Module 5 · Chapters 5 and 6
AEFI Investigation and Investigation of Deaths
Why and which AEFIs to investigate, the stepwise field investigation, the EPID number, cluster logic, verbal autopsy and post-mortem guidance.
5.1 Why investigate? · 5.2 Which AEFIs?
The ultimate goal is to make a valid diagnosis and establish whether the event is related to immunization — from the chronology, circumstantial evidence, cold-chain status, detailed medical history, epidemiological and laboratory evidence. This drives causality assessment and action. Investigations must find immunization errors (preventable) and recognise and communicate coincidental events to keep public confidence.
Objectives of investigating an AEFI case
- Confirm/refute the reported diagnosis or establish a valid diagnosis.
- Document the outcome.
- Identify the implicated vaccine(s) and the interval between vaccination and onset.
- Examine operational aspects of the programme/administration — even if the event seems product-related or coincidental, because an error may have triggered or worsened it.
- Decide if it was a single incident or part of a cluster (then follow the cluster protocol).
- Find whether similar events occurred in people not immunized with the implicated vaccine.
Which AEFIs must be investigated?
All reported severe and serious cases. Per WHO, serious events of known or unknown cause include:
- Death or life-threatening cases
- In-patient hospitalization or prolongation of hospitalization
- Persistent/significant disability or incapacity, or needing intervention to prevent permanent impairment
- Congenital anomaly/birth defect
- Cluster of AEFIs (including minor AEFIs)
- Events causing significant parental or public concern
Also: suspected immunization error; signals and events of special interest or events with newly introduced vaccines; significant events of unexplained nature or cause; events in the reportable AEFI list (Ch 4).
5.3 Steps in investigating AEFIs
- MO visits the case, confirms and fills the CRF (may phone the treating physician if hospitalised) — CRF to DIO within 24 h.
- DIO evaluates the CRF and submits it on SAFE-VAC immediately, not later than 48 h of notification.
- DIO starts planning the investigation once the case is verified as serious/severe; consults the District AEFI Committee where needed (especially clusters or community concern).
- Findings go into the CIF, submitted on SAFE-VAC within the timeline; local action as advised by the DAC.
- If everything is complete and the outcome known, the CIF is final; otherwise pending records are uploaded later.
- CRF, CIF and case records uploaded before 21 days. If the DAC has not met within 21 days, upload the CIF without committee details/signatures; once it meets, upload the updated CIF (diagnosis + signatures) in the “additional records” folder.
- Thorough, timely investigation lets the State AEFI Committee decide causality and follow-up.
5.3.1 Initial response of the health worker
- Inform the MO I/C of the PHC; start management including referral to the nearest facility.
- Assure parents/guardians that an enquiry is being initiated; collect basic event and demographic details for the MO.
- All used vaccine vials are preserved under cold chain at the nearest cold-chain point for 48 hours or until the next session; if an AEFI is reported, they are not discarded until further directions.
5.3.2 MO confirms and reports
Ensure treatment/referral; identify the session and obtain its records including the due list cum tally sheet; visit, examine and interview the family; collect data on patient, vaccines, session, manufacturer and batch numbers. Complete Sections A–D after framing a provisional diagnosis (may not yet be accurate) and send to the DIO within 24 h. Section D must describe a clear sequence of onset of signs/symptoms with the initial diagnosis.
5.3.3 Decision on investigation by the district · EPID number
DIO verifies Sections A–D, fills Section E, confirms onset date and vaccine details, and enters the case in SAFE-VAC — which generates the case ID (EPID number). The DIO writes it on the CRF, scans and uploads the signed CRF; the case reaches state and national levels simultaneously. Also include in the monthly HMIS report and the PHC AEFI register.
EPID formatIND (AEFI) – ST – DIS – YR – NUM (similar to AFP EPID numbers)
IND (AEFI) = India + condition · ST = state code (two letters) · DIS = district code (three letters) · YR = year of event onset (e.g. 14 for 2014) · NUM = serial number in that district that year.
Example: IND (AEFI)-UP-GZA-19-001 = first AEFI case of 2019 from Ghaziabad, Uttar Pradesh. COVID-19 cases use IND (CO-AEFI) from Co-WIN SAFE-VAC.
IND (AEFI) = India + condition · ST = state code (two letters) · DIS = district code (three letters) · YR = year of event onset (e.g. 14 for 2014) · NUM = serial number in that district that year.
