Medical Laboratory Technology and Diagnostics Case Studies
Healthcare5 Case StudiesUpdated Sep 2026
Laboratory work sits between the bedside and the decision, and much of its risk lives in the gaps between them. These five cases follow specimens, results, and the people who move them through hospital and reference settings. Each one centers on a fictional organization with its own staffing pattern, protocols, and reporting habits. Read them for patient needs, the evidence available at each point, how the team coordinates when information arrives unevenly, and which local protocols apply. The discussion questions are meant to be argued, not answered quickly.
Case Study 1
A slow trend that never crossed one threshold
Harrowgate Regional Hospital, a 410-bed acute care hospital with one shared laboratory serving four medical wards and an emergency department.
The situation
Over three days, Ruth, a patient in her seventies, was admitted to the medical ward with a respiratory illness and a history of kidney disease. Her routine blood results moved in small steps. None of the individual values crossed the critical threshold set by the laboratory, and each report went to the ward with the standard comment. Danielle, her ward nurse, noted that Ruth was less alert in the afternoons and was eating less at each meal. Tomas, the registrar, reviewed the results each morning but was covering two wards that week and saw the printed trend only once. Priya, the day shift technologist, noticed the sequence across reports while preparing the weekend batch.
The challenge
Priya had no authority to order investigations, and her role was to report accurately, not to interpret symptoms. The ward had no single document that showed the trend beside the nursing observations. Local escalation rules tied a call to a critical value, and Ruth's values were not critical. The team had to decide whether a pattern without a threshold breach justified a call, and who should make that call.
Analysis
The first question is what Ruth needed at that point. She was not in immediate danger on any single result, but her needs were changing in ways the reports did not capture well. Teams grounded in Medical Laboratory Technology and Diagnostics usually start by separating what the numbers show from what the bedside observations suggest, and then asking whether the two agree. Here they did. Falling intake, reduced alertness, and a slow drift in kidney-related values pointed in the same direction, even though each value sat inside its reference interval. The laboratory had reported accurately. What was missing was a shared view of the sequence.
Local protocols gave the team a clear path once the question was framed correctly. The hospital's escalation policy allowed any clinician or laboratory staff member to request a senior review when a trend raised concern, even without a critical value. Reflex and repeat testing were permitted under the laboratory's own rules, but they would only answer the question if someone with clinical authority asked it. The strongest evidence for escalation was the combination of trend and observation, and the policy named that combination explicitly. Coordination depended on whether the trend reached the person who could act on it before the next scheduled round.
- 1Three-day sequence noticedLaboratory reviewer saw small changes across reports
- 2Observations addedNursing notes showed falling intake and alertness
- 3Trend escalatedCall made under the trend clause, not a critical value
- 4Senior reviewRegistrar reassessed the patient and documented reasoning
Actions taken
- Priya phoned the ward charge nurse with the three-day trend, read out in plain values and collection times, and asked for senior review under the trend clause of the escalation policy.
- The registrar reviewed Ruth at the bedside within the hour, compared her observations with the trend, and recorded the reasoning in the notes so the next shift could follow it.
- The laboratory added a trend comment to later reports so that each new result was read against the prior sequence, and the duty scientist confirmed the pattern showed on the ward view.
- The ward began a short daily huddle for patients with three or more linked changes, using a shared trend sheet so the sequence no longer depended on one person's memory.
Results
Ruth was reviewed the same day the trend was raised, and the senior team changed her plan that afternoon. Her condition improved over the following days, though many factors shaped that outcome and the call was only one of them. The audit later found that the trend had been visible in the system for nearly two days before anyone raised it. Over the next quarter, the ward logged eleven trend escalations, and nine led to a documented review within the hour. The laboratory also received two complaints that its trend comment was unclear, and it revised the wording.
Questions for discussion
- What threshold should trigger a call from the laboratory when no single result is critical, and who should own that judgment?
- How would the ward have handled Ruth's trend if the nursing observations had been recorded only on paper?
- Which parts of the escalation policy would you change after reading this case, and what evidence would support the change?
Case Study 2
A pending request lost between two shifts
Brackenridge Community Medical Center, a 260-bed hospital whose general medical units share a single phlebotomy queue and an electronic order list.
