CBGC Genetic Counselling Certification (CCGC) common mistakes
CBGC Genetic Counselling Certification (CCGC) · Understand the traps and how to avoid them.
When calculating recurrence risk from a pedigree, the candidate stops at the first probability (for example, the chance a consultand is a carrier) and reports it as the final risk, without multiplying through the remaining steps (carrier × carrier × 1/4 for an affected child, or adding the separate pathways that can produce the same outcome).
Pedigree problems are taught as single-step textbook examples, and under time pressure the first number that 'looks like an answer' gets written down. Bayesian and pathway problems also require holding several conditional probabilities in working memory, and the final multiplication or summation step is the one most often dropped.
What to do instead
Adopt a fixed written template for every risk question: (1) draw the full pedigree and label each person's prior probability; (2) list every pathway that leads to the outcome; (3) compute each pathway as an explicit chain of fractions; (4) sum the pathways; (5) sanity-check the answer against the boundary conditions (must be between 0 and 1, and consistent with the number of affected relatives). Practise with at least 15 two- and three-pathway problems until the template is automatic.
In a Bayesian risk problem, the candidate uses the general population carrier frequency as the posterior probability for a person with a negative test result, instead of updating the prior with the test's sensitivity and the family history, or uses the detection rate as if it were the residual risk.
The terms prior, conditional, joint, and posterior blur together when the question is worded clinically rather than statistically. Candidates also forget that a negative test reduces but does not eliminate carrier risk, and that the family-history prior (for example, a 2/3 chance for an unaffected sibling of an affected child) overrides the population figure.
What to do instead
Before computing, write the four-row Bayesian table headings and fill each cell explicitly: prior, conditional (test result given carrier status), joint, and posterior. State in words what each conditional probability represents. Memorise the residual-risk shortcut (posterior carrier risk after a negative test = prior × (1 − detection rate) divided by the total probability of the observed result) and verify it against a full table calculation at least once per study session.
The candidate answers a variant-classification or test-result question by naming the gene and the clinical syndrome but never addresses the specific evidence categories (population frequency, segregation, functional data, de novo status, computational predictors) or the difference between a variant of uncertain significance and a pathogenic variant for clinical action.
Clinical knowledge is over-rehearsed relative to laboratory literacy, so candidates default to phenotype recall. Classification frameworks are memorised as a list of criteria names without practice applying them to a described variant, so the candidate cannot translate the stem's evidence into a category.
What to do instead
For every variant described in a practice question, practise writing a two-line evidence summary that assigns each piece of data to a category (population, computational, functional, segregation, de novo, allelic) and then states the resulting classification and its clinical implication. Build a one-page table of the classification categories with the minimum evidence needed for each, and drill 10 short vignettes where the answer hinges on whether the evidence is sufficient.
When a question describes a test with a certain detection rate, the candidate interprets a negative result as ruling out the condition, or quotes the detection rate as the negative predictive value, without stating the residual risk or the limits of the technology (for example, that a sequencing panel does not detect copy-number changes or deep intronic variants).
Candidates conflate analytic validity, clinical validity, and clinical utility, and treat 'negative' as a binary outcome. The vocabulary of sensitivity, specificity, detection rate, and residual risk is learned as definitions rather than as quantities to be applied.
What to do instead
For each test technology on the exam (karyotype, chromosomal microarray, targeted panel, exome, cfDNA screening, expanded carrier screening), write a card listing what it detects, what it misses, and the residual risk after a negative result. In practice questions, force yourself to include the phrase 'residual risk remains' with a number or a stated range in every negative-result answer before selecting an option.
The candidate selects the medically correct recommendation but ignores the counselling process embedded in the stem - for example, recommending immediate testing to a patient who has clearly stated she does not want it, or giving a risk figure before exploring the client's reasons for seeking counselling.
Exam stems are long, and candidates skim for the clinical facts and the answer choices, treating psychosocial content as decoration. Training also rewards knowing the right test, so the reflex is to act rather than to assess understanding and preference first.
What to do instead
Read the last sentence of the stem first to identify what is being asked (often 'the most appropriate next step' or 'the best response'), then read the whole stem and underline every statement of the client's values, prior beliefs, and stated preferences. Eliminate any option that overrides an expressed preference or delivers information before consent and agenda-setting. Practise with questions where the correct answer is a question back to the client rather than an action.
