THE GOLD STANDARD OF GENETICS · SPECIALIST STUDY
Turn a pharmacogenomic result into a bounded prescribing question
A genotype does not prescribe a medicine by itself. Its clinical meaning depends on which variants the assay covered, how alleles and diplotypes were assigned, the evidence connecting a gene to a particular medicine, the current medication list, organ function, age, ancestry, disease and the possibility of phenoconversion. Good clinical decision support keeps those layers visible. It helps a qualified prescriber see a relevant result at the right time without presenting genetics as the only cause of response or toxicity.
This guide belongs to the Genetics subject root, the Gold Standard of Education, and the How X Works hub. It begins with a concrete decision, then tests whether the method still holds when the context changes.
Choose a reading route
Build the genotype-to-phenotype chain · chapters 1, 2, 3
Add medicines and phenoconversion · chapters 4, 5
Evaluate evidence and work a case · chapters 6, 7, 8
Design decision support and monitoring · chapters 9, 10, 11
Govern uncertainty, updates and transfer · chapters 12, 13
The whole problem at a glance
| Layer | Question | Evidence |
|---|---|---|
| Assay | What variants and copy changes were measured? | Validated report and coverage |
| Interpretation | How were alleles and phenotype assigned? | Versioned translation table |
| Clinical | Which medicine and outcome are relevant? | Label or guideline plus patient context |
| Action | What should the prescriber consider? | Bounded recommendation and monitoring |
A correct genotype can still support the wrong action if phenotype, medicine or context is stale.
CHAPTER 1 OF 13 · Build the genotype-to-phenotype chain
1. Define the pharmacogenomic decision before reading the genotype
Start with the medicine question: selection, starting dose, adverse-event risk, efficacy, monitoring or avoidance. Then identify whether a validated gene–drug relationship addresses it. The working unit for this chapter is one patient, one verified genotype or diplotype, one medicine decision and one current clinical context. Keep the question narrow enough that a reader can identify which person, sample, decision or evidence item is actually being compared, and which conclusion remains outside scope.
Worked example: A CYP result is present in the record, but no action is considered until the prescriber identifies the active medicine, indication, timing and current product information. Write the starting information, the decision made from it, the later observation and the consequence in separate rows. That simple chronology prevents a plausible story from replacing a reproducible account.
Evidence to retain: patient identity; specimen; report; active and planned medicines; indication; dose; timing; organ function; relevant label or guideline Link every item to its date, version, source, custodian and decision use. Record missing or inaccessible evidence explicitly; absence is not a negative result and must not be converted into reassurance.
Boundary and misconception: “Actionable genotype” is not a universal property; actionability is medicine-, evidence- and context-specific. Test at least one competing explanation and state what evidence would change the conclusion. The pass condition is a bounded, reviewable claim—not a promise that the method removes every uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Move the same genotype from one medicine to another and rebuild the evidence instead of transferring the original recommendation. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 1 check · what should survive an audit
- Name the decision, unit, time boundary and authority.
- Preserve both the supporting evidence and the strongest rejected explanation.
- State the next action, review date, stop rule and condition that would reopen the question.
CHAPTER 2 OF 13 · Build the genotype-to-phenotype chain
2. Verify assay coverage and technical limits
Read what the laboratory actually measured, including named variants, copy-number capability, structural variants, quality metrics and the possibility of untested alleles. The working unit for this chapter is one patient, one verified genotype or diplotype, one medicine decision and one current clinical context. Keep the question narrow enough that a reader can identify which person, sample, decision or evidence item is actually being compared, and which conclusion remains outside scope.
Worked example: Two assays both report CYP2D6 but only one measures copy-number changes; the same apparent diplotype confidence cannot be assumed. Write the starting information, the decision made from it, the later observation and the consequence in separate rows. That simple chronology prevents a plausible story from replacing a reproducible account.
Evidence to retain: laboratory accreditation; assay version; variant list; copy-number method; no-call criteria; validation population; raw and interpreted result Link every item to its date, version, source, custodian and decision use. Record missing or inaccessible evidence explicitly; absence is not a negative result and must not be converted into reassurance.
Boundary and misconception: A negative finding means no covered variant was detected, not that every functional allele has been excluded. Test at least one competing explanation and state what evidence would change the conclusion. The pass condition is a bounded, reviewable claim—not a promise that the method removes every uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Apply the assay to a population with additional rare alleles and identify why residual uncertainty may increase. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 2 check · what should survive an audit
- Name the decision, unit, time boundary and authority.