The situation
Marcus, a man in his fifties recovering on a general medical unit, had a repeat specimen requested by the night registrar after an earlier result was judged borderline and worth confirming. The request was entered early in the night shift, but the specimen was not collected until much later because the phlebotomy queue was long. The night nurse, Aisha, wrote the pending request on her own sheet but not in the typed handover document the day team used for planning. At seven, the day nurse, Gerald, took over six patients from a document that did not mention the outstanding request.
The challenge
The gap was small and easy to miss. The request existed in the system, the specimen existed in the queue, and each person held a partial picture. The unit's handover template had no field for pending laboratory actions, and the day team relied on the typed sheet rather than the electronic order list during the first hour. Nobody had been careless, but the process allowed the pending action to fall between two records.
Analysis
Marcus depended on a result that had not yet been produced, which meant the evidence he needed was still in transit. The earlier value was the only evidence on file, and it was borderline enough to need confirmation. A repeat specimen only helps if it is collected, transported, and resulted in time to matter for the decision. Pending results should be treated as active clinical items, not background administrative detail. For the day team, the real question was not whether the result would be good or bad, but whether anyone was placed to notice that it had not been collected or reviewed.
The hospital's handover standard, used across the medical units, required every pending investigation to be named. The template had fields for observations and planned care, but none for specimens awaiting collection or results expected during the next shift. Local protocol also asked that the requesting clinician confirm the order was visible on the electronic list. That step had been skipped during the busy night, and the day team had no prompt to check. The safeguard existed on paper, but the template did not make it hard to forget.
- Night, early shiftRepeat specimen requested and entered in the system
- Night, mid shiftQueue delay pushed collection later than planned
- Night, end of shiftRequest noted on a personal sheet, not the template
- Morning, first hourDay team planned without seeing the open request
- Morning, after the roundRequest found on order list and specimen collected
Actions taken
- Add a required pending investigations field to the handover template, listing each specimen requested but not yet collected or resulted, with the time it was ordered and who owns the follow-up.
- Require outgoing and incoming nurses to check open laboratory requests on the electronic order list together during handover, and initial the sheet once the check is complete.
- Ask the phlebotomy team to flag specimens that are waiting more than a set period in the queue, so that ward staff can see delays before the time for a decision has passed.
- Audit handover sheets each month for a sample of at least forty transfers, counting pending investigations that were missing, and report the rate to the unit quality group.
Results
The collection delay was resolved that morning once the day team checked the order list and found the request. The repeat specimen was taken before midday, and the registrar reviewed the result before the afternoon round. No harm was recorded, though the team could not confirm how much the delay would have mattered had the result been more urgent. In the next quarter, the audit found missing pending items on roughly one in five sheets before the change and on fewer than one in twenty after it. The phlebotomy flag was used forty-one times in its first month.
Questions for discussion
- Which pending items should a handover template require, and which should be left to the clinician's judgment?
- How would you balance the time spent checking the order list against the risk of a rushed handover?
- What would a near miss in this unit look like in the audit data, and how would you separate it from a normal delay?
Case Study 3
Triage choices during an analyzer outage
Ashgrove Regional Laboratory, a shared service processing samples for three hospitals and a cluster of outpatient clinics.
The situation
Ashgrove Regional Laboratory processes roughly 1,400 samples a day for three hospitals and a cluster of outpatient clinics. In the second week of the quarter, one of its two automated analyzers failed and could not be repaired for at least five working days. Demand did not fall. A respiratory surge had begun at one of the hospitals, and the emergency department was sending more urgent requests than usual. The section lead, Nadia, had a staffing roster built for two analyzers, with one senior technologist on the bench, two junior staff, and a rota that could not stretch without moving people from other sections.
The challenge
Three questions pulled against each other. Which samples could wait without harm, which ones needed a result within a set time frame, and how could the section avoid running the remaining analyzer so hard that quality slipped? Manual workarounds existed but were slower and needed the senior technologist, who was already the only person trained to review flagged results. Every choice moved risk somewhere else, and the regional quality policy required that any change in priority be documented.
Analysis
The first step was to sort demand by what each result was used for, not by where it came from. Emergency samples and those tied to an active inpatient decision needed priority, but a routine outpatient panel could sometimes wait a day without changing care. The evidence for this came from the laboratory's own turnaround data and the clinical notes attached to each request, not from the order's label alone. The useful question is what decision a result will support and by when, because that answer shows how much delay is tolerable.