The candidate answers a prenatal question by quoting a procedure-related miscarriage risk or a screening detection rate that is outdated or misattributed to the wrong test, or recommends diagnostic testing without first establishing gestational age, the indication, and whether the client wants information that could lead to a decision about the pregnancy.
Numbers for procedure risk and screening performance change with technique and platform, and candidates memorise a single figure without its context. Prenatal questions also compress several decisions (screening versus diagnosis, timing, termination options, legal limits) into one stem, and candidates answer the technical part while skipping the decision context.
What to do instead
Maintain a dated reference sheet for prenatal numbers, each with its source and the population it applies to, and review it weekly. For every prenatal vignette, write the gestational age, the indication, the client's stated goal, and the legal gestational limit in the jurisdiction before looking at the options. Where a figure is uncertain, prefer the option that describes the range or the need to discuss uncertainty rather than a single precise number.
In a cancer genetics question, the candidate applies general population risks or a single gene's penetrance to a family with a different gene, or recommends testing for the wrong relative (testing an unaffected person when an affected relative has not been tested and is the informative sample).
Hereditary cancer syndromes share overlapping phenotypes, and candidates pattern-match on the tumour type rather than on the full pedigree (ages at diagnosis, multiple primaries, sex-specific cancers, Ashkenazi Jewish ancestry, number of affected relatives). The rule that testing should begin with the most informative affected relative is known but not applied under time pressure.
What to do instead
For each hereditary cancer syndrome on the exam, build a comparison table of associated tumours, typical ages, penetrance range, and the gene-specific management implications. For every vignette, first write the single most likely gene and then ask 'who is the best first person to test?' before reading the options. Rehearse the cascade-testing logic explicitly: affected relative first, then predictive testing of at-risk relatives with pre- and post-test counselling.
The candidate treats a question about a minor, a person lacking capacity, or a family with conflicting wishes by choosing the option that maximises medical information gathering, rather than the option that respects assent, confidentiality, and the client's own decision-making authority.
Clinical urgency and the instinct to be thorough pull toward testing and disclosure. Candidates also confuse 'duty to warn' with a general obligation to share genetic information with relatives, and do not distinguish confidentiality within a family from the clinician's legal obligations in the relevant jurisdiction.
What to do instead
Learn a decision framework rather than case-by-case rules: identify who holds decision-making authority, what consent is required, whether the information is the client's to disclose, and whether any legal exception applies. For each ethics question, write the competing principles (beneficence, autonomy, confidentiality, non-maleficence) and the option that best balances them. Flag any option that discloses information to relatives without consent as suspect unless the stem establishes a specific legal duty.
The candidate describes a management plan or referral pathway that is out of date or not specific - for example, recommending 'increased surveillance' without naming the modality, starting age, and interval, or omitting the referral to the relevant specialty (cardiology, oncology, neurology, reproductive endocrinology) that the stem's findings require.
Management recommendations are revised frequently, and candidates learn them as general principles rather than as concrete protocols. Exam stems often include a finding (an echocardiogram, a hormone profile, a family history of early cancer) that should trigger a specific referral, and candidates answer the genetic question while missing the clinical prompt.
What to do instead
Build a gene-to-management table with modality, starting age, interval, and the responsible specialty, and date each entry. In every vignette, scan for abnormal findings and ask 'which specialty needs to see this person, and what exactly should be done?' before answering. Where a protocol has changed recently, note both the current and prior recommendation so you can recognise a distractor built on the old one.
The candidate answers a professional-practice or research-literacy question by choosing the option that sounds most reassuring or most thorough, rather than the one that reflects the actual evidence hierarchy, the limits of a study design, or the counsellor's scope of practice and obligation to report uncertainty.
These domains feel less clinically concrete, so candidates rely on general impressions. Terms such as sensitivity, positive predictive value, and study validity are recognised but not applied to a described scenario, and scope-of-practice boundaries are not rehearsed explicitly.
What to do instead
Practise a short appraisal routine for any study described in a stem: design, population, outcome, comparison group, and the main threat to validity, then state what the result does and does not support. Keep a one-page summary of evidence hierarchy and the difference between screening and diagnostic performance measures. For scope-of-practice questions, write the counsellor's role, the limits, and the appropriate referral before reading the options.