A real-world pharmacovigilance analysis of hepatitis B vaccine using the U.S. Vaccine Adverse Event Reporting System (VAERS) database

Abstract

Hepatitis B vaccines (HBVs) are widely used duo to their high clinical use and mild effects. However, as post-marketing data accumulate, several serious adverse events (SAEs) following HBV have been reported. Currently, quantitative studies based on real-world data are lacking, and information on their adverse events is limited. Adverse reaction signals of the HBV were mined and analyzed using the U.S. Vaccine Adverse Event Reporting System (VAERS) to provide a reference for the safe clinical use of this vaccine. Multiple statistical methods, including the reporting odds ratio (ROR) method, the Medicines and Healthcare Products Regulatory Agency (MHRA) method, and the Bayesian Confidence Propagation Neural Network (BCPNN) method, were utilized to identify signals of HBV-associated adverse reactions, and positive signals consistent with designated medical events (DMEs) were singled out for focused comparison and discussion. Analysis of 54,136 HBV-related adverse events (AEs) identified 254 positive signals across 22 System Organ Classifications (SOCs), with General disorders and administration site conditions being the most common. Three potential positive new signals consistent with Preferred Terms (PTs) were identified in DME: Aplastic anaemia, Dermatitis exfoliative, and Haemolytic anaemia. This study suggests that HBV has a potential risk in terms of causing Aplastic anaemia, Dermatitis exfoliative, and Haemolytic anaemia. Some subtypes of Aplastic anaemia, Dermatitis exfoliative, Haemolytic anaemia are autoimmune diseases, and immunization may stimulate potential autoimmune genetic predisposition, people with autoimmune diseases or a family history of hereditary immune diseases should be monitored after receiving HBV. Health professionals should be contacted to take measures to help if anaemia, palpitations, and high fever occur.

Introduction

Hepatitis B virus infection is globally prevalent. According to the "Global progress report on HIV, viral hepatitis and sexually transmitted infections, 2021" released by the World Health Organization (WHO), the general population prevalence of HBsAg in 2019 was 3.8%, with approximately 1.5 million new infections and 296 million chronic HBV infections. Additionally, 820,000 people died from liver failure, cirrhosis, or hepatocellular carcinoma (HCC) caused by HBV infection¹. Vaccination against hepatitis B provided over 95% protection against infection. WHO estimates that prior to widespread vaccination in 2000, the proportion of children under 5 years old infected with HBV was around 5%, decreasing to less than 1% in 2019².

A “vaccine adverse event,” also referred to as an “adverse event following immunization” (AEFI), is an adverse health event or health problem that occurs following or during administration of a vaccine. The Council for International Organizations of Medical Sciences (CIOMS) defines an AEFI as “… any untoward medical occurrence which follows immunization and which does not necessarily have a causal relationship with the usage of the vaccine^3,4. Due to the widespread use of the HBV and its large clinical application, with the main recipients being newborns and high-risk populations, its safety has always been a focus of attention. The HBV has a good safety record in infants, children, adolescents, and adults. The most commonly reported AEs are injection site pain and fever^5,6. However, with the continuous accumulation of post-market data, the adverse reaction section of the HBV label has also been continuously supplemented, including severe adverse reactions such as Arthus reaction, Angioedema, and Erythema multiforme^7,8. Additionally, most studies to date have been statistical analyses of AEs of the HBV through systematic reviews or retrospective studies^9–11, lacking quantitative research based on real-world data.

The VAERS is a national system that collects information from the entire U.S. population. Due to its large and diverse reporting base, VAERS is capable of rapidly identifying potential safety concerns and detecting rare adverse events associated with vaccines^12–14. In addition, a vaccine’s safety signal is information from one or more sources, including observations and experiments. This information indicates a new potential causal association or a new aspect of a known association between a drug product and an adverse event. It was determined that the information was sufficient to justify verification action¹⁵. Finally, the lopsided method remains unmatched when it comes to the timely detection of safety signals for rare and unpredictable adverse reactions¹⁶. Therefore, this study intends to use the proportion imbalance method on the basis of VAERS database to excavate and analyze the adverse reaction signals of hepatitis B vaccine as a whole, so as to explore the safety of hepatitis B vaccine and provide basis and reference for vaccination.

