Skip to main content

Look-Alike, Sound-Alike Medicines: A Practical LASA Review for Clinics

A useful LASA programme starts with the medicines your clinic actually buys, stores, searches and hands over. Identify confusing pairs by name, packaging, strength, formulation or route; then change the point where the confusion occurs and test the change. Shelf separation, distinctive lettering and warnings can help, but none replaces verification of the intended medicine against the actual product.

By Bractis · Updated · 15 min read

Indian clinic owners, pharmacists and authorised dispensary staff reviewing look-alike, sound-alike medicine risks in local stock and prescribing searches.

Build a local confusion-pair register

WHO's 2023 LASA publication describes confusion from similar medicine names and appearances, including packaging, shape, colour, doses and strengths. It recognises brand-to-brand, brand-to-generic and generic-to-generic confusion. A national or international list can prompt a review, but it cannot describe every pack on a particular clinic's shelf.

Start with the actual purchase catalogue, recent stock changes and near misses. Ask which two items could be selected instead of one another and at which step. 'All medicines are high risk' is too broad to guide a change; 'these two strengths share a bin and nearly identical pack fronts' identifies an actionable problem.

Include discontinued stock still present in cupboards and patient-returned packs awaiting a disposition decision. A product removed from the ordering screen may remain physically reachable. Equally, a newly purchased brand can create a visual conflict even when the active ingredients in the formulary did not change.

Source: WHO: Medication safety for look-alike, sound-alike medicines, 2023

Locate the confusion before choosing the control

Name confusion during a spoken request needs a different response from a selection error between two strengths on a screen. Record whether the failure occurred at ordering, receiving, shelf replenishment, picking, labelling or patient handover. Otherwise a new sticker may be placed on the shelf while the wrong product continues to enter the basket upstream.

Treat the medicine name as only one part of identity. Compare active ingredient or combination, strength, dosage form, route and pack presentation with the authorised request. A familiar brand name does not establish that a tablet, modified-release tablet and liquid are interchangeable.

At goods receipt, compare the supplied product with the ordered item before putting it in the existing bin. A supplier substitution or redesigned package should trigger a fresh pair review. Staff who work only the evening shift need the change explained, because their memory may still match yesterday's pack.

Use a control matched to the actual failure

The following original matrix is a design aid. Each control should be tested with the affected staff and products; it is not evidence that a particular label format prevents errors in every setting.

On a narrow screen, scroll the table horizontally to read all columns.

LASA failure points and practical controls to test
Confusion mechanismControl to evaluateResidual question
Similar spoken namesRead back full identity and clarify the authorised requestWas the intended formulation and strength confirmed?
Two strengths share a binSeparate, clearly identified locations and a pack-to-request checkCan a restocking error put the wrong strength back?
New packaging resembles another productReview receipt, storage and picking with actual packsDid both morning and evening staff see the change?
Search truncates distinguishing textExpose full name, strength, formulation and routeIs the accepted selection preserved in the final output?
Loose strips lose product informationRetain traceable identity through the dispensing processCan the handover still identify batch and expiry where needed?
Route confused from container shapeSeparate route-specific storage and explicit instructionsCan the user explain what device or route applies?
Warning is routinely dismissedReview why the alert fires and what decision it requiresDoes the warning identify this pair rather than everything?

Make separation visible without creating a second naming system

Place the distinguishing information where the item is chosen, not only inside a policy folder. If the confusing feature is strength, an oversized brand logo with a tiny strength label has not addressed it. Use the approved product identity and avoid local nicknames that differ between prescriber, pharmacist and patient.

Distinctive lettering or colour can draw attention, but the full text must remain readable. Colour alone is unreliable when the pack changes, the label is photocopied or the user cannot distinguish it. Do not let a supplementary label cover the original strength, route, batch or expiry information.

Physical separation also needs a restocking rule. Two bins placed apart are useful only if their contents are verified when replenished. If a cabinet has limited space, document the workable separation and checking process rather than adding a 'LASA' sticker to a mixed basket and declaring the issue closed.

Test medicine search and output as well as the shelf

A screen may show several similar entries with the distinguishing suffix cut off. Test with the actual viewport used at reception or in the consulting room, and inspect the accepted medicine row, prescription and dispensing view. A clear search result is insufficient if the final printed line loses the formulation.

Search aliases can improve retrieval but should not collapse different products into one entry. A spelling variant may refer to the same medicine; a strength or release-form difference may not. Keep the catalogue owner responsible for deciding which aliases are legitimate and for reviewing supplier-driven additions.

Use fictional product labels in a demonstration rather than inventing a real hazardous pair or entering patient prescriptions. Ask how a correction propagates to an unsent draft and how an already issued document is handled. A catalogue edit should not rewrite the historical identity of a medicine previously recorded.