Example: IND (AEFI)-UP-GZA-19-001 = first AEFI case of 2019 from Ghaziabad, Uttar Pradesh. COVID-19 cases use IND (CO-AEFI) from Co-WIN SAFE-VAC.
After deciding to investigate: inform DAC members, fix date/time; in death send CRF copies to the autopsy team and other stakeholders; contact a DAC member — preferably a paediatrician for children — to conduct the verbal autopsy. Only if appropriate, collect the implicated vial and/or logistics (e.g. AD syringes) and dispatch with the Laboratory Request Form (Annexure 13).
5.3.4 Investigation and data collection
- Team = members of the District AEFI Committee, preferably including a paediatrician, coordinated by the DIO; the PHC MO and the session’s health worker provide background and records.
- Collect pre-vaccination health status, treatment and hospital records; verbal autopsy and post-mortem in deaths.
- Verbal autopsy in all deaths even with post-mortem, except hospitalised cases in which the cause of death is certain.
- Drug inspector joins whenever a vaccine quality defect is suspected and in clusters.
- Use the CIF as a guide; obtain all medical files; fill gaps in CRF/CIF; identify other cases.
- DAC meeting synthesises evidence → provisional diagnosis; members sign the CIF; upload with treatment, investigations, discharge slip, post-mortem and/or verbal autopsy.
- Identify possible causes, especially immunization errors, and act on them and on state/national feedback.
WHO e-learning: openwho.org/courses/vaccine-safety-basics · investigation.gvsi-aefi-tools.org.
Be prepared for
- Session-site details ready before the team’s visit — due list cum tally sheet, vaccine details, stock registers.
- Visiting the immunization site, vaccine storage points, the patient’s residence and neighbourhood, and treatment centres.
- Interviewing the patient/parents, treating staff, the vaccinator and community members.
Field investigation — steps 1 to 5
Steps
- Confirm information in the report
- Investigate and collect data — patient, event, suspected vaccine(s), other people in the area
- Assess the immunization service — enquiries and observation
- Specimen collection when applicable — patient; vaccine and logistics
- Conclude the investigation
Sources of information
- Interaction with family/case
- Visit to hospital/clinic/ward; treating physician
- Visit to the immunization site; vaccinator and other frontline workers
- Review of vaccine storage and handling
- Community and other vaccination sites in the block/district if needed
- Investigation reports (blood, urine, CSF…)
| Step | What to do |
|---|---|
| 1. Confirm | Obtain medical file/clinical records/lab reports; check patient and event details; confirm CIF information; verify the time sequence of vaccination and event; obtain missing details; identify other cases to include. |
| 2a. Patient | Immunization history; previous medical history incl. prior similar reaction or allergies; family history of similar events. |
| 2b. Event | Chronological history; clinical description (pre-vaccination condition, interval to onset, sequence of manifestations, response to treatment); lab and other tests (X-ray, ECG); treatment and outcome. |
| 2c. Suspected vaccine(s) | Shipping from manufacturer to last major store; storage-point (refrigerator) conditions, documentation and transport to site; handling at site (cold boxes, ice packs, ambient exposure); VVM condition; time/date of opening the vial; time of each vaccination (beginning/end of session); label condition and date of previous use for reopened vials. |
| 2d. Other persons | Similar illness in the community — use a case definition, categorise, check vaccination status; details of other beneficiaries (from due list cum tally sheet) vaccinated from the same distribution point, centre, day and vial. |
| 3a. Service — enquiries | Dosage, person, site and technique; storage, distribution and disposal; reconstitution procedure; time between reconstitution and administration; number/type of immunizations and other medicines (e.g. vitamin A) in the session; staff training. |
| 3b. Service — observation | How vaccines are placed in the cold chain; other drugs stored with vaccines/diluents; vials without labels; batch numbers and expiry; contaminated-looking opened vials; directly observe reconstitution, drawing up, injection technique, needle/syringe safety, disposal; open vial policy followed. Sense the ambience — organised and child-friendly? |
| 4a. Specimens — patient | Table 5.1 (below; details in Ch 7). |
| 4b. Specimens — vaccine | Testing vaccine quality is RARELY needed — only on clear suspicion, not routine, never before a working hypothesis. DIO consults the SIO and conveys the SIO’s opinion to the DAC, which then decides on testing at CDL Kasauli. Tests: sterility (vaccine/equipment); chemical composition; preservative/adjuvant levels (e.g. aluminium); abnormal components (drug instead of vaccine); biological tests (abnormal toxicity). Some (e.g. abnormal components such as other drugs) can only be done in specialised labs outside the country. |