Material and methods

Source and handling of data

This study was conducted using data from the Vaccine Adverse Event Reporting System (VAERS) in the United States. The VAERS database was established in 1990 and is jointly managed by the Centers for Disease Control and Prevention (CDC) and the Food and Drug Administration (FDA) to identify potential safety issues with licensed vaccines in the United States^17–19. This includes the detection of new, unusual, or rare vaccine AEs; monitoring increases in known AEs; identifying potential patient risk factors for specific types of AEs; and evaluating the safety of newly licensed vaccines²⁰. The VAERS database uses the Medical Dictionary for Regulatory Activities (MedDRA) preferred term (PT) codes to record the symptoms of adverse events, with each VAERS ID containing up to 5 symptoms²¹.

Our study extracted all vaccine adverse events uploaded in the United States between 1990 and 2023 from the VAERS database. The data were imported into Microsoft® Excel® 2021 MSO for processing. Initially, reports originating from within the United States were selected by excluding entries marked as ‘Foreign’ in the ‘STATE’ field, as non-domestic VAERS data were not included in this analysis. Subsequently, completely duplicate reports and those lacking critical information, such as age and gender, were removed to ensure data quality and relevance for the study. After data cleaning was completed, the ‘HEP B’ was used to screen in the ‘VAX_NAME’ field. To eliminate the potential influence of interactions between different vaccines, reports involving the co-administration of hepatitis B vaccine with other vaccines were excluded. This was achieved by restricting the ‘vaccine type’ field to ‘HEP’ and the ‘VAX_NAME’ field to ‘HEP B’ alone. Specifically, individuals who received combination vaccines such as HEP A + HEP B (TWINRIX), HIB + HEP B (COMVAX), DTP + HEP B (NO BRAND NAME), and DTP + HEP B (TRITANRIX) were excluded. Only reports of individuals who received the hepatitis B vaccine alone were retained for analysis. These reports were then compared with all other adverse event reports in the database, which served as the reference group. Statistical analysis and signal detection were performed using SAS 9.4 (Statistical Analysis System Institute Inc.).

In this study, MedDRA version 26.0 was used to encode each AE’s PT, which was then mapped to their respective SOC for further exploration. Specific SOCs (such as product issues, social circumstances, surgical and medical procedures, etc.) were excluded from the analysis to ensure identified signals are indicative of ADRs caused by the vaccine and to eliminate the impact of unreasonable or irrelevant medication reports.

Descriptive analysis

Based on the collected reports, descriptive statistics was conducted to classify the AE reports of HBV by gender, age, dosage, onset time, reporting year, and outcome category. In addition, the distribution of reports on serious AEs was statistically analyzed year by year. Among them, reports with outcomes classified as death, permanent disability, life-threatening, hospitalization, prolonged hospital stay, congenital anomalies, or birth defects were categorized as SAEs.

Statistical analysis

Disproportionation analysis stands as the primary data mining method in the current landscape of ADR signal detection. This method hinges on the principle of comparing the disparity between the occurrence frequency of a specific drug-event combination and the background frequency. A signal is deemed to have manifested when it significantly surpasses a predefined threshold (refer to Table 1). At this juncture, the total number of AEs for each PT constituted the 'A' value. Subsequently, 'B,' 'C,' and 'D' values were calculated based on all the AEs across different drugs.

Table 1.

Two-by-two contingency table for measure of disproportionality.

	Number of target adverse reaction reports	Number of other adverse reaction reports	Total
Target drug	A	B	A + B
Other drugs	C	D	C + D
Total	A + C	B + D	A + B + C + D

To minimize the generation of signals with high false positives, this study employed three methods for signal detection: Reporting Odds Ratio (ROR), MHRA, and Bayesian Confidence Propagation Neural Network (BCPNN)^22–24. A positive signal is generated when all three methods detect a positive result. Additionally, a stronger signal is indicated by higher values of ROR, PRR, IC-2SD, among others. The specific algorithms and threshold values are shown in Table 2.

Table 2.

Measure of disproportionality and signal generation criteria.

Methods	Formula	Signal standard
ROR		A ≥ 3 95%CI > 1
MHRA		A ≥ 3 PRR ≥ 2 ≥4
BCPNN	= IC-2SD = E(IC)-2	A ≥ 3 IC-2SD > 0

, are the Dirichlet distribution parameters, , ,, are Beta distribution parameters, IC-2SD is the lower limit of IC 95%CI, hypothesis , , E is the expected number of cases.

SD standard deviation.

Designated medical events (DME) list screening

In 2016, the European Union introduced the DME list, which consisted of 62 PTs, that is, a collection of PTs regarded as ‘inherently serious’ and considered to be ‘often medicine-related’. Crucially, the purpose of the list of DMEs is to prioritize adverse events in signal detection, and it is described by the EMA as a ‘safety net that ensures signals are not missed’ ²⁵. This study focused on evaluating serious and specific safety events associated with HBVs. The screening process is based on the DME list to identify valuable positive signals related to HBVs, followed by further in-depth analysis in conjunction with the corresponding System Organ Classes (SOCs).