- Preserve both the supporting evidence and the strongest rejected explanation.
- State the next action, review date, stop rule and condition that would reopen the question.
CHAPTER 3 OF 13 · Build the genotype-to-phenotype chain
3. Translate variants into alleles and diplotypes transparently
Preserve the allele-definition and translation version used to combine observed variants into star alleles or other haplotype labels. The working unit for this chapter is one patient, one verified genotype or diplotype, one medicine decision and one current clinical context. Keep the question narrow enough that a reader can identify which person, sample, decision or evidence item is actually being compared, and which conclusion remains outside scope.
Worked example: A future definition changes how one variant combination is named. The earlier report remains historically traceable and any reinterpretation is issued as a new version. Write the starting information, the decision made from it, the later observation and the consequence in separate rows. That simple chronology prevents a plausible story from replacing a reproducible account.
Evidence to retain: variant calls; phasing assumptions; allele definition source and version; copy number; diplotype alternatives; laboratory comment Link every item to its date, version, source, custodian and decision use. Record missing or inaccessible evidence explicitly; absence is not a negative result and must not be converted into reassurance.
Boundary and misconception: A familiar star-allele label can conceal uncertainty from phasing, hybrid genes or incomplete coverage. Test at least one competing explanation and state what evidence would change the conclusion. The pass condition is a bounded, reviewable claim—not a promise that the method removes every uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Change one unphased variant and show how more than one diplotype can fit the same genotype. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 3 check · what should survive an audit
- Name the decision, unit, time boundary and authority.
- Preserve both the supporting evidence and the strongest rejected explanation.
- State the next action, review date, stop rule and condition that would reopen the question.
CHAPTER 4 OF 13 · Add medicines and phenoconversion
4. Assign phenotype with a versioned rule
Map the diplotype to a predicted metabolizer or functional phenotype using the relevant gene-specific framework and retain any activity-score assumptions. The working unit for this chapter is one patient, one verified genotype or diplotype, one medicine decision and one current clinical context. Keep the question narrow enough that a reader can identify which person, sample, decision or evidence item is actually being compared, and which conclusion remains outside scope.
Worked example: A diplotype maps to intermediate metabolism under the current table; the record stores both the diplotype and mapping version rather than only the phenotype word. Write the starting information, the decision made from it, the later observation and the consequence in separate rows. That simple chronology prevents a plausible story from replacing a reproducible account.
Evidence to retain: diplotype; activity values; phenotype table; guideline version; uncertainty; date of interpretation Link every item to its date, version, source, custodian and decision use. Record missing or inaccessible evidence explicitly; absence is not a negative result and must not be converted into reassurance.
Boundary and misconception: Phenotype categories are models that can be revised; they should not erase the underlying genotype or imply measured enzyme activity. Test at least one competing explanation and state what evidence would change the conclusion. The pass condition is a bounded, reviewable claim—not a promise that the method removes every uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Apply a revised activity-score table and identify which patients need re-evaluation rather than silent relabelling. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 4 check · what should survive an audit
- Name the decision, unit, time boundary and authority.
- Preserve both the supporting evidence and the strongest rejected explanation.
- State the next action, review date, stop rule and condition that would reopen the question.
CHAPTER 5 OF 13 · Add medicines and phenoconversion
5. Detect phenoconversion from inhibitors, inducers and disease
Compare the genetically predicted phenotype with the person’s current functional context, including strong inhibitors, inducers and clinically relevant physiological change. The working unit for this chapter is one patient, one verified genotype or diplotype, one medicine decision and one current clinical context. Keep the question narrow enough that a reader can identify which person, sample, decision or evidence item is actually being compared, and which conclusion remains outside scope.
Worked example: A predicted normal metabolizer takes a strong enzyme inhibitor and functions more like a poor metabolizer for the affected pathway; the decision support records the temporary cause. Write the starting information, the decision made from it, the later observation and the consequence in separate rows. That simple chronology prevents a plausible story from replacing a reproducible account.
Evidence to retain: complete medication list; start and stop dates; interaction strength; adherence; liver and kidney function; inflammation or disease factors; monitoring Link every item to its date, version, source, custodian and decision use. Record missing or inaccessible evidence explicitly; absence is not a negative result and must not be converted into reassurance.