Run a small retrieval exercise and report its denominator honestly

In a fictional training exercise, four staff each complete six labelled retrieval tasks using empty demonstration packs: 24 attempts in total. Three attempts initially select the wrong demonstration item, and two of those are caught before the mock handover. That is three selection errors and one error reaching the mock handover, not three harmed patients.

Record which pair and step caused each problem. Repeating the same exercise after moving a bin may show whether staff can use the new layout, but it is not a clinical error-rate study. Participants know they are being tested, and a small set cannot establish performance during a busy evening OPD.

Include a new-staff and restocking scenario. If only the person who designed the shelf can retrieve correctly, the arrangement is not self-explanatory. Revise the label or process that produced uncertainty and retain the unresolved issue instead of converting every attempt into a training-complete tick.

Worked example: two medicine identities, three pack designs

Fictional dispensary review L-12 in Mysuru uses empty demonstration packs for a made-up ingredient Z. Product A is a 5 mg immediate-release tablet; product B is a 50 mg immediate-release tablet. B has both an older pack and a redesigned pack. These are training identities, not real medicines or prescribing choices. The tenfold strength difference is a label comparison, not a dose-conversion instruction.

The exercise has two product identities and three pack designs. B's old and new packs represent the same verified product identity in this scenario, but their packaging evidence should remain distinguishable. A and B must remain separate catalogue choices. A search alias that collapses their shared ingredient name into one selectable item would remove the very distinction the clinic needs to preserve.

The redesigned B pack resembles A's front panel. The receipt team notices the resemblance before restocking, but the current search list also truncates the strength on the consulting-room display. Moving the shelf addresses only one part of this problem. Record the physical-pack issue and the software-display issue separately, with an owner for each. Do not close the second because staff have already attended a shelf briefing.

Match the requested item to the actual full label rather than recognising its colour. For this mock case, the authorised request names A; a participant selects the new B design. Capture both identities and the stage at which the discrepancy is detected. A description such as 'wrong brown box' becomes useless when packaging changes again. The full product and presentation information makes the near-miss record reviewable.

Preserve old-pack stock in the review until it is actually removed through the authorised process. Changing the online catalogue image does not change the packs already on the shelf. Likewise, an evening staff member who has only seen the old design needs access to the revised information. A short labelled comparison at receipt and selection can help; the comparison should keep original strength, form and route visible.

On a narrow screen, scroll the table horizontally to read all columns.

L-12: controls for the same pair at three separate selection points
Selection pointObserved problemSpecific change to testEvidence still needed
Goods receipt and replenishmentNew B design resembles A; old B stock remainsCompare ordered and delivered identities; separate and identify the actual binsCorrect replenishment by someone who did not design the layout
Consulting-room searchStrength disappears from the visible candidate labelExpose full identity before accepting a candidateCorrect accepted row and patient output, including the strength
Mock patient handoverParticipant recognises pack colour rather than the requestCompare the chosen pack with the full authorised requestDetection before handover, recorded separately from first selection

Source: WHO: Medication safety for look-alike, sound-alike medicines, 2023

Show the errors a successful mock handover can hide

Extend the earlier four-person, six-task exercise. The original round contains 24 attempts: 21 correct first selections, two initial errors intercepted before mock handover and one initial error that reaches the mock handover. Thus three attempts contain an initial selection error, even though only one finishes with the wrong demonstration pack. The two recovered attempts remain useful evidence of where the process needed a second chance.

After the changes, the same four participants repeat six tasks each. In this fictional retest, 22 first selections are correct and two are initially wrong; both mistakes are intercepted before the mock handover. All 24 handovers finish with the intended demonstration item. Calling that '100% accurate selection' would hide the two errors. The first-selection figure is 22/24, about 91.7%, while the final-handover figure is 24/24.

The initial round's corresponding first-selection figure is 21/24, or 87.5%. The difference is one additional correct first selection out of 24, about 4.2 percentage points. It is not proof that the clinic's real-world medicine-error rate improved by that amount. The same people know the task, the items are empty training packs and the sample is small. Familiarity, observation and changed conditions all limit the comparison.

Record the two remaining errors by their actual cause. If both involve a replenished bin, further search-screen editing will not resolve them. If staff selected correctly but the printed line lost the distinguishing text, that is an output defect even when the physical pack was right. Keep the stages visible so a single overall score does not conceal the next useful action.

The retest has not assessed new joiners or the evening team in this example. Add a separate exercise for those roles and a replenishment task before treating their readiness as established. Do not add their names to a training-complete list because a colleague described the change. Record assistance honestly: a correct choice after a prompt is different from an independent first selection.

On a narrow screen, scroll the table horizontally to read all columns.