The regional resource policy set out a tiered priority system with clear rules for when a section could suspend routine work. Tier one covered time-critical requests from emergency and critical care areas, and the policy allowed the section lead to pause routine testing for a defined period once a backlog passed an agreed threshold. Changes had to be communicated to the three hospitals and logged, so that clinicians knew which results would be late and why. Coordination mattered as much as bench capacity. Without a shared message, hospital teams would keep chasing results that were deliberately deferred.
Actions taken
- Triage every incoming request against the tiered priority list at intake, recording the tier on the request so bench staff and clinicians see the same priority without asking.
- Move the senior technologist onto review of flagged results only, and assign one junior staff member to run routine batches overnight when the remaining analyzer is least in demand.
- Send a written notice to each hospital and clinic stating which routine tests would be deferred, the expected delay, and the contact for urgent exceptions, and log the notice in the regional record.
- Review turnaround times daily for the first two weeks and lift the routine pause as soon as emergency results return to the agreed time frame, rather than waiting for the repair date.
Results
Emergency and inpatient results stayed within their agreed time frames across the outage, with one exception on the second day when a sample arrived damaged and had to be recollected. Routine outpatient results were delayed by up to two days in most cases. Two clinics reported confusion over which tests were paused, so the laboratory revised its notice to list them by name. The analyzer returned to service in the third week, earlier than the supplier's estimate, and the routine backlog cleared within four working days. The section logged one complaint, which was resolved the same day.
Questions for discussion
- Which samples should be counted as time-critical in a regional policy, and who should be able to change that list?
- What signals would tell a section lead that a routine pause is lasting longer than it should?
- How should a laboratory weigh the cost of a notice that clinics find confusing against the cost of an unannounced delay?
Case Study 4
Rejected specimens and an uneven collection network
Thornbury Health Partners, a network of three hospitals and eleven outpatient collection sites sharing one laboratory information system.
The situation
Thornbury Health Partners ran outpatient specimen collection at eleven sites over the previous six months. Specimen rejection rates varied widely between them. The best site rejected about one collection in a hundred, while the highest rejected more than one in ten. The network's quality lead, Kwame, was asked to explain the spread before the next committee meeting. The indicators available included rejection reasons, recollection rates, time from collection to receipt, and time from receipt to result. Staff at the high-rejection sites were experienced but used different labeling routines, and two sites had changed courier timing partway through the period.
The challenge
The indicators were anonymized, so Kwame could not contact individual collectors directly, and the sites differed in patient mix as well as process. A high rejection rate could reflect labeling habits, late transport, a difficult patient population, or a recording problem in the information system. Changing a process across eleven sites at once would be fast but could mask which change mattered. The committee wanted a plan that would show cause as well as effect.
Analysis
Every rejected specimen meant a patient had to come back, sometimes after a delay that mattered for their care. Kwame's evidence was aggregate, so the first step was to read the rejection reasons instead of treating the headline rate as the problem. Usually the useful split is between reasons that point to the collection step, such as labeling or volume, and reasons that point to transport or recording. The reason codes showed that most rejections at the high sites were labeling errors, while the lowest sites recorded more transport delays. That pattern pointed to different causes needing different fixes.
The network's quality policy required any change to be piloted, measured against a baseline, and reviewed by the committee before wider adoption. Two indicators were agreed in advance: the rejection rate per site and the share of rejections carrying a labeling reason. Process owners at each site were asked to record any change in courier timing or staffing, so that effects could be separated from routine variation. Without that record, a fall in rejections after a courier change could easily be mistaken for the effect of a labeling fix.
- Pilot one step Labeling checklist tested at the two highest-rejection sites
- Measure together Rejection rate, labeling share, and transit time tracked weekly
- Compare changes Courier and staffing shifts logged alongside each result
- Adopt or adjust Network-wide use only if pilot falls and comparisons do not
Actions taken
- Pilot a standard labeling check at the collection point at the two sites with the highest labeling rejections, using a checklist agreed with the local collection leads.
- Record courier timing and staffing changes at every site in a shared log, so that any shift in rejection rates can be compared with what changed at the same time.