Boundary and misconception: There is no single consensus adjustment for every gene, inhibitor and clinical situation; phenoconversion should not be converted into an unqualified new genotype. Test at least one competing explanation and state what evidence would change the conclusion. The pass condition is a bounded, reviewable claim—not a promise that the method removes every uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Remove the inhibitor and define when the functional interpretation should be reviewed again. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 5 check · what should survive an audit
- Name the decision, unit, time boundary and authority.
- Preserve both the supporting evidence and the strongest rejected explanation.
- State the next action, review date, stop rule and condition that would reopen the question.
Worked drug–gene review
| Observation | Competing explanation | Protection | Changed-context test |
|---|---|---|---|
| Predicted poor metabolism | Genotype or inhibitor effect | Verify phenotype and medicine list | Remove inhibitor and reassess |
| Unexpected toxicity | Exposure, organ function or interaction | Clinical review and monitoring | Change dose or agent under prescriber |
| No response | Ultrarapid metabolism or non-genetic cause | Check adherence and indication | Repeat decision with new medicine |
| Alert fires repeatedly | Relevant or stale result | Reconcile active therapy and override reason | Test after guideline update |
The table is educational; individual treatment decisions belong to qualified clinicians using current product information.
CHAPTER 6 OF 13 · Evaluate evidence and work a case
6. Separate drug–gene from drug–drug–gene interactions
Map whether another medicine changes the same enzyme or transporter that connects genotype to exposure, and whether effects reinforce or oppose one another. The working unit for this chapter is one patient, one verified genotype or diplotype, one medicine decision and one current clinical context. Keep the question narrow enough that a reader can identify which person, sample, decision or evidence item is actually being compared, and which conclusion remains outside scope.
Worked example: A substrate medicine, a CYP inhibitor and a reduced-function genotype combine to raise exposure beyond what the genotype alone predicts. Write the starting information, the decision made from it, the later observation and the consequence in separate rows. That simple chronology prevents a plausible story from replacing a reproducible account.
Evidence to retain: medicine roles; dose; timing; pathway; interaction evidence; genotype; measured concentrations where available; clinical effects Link every item to its date, version, source, custodian and decision use. Record missing or inaccessible evidence explicitly; absence is not a negative result and must not be converted into reassurance.
Boundary and misconception: A static two-column drug–gene table can miss the multi-drug mechanism and should not replace a full interaction review. Test at least one competing explanation and state what evidence would change the conclusion. The pass condition is a bounded, reviewable claim—not a promise that the method removes every uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Add an inducer to the regimen and explain why direction and timing cannot be guessed from the genotype label. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 6 check · what should survive an audit
- Name the decision, unit, time boundary and authority.
- Preserve both the supporting evidence and the strongest rejected explanation.
- State the next action, review date, stop rule and condition that would reopen the question.
CHAPTER 7 OF 13 · Evaluate evidence and work a case
7. Use an evidence hierarchy for each gene–drug pair
Distinguish regulatory labelling, companion-diagnostic requirements, implementation guidelines, consensus recommendations and emerging association studies. The working unit for this chapter is one patient, one verified genotype or diplotype, one medicine decision and one current clinical context. Keep the question narrow enough that a reader can identify which person, sample, decision or evidence item is actually being compared, and which conclusion remains outside scope.
Worked example: A gene–drug pair appears in an evidence table, but the FDA page notes that inclusion does not always mean testing is recommended before prescribing. Write the starting information, the decision made from it, the later observation and the consequence in separate rows. That simple chronology prevents a plausible story from replacing a reproducible account.
Evidence to retain: current approved label; FDA association table; CPIC or other guideline; evidence grade; population; outcome; implementation date Link every item to its date, version, source, custodian and decision use. Record missing or inaccessible evidence explicitly; absence is not a negative result and must not be converted into reassurance.
Boundary and misconception: Association, altered pharmacokinetics and proven clinical benefit are different claims and must not be collapsed into one “precision medicine” promise. Test at least one competing explanation and state what evidence would change the conclusion. The pass condition is a bounded, reviewable claim—not a promise that the method removes every uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Move from a pharmacokinetic association to a toxicity endpoint and identify the additional evidence needed for action. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 7 check · what should survive an audit
- Name the decision, unit, time boundary and authority.
- Preserve both the supporting evidence and the strongest rejected explanation.
- State the next action, review date, stop rule and condition that would reopen the question.