L-12: initial selection and final handover use different numerators
MeasureOriginal round: 24 attemptsRetest: 24 attemptsInterpretation
Correct first selection212287.5% versus about 91.7%; one extra correct first selection
Initial selection errors32Errors still occurred in the retest
Errors intercepted before handover22Recovery is recorded, not erased
Wrong item at mock handover10No wrong final item observed in these 24 retest attempts
Correct final mock handover2324A 100% final-handover result is not 100% correct initial selection

Document the change, its limits and the next person to test it

The L-12 record should identify which physical and digital changes were actually made, who checked them and which products remain in scope. 'LASA labels applied' is too broad when the main failure was a hidden strength on screen. Save the relevant before/after label or display evidence using invented records in a demonstration, and make the exact corrected field easy to locate.

An owner should also check the endpoints. A full search result can become an abbreviated accepted medicine row, a truncated print line or an unclear dispensing label. These are separate representations of the selected identity. A correction to an unsent draft should be visible before issue; a change to the catalogue must not silently alter a historical prescription. Existing issued-document correction procedures still apply.

Keep the mock exercise separate from any real patient incident. If a real wrong item may have been supplied, the responsible clinical and pharmacy teams need the actual product, recipient, use history and response. Improving a bin or recording a training result does not complete that clinical follow-through. L-12 contains no patient prescriptions, dispensing event or harm outcome.

The next review trigger should be concrete: another supplier pack, a new strength, changed display width, a changed bin location or an observed near miss. An annual review date alone will miss a relevant change that happens tomorrow. Retain both resolved and unresolved issues so a new pharmacist can understand why the controls exist instead of inheriting unexplained stickers.

The filled note is a modest completion claim. It says which training tasks were observed and what remains untested. The aim is to make a confusing selection harder to make and easier to catch, then examine the observed result. No local colour scheme, lettering convention or small exercise is presented as WHO certification or a validated prediction of patient safety.

Use near misses to improve the process, not hide the count

A near miss is valuable because it shows where the process nearly failed and where it recovered. Record what was selected, what was intended, who detected the mismatch and what made detection possible. Avoid blame-focused language that discourages reporting ordinary slips and confusing packaging.

Separate the immediate patient or stock response from the system change. If a wrong item may have reached a patient, clinical assessment and appropriate contact take priority. Moving the shelf later does not complete that response.

Review the register when stock, software search or staffing changes. A fixed annual list may miss a new procurement problem for months. The owner should be able to explain why a pair remains on the list, which controls apply and what evidence supports closing a resolved local issue.

Fictional example: a supplier changes the pack, not the ingredient

A supplier delivers the same authorised medicine in redesigned packaging that resembles another strength already stocked. The receiving team identifies the visual similarity before replenishing the old bin. They separate the locations, revise the local pair record and test retrieval with evening staff. The change is recorded as a local control improvement; the clinic does not claim that a new label eliminates all medicine-selection errors.

Local LASA Pair Review Worksheet

Use these prompts in your clinic’s own notes. Record the result, responsible person and next action for each item.

Real pair
Which two actual products could be confused, and what feature makes that likely?
Failure point
At which step must the control change the selection or handover?
Control test
Can someone unfamiliar with the revised shelf or search identify the correct item?
Follow-through
Are patient consequences, stock corrections and system changes tracked separately?

Questions doctors ask

Is there one complete LASA list for every Indian clinic?

No list can cover every local brand, presentation, supplier change and storage arrangement. Use published examples to prompt a review of actual stock and near misses.

Are coloured stickers enough to prevent a LASA error?

They are only a supporting cue. Keep full product identity readable and verify the chosen item against the authorised request; colour may change, be missed or reproduce poorly.

Should different strengths of the same medicine be reviewed?

Yes, if their names, packs, locations or screen entries make selection confusion plausible. The register should describe the specific local failure rather than only list different ingredients.

Can a mock retrieval exercise prove the clinic's medication error rate?

No. It can expose confusing pairs and test a proposed control. Report attempts and detection stages as training observations, not patient harm or measured clinical outcomes.

When should the LASA register be revisited?

Revisit it after relevant purchasing, packaging, stock-layout or search changes and after near misses. Assign an owner so the list remains connected to current practice.

Can all mock handovers be correct while medicine-selection errors still occur?

Yes. A wrong first selection can be caught and corrected before handover. Report initial selections, intercepted errors and final handover outcomes separately; a perfect final result in a small exercise does not establish error-free routine practice.

Should redesigned packaging automatically create a new medicine identity?

No. Verify the ingredient, strength, formulation, route and presentation first. A packaging change may need a new local risk review even when the medicine identity is unchanged. Different strengths or formulations must not be merged merely because their names resemble each other.

Sources and scope

Updated on . This article combines the cited references with Bractis’s original workflow analysis and illustrative examples. Product capabilities depend on the configured scope. No independent clinical or legal review is claimed.

Try the workflow
with Bractis.

Compare two fictional similar products through purchasing, search, shelf retrieval and prescription output; show how a corrected selection and its history remain understandable.

Request a clinic demo