially after changing an answer.
Candidates spend too long on a few difficult stems, then rush the final section and misalign or double-mark answers under pressure.
What to do instead
Set a per-question time budget from the total number of items, flag and move on after the budget, and reserve the last few minutes for a transfer and alignment check. Change an answer only when you can state a concrete reason, since first-instinct changes are as often wrong as right.