CHAPTER 8 OF 13 · Evaluate evidence and work a case
8. Work a complete genotype-to-decision case
Build a fictional case from verified report through current medicines, phenotype, evidence source, options, prescriber decision and follow-up. The working unit for this chapter is one patient, one verified genotype or diplotype, one medicine decision and one current clinical context. Keep the question narrow enough that a reader can identify which person, sample, decision or evidence item is actually being compared, and which conclusion remains outside scope.
Worked example: A patient with a CYP2D6 reduced-function result begins a medicine affected by CYP2D6 while taking an inhibitor. The prescriber reviews alternatives, product information and monitoring rather than applying a genotype-only dose. Write the starting information, the decision made from it, the later observation and the consequence in separate rows. That simple chronology prevents a plausible story from replacing a reproducible account.
Evidence to retain: result provenance; assay limits; phenotype mapping; full medicines; indication; evidence source; options; shared decision; monitoring and outcome Link every item to its date, version, source, custodian and decision use. Record missing or inaccessible evidence explicitly; absence is not a negative result and must not be converted into reassurance.
Boundary and misconception: The educational case cannot determine a real dose. It demonstrates which evidence must be joined before the qualified prescriber acts. Test at least one competing explanation and state what evidence would change the conclusion. The pass condition is a bounded, reviewable claim—not a promise that the method removes every uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Change the inhibitor, organ function and therapeutic alternative one at a time and show which part of the reasoning changes. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 8 check · what should survive an audit
- Name the decision, unit, time boundary and authority.
- Preserve both the supporting evidence and the strongest rejected explanation.
- State the next action, review date, stop rule and condition that would reopen the question.
CHAPTER 9 OF 13 · Design decision support and monitoring
9. Design clinical decision support around the decision moment
Place the result where a prescriber can see it when selecting or ordering the relevant medicine, with concise rationale, alternatives, limits and source date. The working unit for this chapter is one patient, one verified genotype or diplotype, one medicine decision and one current clinical context. Keep the question narrow enough that a reader can identify which person, sample, decision or evidence item is actually being compared, and which conclusion remains outside scope.
Worked example: A pre-test alert explains that a result exists and offers a link to the verified report and current recommendation; it does not block every order automatically. Write the starting information, the decision made from it, the later observation and the consequence in separate rows. That simple chronology prevents a plausible story from replacing a reproducible account.
Evidence to retain: trigger logic; active medication; patient result; source version; recommended options; acknowledgement; override reason; alert performance Link every item to its date, version, source, custodian and decision use. Record missing or inaccessible evidence explicitly; absence is not a negative result and must not be converted into reassurance.
Boundary and misconception: An alert that fires too broadly creates fatigue; one that hides uncertainty can create false authority. Both failures should be measured. Test at least one competing explanation and state what evidence would change the conclusion. The pass condition is a bounded, reviewable claim—not a promise that the method removes every uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Move the alert from prescribing to dispensing and identify which actor, information and action are now available. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 9 check · what should survive an audit
- Name the decision, unit, time boundary and authority.
- Preserve both the supporting evidence and the strongest rejected explanation.
- State the next action, review date, stop rule and condition that would reopen the question.
Clinical decision-support gate
| Gate | Pass evidence | Remain open when |
|---|---|---|
| Identity and result | Patient and report verified | Provenance uncertain |
| Translation | Allele and phenotype version known | Coverage or copy number unresolved |
| Evidence | Medicine-specific source current | Association unsupported or outdated |
| Action | Prescriber sees limits and monitoring | Alert implies automatic treatment |
Silencing an alert and changing therapy are both decisions that require a recorded reason.
CHAPTER 10 OF 13 · Design decision support and monitoring
10. Preserve human review and justified overrides
Define who may override a recommendation, what reason is recorded, what urgent pathway exists and how outcomes are reviewed without punishing legitimate clinical judgment. The working unit for this chapter is one patient, one verified genotype or diplotype, one medicine decision and one current clinical context. Keep the question narrow enough that a reader can identify which person, sample, decision or evidence item is actually being compared, and which conclusion remains outside scope.
Worked example: A prescriber overrides because prior treatment response and current monitoring support continuation. The decision and follow-up are visible for later review. Write the starting information, the decision made from it, the later observation and the consequence in separate rows. That simple chronology prevents a plausible story from replacing a reproducible account.
Evidence to retain: user identity; recommendation shown; override option; reason; patient discussion; monitoring; later outcome; escalation Link every item to its date, version, source, custodian and decision use. Record missing or inaccessible evidence explicitly; absence is not a negative result and must not be converted into reassurance.
