Drug Allergy Records: Separate Unknown from None Known
Record uncertain allergy histories with source-preserving examples, dated assessments and checks across prescribing, referral and export.
Medication reconciliation compares the medicines a patient is actually using with the available prescriptions and clinical record, then resolves discrepancies through clinician review. A collected list is the starting point, not the finished result. In an Indian OPD, the process must handle outside prescriptions, hospital discharge papers, medicine strips, caregiver reports and incomplete information without turning any of them into an automatic new order.
For doctors, clinic assistants and owners designing medicine-history intake and the handover into prescribing.
A medicine history records information from the patient and available sources. Reconciliation adds comparison, assessment of differences and a documented current plan. AHRQ’s MATCH toolkit describes reconciliation around comparing the current regimen with clinical orders at transitions. WHO also identifies medication discrepancies during transitions of care as a safety concern.
For a small OPD, the practical distinction is simple: reception may have copied every line on a discharge paper accurately, yet the doctor may still need to determine what the patient is taking and whether the current plan differs. A scanned document, a completed intake form and a reconciled list are three different outcomes.
Keep those outcomes visible in the record. Otherwise a field labelled “current medicines” can contain a mixture of old orders, patient statements and new decisions with no indication of which is which. The next clinician then has to repeat the investigation or risk interpreting historical information as the present plan.
Patients may consult more than one doctor, attend a hospital between OPD visits, buy non-prescription products or change what they take without the clinic’s old record reflecting it. Ask what information is available today and identify the source of each item. A caregiver’s account can be useful, but it should be attributed rather than presented as a directly verified fact.
Medicine strips and photographs can help identify a product, while a prescription explains what was ordered at a particular time. Neither automatically proves present use. Ask about reported use separately and preserve disagreement between sources for the doctor to assess.
On a narrow screen, scroll the table horizontally to read all columns.
| Source | Useful information | Limitation to retain |
|---|---|---|
| Patient or caregiver account | What the person reports taking and difficulties following the plan. | Names, strengths or dates may be uncertain; identify who supplied the account. |
| Medicine strip or container | Product information visible on the actual packaging. | Possession alone does not establish current use, the prescribed amount or the intended duration. |
| Outside prescription | What another prescriber recorded at that encounter. | A later change may exist; it is not automatically the present plan. |
| Hospital discharge summary | Instructions associated with a transition from hospital care. | Clarify subsequent consultations and reported use since discharge. |
| Previous local EMR list | Historical orders and assessments available in the clinic. | Check the dates and status; do not equate “last recorded” with “currently taking.” |
A patient may answer “yes” when asked whether they are taking the medicines on a prescription, then describe a different routine when invited to explain it. Start with the actual routine and available evidence. An original opening question could be: “Please tell me which medicines you are using at home now, including anything from another doctor or bought without a prescription.”
A Hindi question to adapt is: “कृपया बताइए, घर पर आप अभी कौन-कौन सी दवाएँ ले रहे हैं?” It asks which medicines the person currently uses at home; it does not ask them to confirm an old list. Have staff fluent in the patient’s preferred language check the wording used in practice, and ask about relevant outside sources without assuming every patient can read a strip or medicine name.
If the answer is “the tablet in the blue strip,” keep that description linked to the packaging or photograph available. Colour is not enough to establish product identity. Record an unknown name as unknown and ask for an appropriate clearer source. An assistant should not make the list look complete by selecting a familiar medicine with similar packaging.
Use neutral follow-up when reported use differs from an earlier order. The patient may describe an outside consultation, difficulty obtaining a product or an instruction they understood differently. Record the explanation as reported. The doctor needs that context to assess the discrepancy; an administrative label such as non-compliant can conceal it.
Give staff a collection task they can complete reliably: record the item as reported, identify its source and mark what is unclear. They should not be required to select a precise medicine from a search result when the name is unknown. An uncertainty field is more truthful than a guessed code.
Keep the source document accessible beside the entry. If a photograph is unreadable, record that limitation and seek a clearer source. If the patient says they are not using an item on the old list, preserve that statement rather than silently deleting the earlier order. The doctor needs to see the discrepancy.