| 5. Conclude | Review epidemiological, clinical and lab findings; formulate a likely valid diagnosis; reach a provisional conclusion on cause; complete the CIF. DIO ensures investigation complete, DAC review of the sequence of events, and CIF + documents on SAFE-VAC within 21 days. |
Table 5.1 · Specimens from the patient
| Event | Specimen |
|---|---|
| Severe local reaction | Blood |
| Abscess | Swab, blood, pus |
| Lymphadenitis | Blood, FNAC |
| CNS symptoms with/without paralysis | CSF, blood and stool (ensure AFP reporting for all cases with paralysis) |
| Anaphylaxis | Blood |
| Toxic shock syndrome | Blood |
| Death | Post-mortem tissue specimens |
Documents to upload with the CIF
- Hospital records
- Pathology/microbiology (blood, CSF, urine)/radiology results
- Prescription/treatment record for this AEFI
- Prescription/treatment record for other illness
- Vaccine/diluent lab report (if tested)
- Syringe/other drug lab result
- In death: death certificate (if available), completed verbal autopsy form, preliminary and final post-mortem reports with cause-of-death report
- Any other relevant document
5.3.5 COVID-19 AEFIs · 5.3.6 Reported AEFI deaths
COVID-19 vaccination targeted a new epidemic disease in adults, so the CIF gained: (a) history of COVID-19 infection or contact with a positive person; (b) history of pre-existing illness/comorbidity/congenital disorder/treatment. Additional tests may be required by the epidemiological situation — e.g. RT-PCR/RAT for COVID-19 in all deaths and hospitalisations reported after COVID-19 vaccines.
A reported death needs a field investigation without delay (serious event, community concern, vaccine confidence); upload the CRF on SAFE-VAC; team of clinical, laboratory and forensic experts.
Death — key rules (5.3.6)
- A post-mortem within 72 hours gives the best information.
- In a medical college or District Hospital (if no MC), by a panel of at least a forensic medicine specialist, a pathologist and a paediatrician/medical specialist.
- Send viscera for histopathology (HPE) and toxicology per state guidelines; do videography.
- If post-mortem is impossible, try to get biological specimens from the body.
- A focused verbal autopsy by trained personnel even when post-mortem is done.
5.3.7 Investigating AEFI clusters
Cluster = two or more cases of the same adverse event related in time, place, vaccinator or vaccine. What counts as related “time” or nearby “place” depends on the event. A cluster may be in one district or linked to one vaccine, batch or vaccinator. Start by establishing a case definition and finding all cases. The DIO then:
- Identifies the cluster cases — when, where, which vaccines: detailed data on each patient; programme data (storage/handling); health workers’ preparation, handling, reconstitution and administration practices.
- Identifies common exposures — vaccine (name, lot); same vaccinator; same session; common recipient characteristics (age, nutrition, same family); data on unexposed people with similar symptoms.
Table 5.2 · Cluster characteristics pointing to a cause
| Cause | Cluster characteristics |
|---|---|
| Vaccine product-related | All cases got the same vaccine and no similar cases in the community; increased frequency reported from multiple settings |
| Vaccine quality defect | All cases got the same vaccine or lot; increased frequency from multiple settings |
| Immunization error | All cases got vaccine from the same health worker/facility; no other cases |
| Coincidental | Cases include unvaccinated people of the same area and age group |
| ITSR (anxiety) | Clusters of fainting after immunization are well recognised in mass campaigns |
Key considerations: usually an immunization error, much less commonly a quality defect. For new vaccines or new target populations, a cluster may represent a previously unrecognised product-related reaction. Knowing vaccine reaction rates and background rates is essential to judge signal strength.
Figure 5.1 · Identifying causes of an AEFI cluster
Q1
All cases from only one facility? (assume same lot used elsewhere) → Yes: immunization error. No: go to Q2.
Q2
All cases got the same vaccine or lot? → Yes: ask similar illness in others who did not get the vaccine? — Yes → coincidental event; No → immunization error (transport/storage) or vaccine quality defect. No: go to Q3.
Q3
Known vaccine reaction? → Yes: rate within expected? — Yes → vaccine product-related reaction; No → immunization error or vaccine quality defect. No: go to Q4.
Q4
Similar illness in others who did not get the vaccine? → Yes: coincidental event. No: immunization error, coincidental or unknown.