Boundary and misconception: Override rate alone is not a quality score; high or low rates can reflect poor alert targeting, workflow or case mix. Test at least one competing explanation and state what evidence would change the conclusion. The pass condition is a bounded, reviewable claim—not a promise that the method removes every uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Test the same alert in an emergency context where delay carries a different risk. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 10 check · what should survive an audit
- Name the decision, unit, time boundary and authority.
- Preserve both the supporting evidence and the strongest rejected explanation.
- State the next action, review date, stop rule and condition that would reopen the question.
CHAPTER 11 OF 13 · Design decision support and monitoring
11. Monitor clinical outcome and non-genetic explanations
Follow the intended benefit, adverse effects, adherence, exposure markers and alternative causes after a decision, with a plan for reassessment. The working unit for this chapter is one patient, one verified genotype or diplotype, one medicine decision and one current clinical context. Keep the question narrow enough that a reader can identify which person, sample, decision or evidence item is actually being compared, and which conclusion remains outside scope.
Worked example: Symptoms persist despite a genotype-informed change. The team checks indication, adherence, interacting medicines and disease progression rather than declaring pharmacogenomics ineffective. Write the starting information, the decision made from it, the later observation and the consequence in separate rows. That simple chronology prevents a plausible story from replacing a reproducible account.
Evidence to retain: baseline outcome; dose and timing; adherence; co-medications; laboratory or clinical monitoring; adverse events; patient-reported effects Link every item to its date, version, source, custodian and decision use. Record missing or inaccessible evidence explicitly; absence is not a negative result and must not be converted into reassurance.
Boundary and misconception: Genetics may explain part of variation, but response remains multi-factorial and a normal outcome does not validate every link in the chain. Test at least one competing explanation and state what evidence would change the conclusion. The pass condition is a bounded, reviewable claim—not a promise that the method removes every uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Change the disease state while keeping genotype and medicine fixed; identify which monitoring and causal claims must change. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 11 check · what should survive an audit
- Name the decision, unit, time boundary and authority.
- Preserve both the supporting evidence and the strongest rejected explanation.
- State the next action, review date, stop rule and condition that would reopen the question.
CHAPTER 12 OF 13 · Govern uncertainty, updates and transfer
12. Address ancestry, rare alleles and unequal evidence
Examine whether assay coverage, allele frequencies, function assignments and clinical validation represent the population in which the result is used. The working unit for this chapter is one patient, one verified genotype or diplotype, one medicine decision and one current clinical context. Keep the question narrow enough that a reader can identify which person, sample, decision or evidence item is actually being compared, and which conclusion remains outside scope.
Worked example: A common-panel assay has lower certainty for a person whose population carries functional alleles not included in the panel; the report’s residual risk is made explicit. Write the starting information, the decision made from it, the later observation and the consequence in separate rows. That simple chronology prevents a plausible story from replacing a reproducible account.
Evidence to retain: validation ancestry and sample size; covered alleles; rare and novel variants; function evidence; access to confirmatory methods; outcome equity Link every item to its date, version, source, custodian and decision use. Record missing or inaccessible evidence explicitly; absence is not a negative result and must not be converted into reassurance.
Boundary and misconception: Race is not a genetic proxy, and ancestry-related evidence gaps should not be used to deny useful testing or overstate certainty. Test at least one competing explanation and state what evidence would change the conclusion. The pass condition is a bounded, reviewable claim—not a promise that the method removes every uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Move the same panel across populations and identify which technical and clinical validation claims need rebuilding. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 12 check · what should survive an audit
- Name the decision, unit, time boundary and authority.
- Preserve both the supporting evidence and the strongest rejected explanation.
- State the next action, review date, stop rule and condition that would reopen the question.
CHAPTER 13 OF 13 · Govern uncertainty, updates and transfer
13. Govern reinterpretation, privacy and long-term reuse
Set rules for reanalysis, notification, data access, consent, retention, audit and correction because one result may affect many future medicines. The working unit for this chapter is one patient, one verified genotype or diplotype, one medicine decision and one current clinical context. Keep the question narrow enough that a reader can identify which person, sample, decision or evidence item is actually being compared, and which conclusion remains outside scope.
Worked example: A guideline changes three years later. The service identifies affected patients, verifies active therapy and sends a bounded update without silently rewriting the historical decision. Write the starting information, the decision made from it, the later observation and the consequence in separate rows. That simple chronology prevents a plausible story from replacing a reproducible account.