Results

Descriptive overview of cases

A total of 54,136 AE reports related to HBVs was identified in this study, including 201,978 adverse events (Table 3). Among all AEs, males (36.69%) were less prevalent than females (56.63%). In terms of age distribution, children under 1 year old (28.79%) accounted for the highest proportion. Regarding dosage, excluding cases with unknown information, most patients experienced adverse events after receiving 1 dose (29.55%) 2 doses (25.45%) or 3 (20.66%) doses of HBV. Furthermore, over half of the patients had an unknown outcome (57.87%), with the highest number of patients experiencing emergency outcomes (31.78%). Half of the patients developed adverse events on the same day after vaccination, while the majority of the rest experienced symptoms within a month, with an average onset time of 34.9 days.

Table 3.

Characteristics of AE reports associated with HBV from VEARS between 1990–2023.

Characteristics		N	Proportion (%)
Total		54,136	100%
Gender	Male	19,862	36.69%
	Female	30,657	56.63%
	Unknown	3617	6..68%
Age(year)	≤ 1	15,638	28.79%
	> 1, < 2	2650	4.88%
	≥ 2, < 12	6287	11.61%
	≥ 12, < 22	7418	13.70%
	≥ 22, < 42	12,933	23.81%
	≥ 42, < 62	7815	14.39%
	≥ 62, < 120	1049	1.93%
Administration dose	1	15,997	29.55%
	2	13,776	25.45%
	3	11,183	20.66%
	4	1667	3.08%
	5	332	0.61%
	6	133	0.25%
	7 +	41	0.08%
	Unknown	10,966	20.26%
	NA	41	0.08%
Outcome	Died	872	1.61%
	Life threatening	1093	2.02%
	Emergency room	17,206	31.78%
	Hospitalized	3801	7.02%
	Prolonged hospitalization	323	0.60%
	Disability	1326	2.45%
	Congenital anomaly or birth defect	1022	1.89%
	healthcare professional office/clinic visit	1302	2.41%
	Emergency room/department or urgent care	1250	2.31%
	Unknown	31,330	57.87%
Onset time(day)	0	26,022	/
	> 0, < 30	18,853	/
	≥ 30, < 60	985	/
	≥ 60, < 90	443	/
	≥ 90, < 120	307	/
	≥ 120	1631	/
	average	34.9	/

AE adverse event, HBV hepatitis B vaccine, VAERS vaccine adverse event reporting system.

As illustrated in Fig. 1, we normalized and scaled the total number of hepatitis B reports to a standard population size (per 100,000 VAERS reports). The results indicate that the annual frequency peaked in 1999, followed by 1995 and 2011. Between 2000 and 2015, the annual frequency remained relatively stable, while a general declining trend was observed after 2016.

Fig. 1.

Yearly frequencies of overall HBV AEs and SAEs reports..

Among the 54,136 AE reports collected related to HBVs, the main associations were observed with General disorders and administration site conditions, Nervous system disorders, Skin and subcutaneous tissue diseases, and gastrointestinal system disorders. Common symptoms included fever, fatigue, chills, seizures, drowsiness (MedDRA term), erythematous rash (MedDRA term), diaphoresis (MedDRA term), and nausea (Table 4).

Table 4.

Distribution of AEs in the report (n ≥ 1000).

SOC	PT (n ≥ 1000)
General disorders and administration site conditions (56,922)	Pyrexia(9623); Asthenia(2698); Chills(1344); Crying(2142); Drug ineffective(2854); Fatigue(1206); Injection site erythema(2270); Injection site hypersensitivity(2154);; Injection site oedema(1936); Injection site pain(3192); Injection site swelling(1434);; Malaise(1734); Oedema peripheral(1203); Pain(3464); Screaming(1761); Unevaluable event(1230)
Nervous system disorders (19,674)	Convulsion(1568); Somnolence(1213); Headache(3769); Dizziness(3338); Paraesthesia(1714); Syncope(1154);
Skin and subcutaneous tissue disorders (18,627)	Rash maculo-papular(1169); Hyperhidrosis(1263); Erythema(1478); Rash(5109); Pruritus(3760); Urticaria(3837)
Gastrointestinal disorders (9837)	Nausea(3759); Abdominal pain(1202); Diarrhoea(2029); Vomiting(3580)
Injury, poisoning and procedural complications (8670)	Incorrect product storage(2615);
Musculoskeletal and connective tissue disorders (7991)	Arthralgia(2416); Myalgia(2699)
Psychiatric disorders (4658)	Agitation(2304)
Vascular disorders (4024)	Pallor(1557); Vasodilatation(1844)
Respiratory, thoracic and mediastinal disorders (3396)	Dyspnoea(1825)
Infections and infestations (3039)	Infection (1287)
Investigations (2869)	–
Metabolism and nutrition disorders (1546)	–

AE adverse event, PT preferred term, SOC system organ classification.