Case study (a) · toxic shock syndrome2018, state A: four deaths within 12 h of vaccination; all received MR and/or JE; fever, diarrhoea and vomiting within 2 h. Only these four had MR/JE; children who got pentavalent, OPV, rotavirus, IPV, PCV in the same session were fine. Cold-chain lapses: MR and JE used beyond the recommended time after reconstitution, not always on ice packs, and one syringe used to reconstitute multiple vials. Valid diagnosis: toxic shock syndrome. Cause: incorrect vaccine handling after reconstitution.
Case study (b) · wrong diluent2019, state C: three infants died after JE vaccine — hypotonia, difficult breathing, unconsciousness within 30 minutes; only JE recipients affected. Other injections including a muscle relaxant were stored with vaccines; vial not tested. Valid diagnosis: muscle paralysis including respiratory muscles. Cause: probable reconstitution of JE vaccine with a muscle relaxant.
WHO resources: investigation.gvsi-aefi-tools.org; WHO aide-memoire on AEFI. ITSU special investigation protocol (itsu.org.in).
Chapter 5 summary
- Investigations must be timely, comprehensive and methodical.
- Lab tests of vaccine/logistics only if indicated.
- Store used vials, syringes and biological samples properly in case needed.
- DIO starts at once and completes/upload within 21 days.
- Verbal autopsy in all deaths except hospitalised cases with complete records, diagnosis and cause of death.
- Cluster investigations must be immediate and detailed — many affected, community anxiety, media attention.
- Quality and completeness of investigation drive causality.
- New vaccines for new diseases may require changes and extra information.
Chapter 6 · Investigation of reported deaths following vaccination
- Vaccines are among the safest medical products, yet temporally associated deaths are often reported and damage the programme — each must be investigated.
- Most reported deaths are coincidental. Truly vaccine-caused deaths are very rare: product-related (anaphylaxis, viscerotropic disease), quality defect (incompletely attenuated live agent) or immunization error (vial contamination → toxic shock syndrome).
- Sudden deaths at home without prior illness are traumatic and raise safety questions. Many are due to previously undiagnosed illness; rarely accidental injury or environmental factors (e.g. hypothermia); increasingly genetic, metabolic or cardiac disorders are recognised — identifying them protects other children and maintains trust.
- Investigation includes history, circumstances of death, medical examination, post-mortem and interviews with doctors, ANM, ASHA and others around the family in the preceding days. Collect all records: OPD slips, in-patient records, discharge summary, investigation reports, death summary; consult post-mortem doctors.
- Multidisciplinary; delay loses evidence. Examine the site of death for trauma, fall from crib/cot or accidental overlaying by parents; photograph if the body has not been moved.
- Separate verbal autopsy forms exist for children and adults.
6.1 Verbal autopsy
- Conducted by a trained MO (DIO) with the customised AEFI form; a paediatrician/physician should assist to probe for a clinical diagnosis.
- In privacy, with empathy; avoid leading questions.
- Mandatory in all deaths, irrespective of post-mortem — including deaths after hospitalisation (then upload all related documents with the CIF).
- Objective: detailed information on the terminal event, especially brought-dead cases, home/non-hospitalised deaths, insufficient records or no possible clinical diagnosis. Where records give the cause, it still adds risk factors and chronology.
6.1.1 Children (<18 y) — Annexure 10
- Personal: name, DOB/age, sex, parents, caregiver, grandparents/others in house, address.
- Event: date/time and place of death; events before death in order.
- Immunization: site, route, location of session, vaccinator, date/time, local reaction, past AE, status of other children vaccinated at the same session.
- Overview: who, what, where, when; team visits the site; verbatim account from 24 h before (or earlier) to declaration of death.
- Environment, family structure, stressors (discord, parental alcohol/drug abuse), other children’s health, overcrowding/unhygienic conditions → overall impression in the summary.
6.1.2 Adults — Annexure 11
- Past history — pre-existing illness, hospitalisation, medication, investigations.
- Comorbid conditions and long-term medication.
- Relevant obstetric and gynaecological history in women.
- Systemic involvement — respiratory, cardiovascular, GI, CNS, genitourinary.
- COVID-19 infection/post-COVID vaccination details.
- Treatment/medication before death.
- Risk factors — smoking, alcohol, drug abuse, occupation.
6.2 Guidance on conducting post-mortem
MedicolegalIt is desirable that all deaths following vaccination be converted into medicolegal cases (MLC). Avoid issuing a death certificate; report to the police as an MLC. Once it is an MLC, family consent is not necessary for post-mortem.