Evidence to retain: consent; result and source versions; access log; reanalysis trigger; affected cohort; notification; correction; retention; patient communication Link every item to its date, version, source, custodian and decision use. Record missing or inaccessible evidence explicitly; absence is not a negative result and must not be converted into reassurance.
Boundary and misconception: Long-term value increases governance burden: a reusable result can become stale, be misunderstood or expose sensitive information if provenance and access are weak. Test at least one competing explanation and state what evidence would change the conclusion. The pass condition is a bounded, reviewable claim—not a promise that the method removes every uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Transfer the record to a new health system and test whether provenance, consent limits and current interpretation survive. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 13 check · what should survive an audit
- Name the decision, unit, time boundary and authority.
- Preserve both the supporting evidence and the strongest rejected explanation.
- State the next action, review date, stop rule and condition that would reopen the question.
Verification lab: work the decision twice
Build one fictional CYP case with genotype, assay coverage, diplotype, predicted phenotype, inhibitor, active medicine and monitoring. Remove the inhibitor and show which interpretation expires and which genotype evidence remains.
Compare an FDA table entry, an approved product label and a CPIC guideline. Write three separate claims: what association is recognised, what prescribing information says and what an implementation guideline recommends.
Design an alert that fires only when the medicine, result and decision context align. Add a justified override, an urgent pathway and a monthly review for false alerts and missed opportunities.
Final review: distinguish a genotype result, a predicted phenotype, a phenoconverted state and an observed clinical response. If any two are treated as synonyms, rebuild the decision chain.
Limits and misconceptions
A genotype is not a prescription
Medicine, evidence, interactions, organ function, disease and patient preference remain part of the decision.
Predicted phenotype is not measured enzyme activity
It is a versioned interpretation that can be modified by current context.
Testing panels do not cover every functional allele
Coverage and residual uncertainty differ across genes, methods and populations.
Clinical decision support must remain reviewable
Alerts should expose evidence and limits, support justified overrides and be monitored for harm.
Questions readers ask
What is phenoconversion?
- It is a mismatch between genetically predicted and current functional phenotype, for example when an inhibitor or inducer changes enzyme activity.
Does an FDA association table mean everyone should be tested?
- No. FDA states that inclusion does not necessarily mean pre-prescribing testing is recommended unless required as a companion diagnostic.
What is a star allele?
- It is a named haplotype defined by a versioned combination of variants for certain pharmacogenes.
Can a result be reused for future medicines?
- Potentially, but the assay, interpretation and medicine-specific evidence must still be current and relevant.
Why record the full medication list?
- Other medicines can inhibit or induce the same pathway and change the functional meaning of a genotype.
Should patients change medicines from an online result?
- No. Medication decisions require a qualified prescriber using the verified report, current product information and the person’s clinical context.
Practise, verify and continue
Create a one-page genotype-to-decision map for a fictional patient. Include assay coverage, diplotype alternatives, phenotype version, interacting medicines, evidence source, options, prescriber decision, monitoring and a future reinterpretation trigger.
Apply the chain to an HLA-associated adverse-reaction risk rather than a metabolizer gene. The need for verified testing, medicine-specific evidence, clinical context and monitoring transfers, but the biological mechanism and action logic must be rebuilt.
Practical evidence check
Before implementation, write a one-page evidence map with the decision, unit of analysis, version or time boundary, inputs, outcomes, uncertainty, authority and stop rule. Attach one worked case that passes and one that fails. A reviewer should be able to reproduce the reasoning without relying on the memory of the person who built the system.
Run one changed-context rehearsal before approval. Alter a mechanism rather than a cosmetic detail, then identify which calculation, safeguard or governance gate must change. Record any conclusion that does not transfer. This short rehearsal is often the fastest way to expose a hidden assumption before it becomes an operational dependency.
Related eduKate reading
Phenotype harmonisation and case-definition drift
Ancestry-aware transferability and calibration
Cross-ancestry rare-variant calibration
Authoritative sources and scope
FDA Table of Pharmacogenetic Associations
Clinical Pharmacogenetics Implementation Consortium guidelines
CPIC serotonin-reuptake-inhibitor guideline and phenoconversion discussion
FDA companion diagnostics overview
These sources support the visible distinctions and evidence practices described here. They do not replace patient-specific counselling, laboratory validation, consent, local regulation or a clinician’s interpretation. All worked cases are fictional and no reader should start, stop or change medical care from this educational article.