Do not ask an assistant to resolve duplicate-looking entries, change a dose or decide whether a medicine should be continued. Similar names can refer to different products, and different names can conceal a relationship that needs clinical assessment. The staff handover should make these questions easy to find.
“Please review medicines” is a weak handover when the clinic already knows what differs. A stronger note states the conflicting sources: “The discharge paper lists item A; the patient reports not starting it,” or “The old chart and the new prescription show different strengths.” This preserves the question without proposing a treatment answer.
Common discrepancies include an item missing from one source, unclear current use, different recorded instructions, an uncertain product identity and a document that cannot be dated. A useful review queue names the issue, the available evidence and the responsible clinician. It does not classify every difference as an error before review.
The clinician may resolve a difference immediately, request more information or document that uncertainty remains. Preserve that outcome. A process that forces every row into “resolved” at the end of intake encourages false certainty and makes later follow-through difficult.
In this fictional encounter M-52, the patient brings a previous clinic prescription, a hospital discharge sheet and medicine packaging. Their account of current use differs from the documents. Item labels A, B and C below are placeholders for this information exercise; no medicine, strength or regimen is being recommended.
Compare each candidate item across sources before deciding whether records belong to the same product. Where identity is not established, retain separate candidate entries and the question linking them. Matching a name or merging two rows is not the same as confirming the medicine and its present use.
On a narrow screen, scroll the table horizontally to read all columns.
| Candidate item | Available sources and disagreement | Question for clinician review |
|---|---|---|
| A: identity established, instructions differ | The previous clinic order and discharge sheet show different instructions; the patient describes following the earlier copy. | Which instruction reflects the intended current plan, and what explanation is needed after the clinician decides? |
| B: present on discharge sheet | The patient reports never starting it. No subsequent instruction resolving that difference is available. | What is the reason for the difference, and what does the clinician decide now? Do not mark it currently used or discontinue it automatically. |
| C: partly legible strip | A product is physically present, but its full identity and relation to either prescription are uncertain. | What further evidence is needed? Keep the candidate unresolved instead of guessing a code or deleting it as a duplicate. |
For item A, a generic “medicines reviewed” note leaves the next clinician guessing whether the competing instructions were considered. The useful record connects the evidence, question and decision. The clinician supplies the treatment decision; the example below deliberately does not choose one.
For item C, review may conclude that more evidence is needed. That is a recorded clinical disposition with an unresolved information task, not permission to invent the product identity. Distinguish a case reviewed with an open question from a case that has not reached the clinician at all.
After the decision, inspect what the system creates. The historical order, patient-reported use and new approved prescription should remain distinguishable. A history import that silently generates active prescription rows defeats the comparison even if the source documents remain attached.
After review, the record should show the clinician’s current decision and what information it was based on. Retain the earlier sources so a future reviewer can understand a change. Avoid using one editable text box for both the patient’s report and the final plan; each later edit can erase how the decision was reached.
Check that the patient-facing prescription, referral summary and next-visit view reflect the intended status. A medicine marked historical in one screen should not automatically reappear as an active order in another. Similarly, a reported item awaiting review should not become a signed prescription merely because it appears in the medicine-history section.
Explain the reviewed plan through the clinic’s normal clinical communication process. When a new document supersedes an earlier one, the patient or appropriate caregiver needs to know which record to use. The information task is not finished merely because the clinician has saved a new list.
Make a missing-information task concrete: what is needed, who will request it, where the result should go and when the responsible clinician wants it reviewed. “Bring old papers next time” may be insufficient if the doctor needs the information sooner; the clinician should determine the appropriate timing.
Link the incoming document to the unresolved question when it arrives. Uploading it into a general attachment folder does not ensure the reviewer knows it is there. Cover arrangements also matter: if the original doctor is away, the team should know the agreed clinical route rather than leaving a task assigned to an unavailable person.
At handover, distinguish “information requested,” “received,” “reviewed” and “plan communicated.” These states describe observable work. They allow the team to identify whether delays are occurring at the patient contact, document receipt or clinician review stage.