- Measure rejection rate, labeling share, and collection-to-receipt time weekly for ten weeks at the pilot sites and two comparison sites, and report all four in one table.
- Adopt the checklist network-wide only if the pilot sites fall below the agreed threshold and the comparison sites show no matching fall, otherwise revise the change and repeat the pilot.
Results
The pilot sites reduced their labeling rejections over the ten-week period, and the comparison sites changed much less. Overall rejection rates at the pilot sites fell from more than one in ten to about one in thirty. Transport delays did not improve, which confirmed that the courier schedule needed a separate review. The committee approved the checklist for network-wide use but kept the courier question open. Two sites reported that the checklist added a little time at busy hours, and the team adjusted the order of steps rather than dropping the check.
Questions for discussion
- Which indicators would you trust most when sites differ in patient mix, and what would you need to adjust for them?
- How would you protect a pilot from being contaminated by other changes happening at the same sites?
- When a checklist adds time at busy hours, who should decide whether the trade-off is acceptable?
Case Study 5
A sister, a consent form, and a disputed disclosure
Calder Reference Laboratory, a regional specialist laboratory providing molecular and genetic testing to hospitals and clinics in its catchment.
The situation
Helen, a patient in her forties, had a specialist test ordered by her referring physician, Ruben, as part of an assessment of family history. The laboratory's report, issued in the usual way to Ruben, showed a result with implications for close relatives. Helen had signed a consent form that described the test but did not state clearly how results might be shared with family members. Two days later, Ruben called the laboratory's clinical liaison, Osei, asking for the full report to be shared with Helen's sister, who had also been seeing him. When the liaison team reached Helen by phone, she said she did not want her sister told yet.
The challenge
The request came from a clinician with a legitimate role in care, but the patient had asked for limits on sharing. The consent form was ambiguous, and the laboratory's own policy did not state who could decide when a patient and a clinician disagreed about disclosure. Osei had to decide what to release, what to hold, and whom to involve, while keeping the patient informed and not making a choice on her behalf.
Analysis
Helen's needs were about control over information that concerned her family as well as herself. The result was genuine evidence with real consequences for relatives, so the question was not whether it mattered, but who should be told, in what order, and with what support. Her duty of confidentiality sat alongside the clinician's legitimate need to act on the result for her care. Reading the consent form closely showed that it did not cover family disclosure, which meant the laboratory could not assume permission. Osei treated the gap as a reason to pause, not as a reason to choose a side.
The laboratory's disclosure policy required that requests to share results beyond the patient and the ordering clinician go to the privacy lead and the medical director. Local guidance also asked staff to record the patient's stated wishes and to offer her a session with a genetic counselor before any further disclosure. Escalation was not a refusal of Ruben's request. It was a way of making sure the decision was made by people with the authority and information to make it, with Helen still involved. Clear communication with Ruben about timing and reasons was part of the same duty.
Actions taken
- Release the report to the ordering physician only, as the original consent allowed, and record that no further sharing is authorized until the review is complete.
- Refer the sharing request to the privacy lead and the medical director the same day, attaching the consent form, the call notes, and Helen's stated wishes.
- Arrange a call between Helen and the laboratory counselor to explain what the result could mean for relatives and which options she has for who is informed and when.
- Ask Ruben to hold the conversation with the sister until the review is finished, explain the reason in plain terms, and agree a decision date with a named contact.
Results
The review found the original consent form ambiguous on family disclosure and concluded that the laboratory could not share the report with the sister without Helen's written agreement. Helen met with the counselor and chose to tell her sister herself, with support, in the following week. Ruben accepted the delay once the reasons were explained, though he asked for a clearer consent template for future cases. The laboratory revised its form to separate general test consent from family disclosure, and added a field for stated preferences. The matter closed without a complaint.
Questions for discussion
- When a patient's wishes and a clinician's reasons for disclosure conflict, what should the laboratory's role be?
- How would you redesign a consent form so that family implications are clear before the test is carried out?
- What should be documented about a disagreement like this, and who should be able to read that record later?
Learning Resources
Healthcare CASE STUDIES
China, India, United States, Indonesia, Pakistan, Nigeria, Brazil, Bangladesh, Russia, Japan, Mexico, Philippines, Vietnam, Egypt, Turkey, Germany, United Kingdom, France, Italy, Canada, South Korea, Australia, Spain.