- Do a post-mortem and related lab work whenever feasible. Routine district post-mortems are not oriented to pathological causes — use the AEFI post-mortem format (Annexure 12).
- Perform as soon as possible to avoid tissue damage, artefacts and lysis of internal organs. DIO ensures a detailed history accompanies the post-mortem form.
DIO ensures
- A medical college in the district or a neighbouring district is identified; if not possible, the district hospital.
- Team with a forensic medicine expert or pathologist; if not feasible (transport/notification delay), any doctor under their consultation/supervision.
- Videography (or photography) arranged for external and internal examination.
When a death is reported
- HW/MO informs the DIO before the body is disposed of.
- DIO informs the police on priority to make it an MLC.
- Convince the family of the advantages of post-mortem — approval smooths the process.
- Inform forensic/pathology experts of the identified medical college; share formats beforehand.
- Arrange transfer of the body.
While conducting
External examination first (age, sex, height, weight, overall condition; for infants length and head circumference), then internal.
| External examination (Table 6.1) | Internal examination |
|---|---|
| Clothes · stains (mud, blood, urine, stools) · identity marks · body orifices · finger and toe nails · injuries/surgical interventions · rigor mortis · post-mortem staining · decomposition/other changes | Three cavities: skull/cranium · thorax · abdomen |
| Sample | Preservation | Sent to |
|---|---|---|
| Organs for histopathology | Whole organ in 10% formalin | DH or medical-college pathology, through the police |
| Blood for microbiology | Sterile test tubes at 4°C | Medical-college microbiology, through the police |
| Viscera/gastric contents (suspected poisoning) | Saturated common salt solution | Forensic Science Laboratory (FSL) |
| Blood for toxicology | Sodium fluoride (NaF) | FSL |
- Send the full post-mortem report or a note on what is suspected and tests needed; guide the police on what was preserved and where to send it.
- The report must state the opinion on cause of death, estimated time between death and post-mortem, pathology found, and articles preserved.
- Follow-up: preliminary report goes with the CIF; chase HPE and chemical reports; upload the final report on SAFE-VAC/CoWIN-SAFE-VAC.
6.3 Additional findings to confirm cause of death
| Suspected cause | Look for |
|---|---|
| Anaphylaxis | Rashes or small petechial haemorrhages on skin and internal organ surfaces; laryngeal oedema; pulmonary oedema; fluid in thoracic or abdominal cavities |
| Aspiration | Gastric contents in distal and terminal bronchi — gastric contents only in the trachea are not evidence |
| Overlaying / suffocation | Cardinal signs of asphyxia: deep congestion of organs, pulmonary oedema, bluish lips and nails |
| TTS (adenovector COVID-19 vaccines) | Thrombus in coronary or pulmonary vessels |
| Nothing significant found | Preserve: viscera in saturated salt (toxicology); lungs, liver, spleen, kidney, brain in formalin (HPE); blood in NaF; blood in sterile tubes at 4°C (microbiology) |
| Pregnant woman | Uterine and foetal size → gestation; signs of miscarriage/haemorrhage; history of abortions/poor obstetric history; retain foetal tissue for DNA analysis if needed |
6.4 Impact of post-mortem quality and timeliness on causality
Challenges: (a) delayed post-mortem/histopathology reports; (b) incomplete or illegible reports; (c) unknown whether preserved viscera were sent for HPE/toxicology; (d) opinion on cause of death not given/reserved; (e) discrepancies among CRF, CIF and post-mortem report.
Mitigation by the State AEFI Committee: review and feedback on post-mortem quality; the committee’s forensic expert/pathologist guides district autopsy surgeons (via the DIO); include a forensic expert in the District AEFI Committee. DIO–police coordination ensures timely histopathology reports.
Chapter 6 summary
- Investigate every death to judge relation to vaccination.
- Investigations are multidisciplinary; document early before details are forgotten.
- Verbal autopsy in all deaths irrespective of post-mortem.
- Convert deaths after vaccination into MLCs.
- Post-mortem in a medical college (district/neighbouring) else the district hospital, by a panel incl. post-mortem surgeon/pathologist/forensic specialist and a paediatrician or physician; videography.
- The AEFI post-mortem is not merely to exclude unnatural death — detailed gross and histopathology of all organs matter.
Quick recall
{{c.q}}
{{c.a}}
Self-test · Chapters 5 and 6
Score: {{score}} / {{total}} · answered {{done}}
{{q.num}}. {{q.text}}
{{q.verdict}} {{q.expl}}
You scored {{score}} of {{total}} ({{pct}}%).