After M-52, an authorized caregiver supplies a clearer image for item C. Link the image to that candidate and record when it arrived. The image’s capture or receipt date is not necessarily the date of the prescription or the date the patient started using the product. Preserve those separate facts where known.
Route the newly available evidence to the responsible clinician or agreed cover. Staff can record that the image is legible or that an outside document has arrived; those observations do not establish the current treatment plan. If the clinician’s review changes an already issued instruction, use the prescription-correction process and the appropriate communication route.
At the next visit, keep the previous reconciliation as history and establish what has changed since it. A single click copying the previous list should not also assert that current use was confirmed today. This matters when the patient has seen another provider between clinic visits or arrives without the relative who supplied the earlier account.
Use a fictional previous clinic prescription, an outside discharge list and a patient account that differs from both. Ask the demonstrator to retain all three, flag one uncertainty and show the doctor recording a decision. Then open the referral and the next-visit prescribing view.
The acceptance question is not “Does the software have medication reconciliation?” It is whether the doctor can distinguish reported use, historical orders, unresolved discrepancies and the reviewed plan throughout the workflow. Ask how that distinction survives export and migration as well as a normal consultation.
For a pilot, record the number of sampled encounters in which a discrepancy was made explicit and assigned for review. Do not infer prevented adverse events from a small software exercise. A useful early measure is whether the information reaches the right person in a form they can act on.
In a fictional follow-up visit, a patient brings a hospital discharge sheet, an older local prescription and two strips. The patient says one discharge item was never started and cannot identify a partly damaged strip. The assistant records these differences with their sources, leaving the unknown item unresolved. The doctor reviews the available information, documents a plan and assigns a task for the missing evidence. When a clearer document arrives later, staff connect it to that task and route it for review. At no point does copying the discharge sheet automatically create a new prescription.
Use these prompts in your clinic’s own notes. Record the result, responsible person and next action for each item.
Record the patient’s account alongside the discharge instruction, including any reason they report, and send the discrepancy to the clinician. Do not automatically label the item currently used, delete it from history or supply a start/stop decision at intake.
It may resolve some missing identity evidence, but receipt alone does not establish current use or the treatment decision. Link it to the original question and route it for the required clinical review, keeping any resulting patient communication visible.
No. It preserves one historical source. Reconciliation requires comparison with current reported use and available information, followed by clinician assessment of differences and a documented plan.
An assistant can collect and organize information within the assigned role. Clinical assessment of discrepancies and treatment decisions belong to the responsible clinician. The software should show that handover instead of treating completed data entry as completed review.
Extracted text needs verification against a sufficiently clear source. OCR does not establish current use, prescribing intent or whether a partially visible product name is correct.
Make the apparent duplication visible with its sources and refer it for clinical assessment. Silent removal can erase the very discrepancy the doctor needs to investigate.
Define completion around the clinician-reviewed plan and the required communication, while retaining explicit ownership of any unresolved information. A populated list or uploaded document alone is not sufficient.
Updated on . The article addresses collection, comparison and handover of medicine information. It does not tell staff or patients to start, stop, substitute or change a medicine. No independent clinical or legal review is claimed.
US toolkit overview checked 28 October 2026 for reconciliation as a comparison process at care transitions. The M-52 case, intake wording and OPD handover are original adaptations, not a treatment protocol or validated history instrument.
Publication overview checked 28 October 2026 for the importance of discrepancies and information quality at care transitions; the full report was not re-reviewed. No clinical outcome or product-effectiveness estimate is inferred.
Record uncertain allergy histories with source-preserving examples, dated assessments and checks across prescribing, referral and export.
Write focused clinic referral letters with an attachment manifest, patient instructions, returned-advice review and redirection history.
Verify outside lab reports with worked OCR errors, missing-page handling, distinct dates and amended-report reviews for Indian clinic EMRs.
Bring three fictional sources that disagree. We can examine how history, unresolved questions and the doctor’s reviewed plan are represented across intake, prescribing and referral.
Request a clinic demo