Disproportionality analysis

The positive signals generated through the analysis of ROR, MHRA, and BCPNN primarily encompassed General disorders and administration site conditions, Nervous system disorders, Skin and subcutaneous tissue diseases, Injury, poisoning and procedural complications, Psychiatric disorders, Infectious and infestations, among others, involving 22 System Organ Classes (SOCs) and a total of 254 Preferred Terms (PTs), as shown in Fig. 2 and 3. General disorders and administration site conditions were the most common (N = 15,813, 24.71%), followed by diverse Nervous system disorders (N = 8,823, 13.79%), and Skin and subcutaneous tissue diseases (N = 6,774, 10.58%). Investigations contained the most PT types (N = 33,12.99%), followed by Nervous system disorders (N = 32, 12.60%), and General disorders and administration site conditions (N = 21, 8.27%).

Fig. 2.

HBV AE SOC positive signals.

Fig. 3.

HBV AE positive signaling SOC includes PT types.

Among the positive PTs, Urticaria, Drugs ineffective, Product storage error, Agitation, Crying, Injections site hypersensitivity, and Injection site oedema were relatively high. In addition, the signals of Recalled products administered (ROR: 965.45; PRR: 964.91; IC-2SD: 4.06; = 4570.97), Hepatitis B DNA assay negative (ROR: 232.66; PRR: 232.65; IC-2SD: 0.57; = 354.70), Hepatitis B antibody positive (ROR: 220.00; PRR: 219.97; IC-2SD: 2.22; = 878.29), and Gamma-glutamyl transferase increased (ROR: 153.70; PRR: 153.65; IC-2SD: 3.46; = 2225.78) were the strongest, followed by Fibrosis tendinous (ROR: 141.00; PRR: 141.00; IC-2SD: 0.38; = 288.56) and Hepatitis (ROR: 139.77; PRR: 139.59; IC-2SD: 4.35; x² = 8518.60), Hyperbilirubinemia (ROR: 122.02; PRR: 121.96; IC-2SD: 3.72; = 2996.61), which deserve attention. See Suppl Appendix 1 for details.

DME list screening results

Among all the detected signals, those related to the DME include Anaphylactoid reaction, Angioedema, Aplastic anaemia, Dermatitis exfoliative, Erythema multiforme, Haemolytic anaemia, and Hepatic failure (Table 5). Additionally, after excluding adverse events already listed in the product insert, Aplastic anaemia, Dermatitis exfoliative, Haemolytic anaemia, and Hepatic failure were identified as potential new signals.

Table 5.

DME screening results for HBV.

PT	A	ROR	PRR		IC-2SD
Skin and subcutaneous tissue disorders
Dermatitis exfoliative	59	14.11 (10.51–18.94)	14.10	528.70	2.25
Blood and lymphatic system disorders
Aplastic anaemia	16	6.57 (3.88–11.13)	6.57	60.57	0.45
Haemolytic anaemia	23	6.16 (3.98–9.54)	6.16	82.19	0.76
Hepatobiliary disorders
Hepatic failure	20	4.57 (2.88–7.25)	4.57	47.14	0.34

DME designated medical events.

Discussion

In this study, we evaluated the safety of HBVs post-licensure through disproportionality analysis of the VAERS database and screening of the DME list. The results of disproportionality analysis showed that the most common AEs associated with HBVs were Urticaria, Agitation, and Injection site hypersensitivity. Additionally, among the 254 positive PTs identified, 7 signals were consistent with PTs in the DME list, including Anaphylactoid reaction, Angioedema, and Erythema multiforme as known signals. Four PTs, namely Aplastic anaemia, Dermatitis exfoliative, Haemolytic anaemia, and Hepatic failure were identified as potential new signals and warrant further monitoring. Cases of liver failure related to HBV have not been reported, possibly due to the lack of protective effects of HBV vaccine on HBV patients, resulting in the occurrence of liver failure cases. Therefore, the possibility of false positive signals cannot be ruled out, and further monitoring is needed, so they are not included in this discussion.

Aplastic anaemia

Aplastic anaemia (AA) is a severe hematologic disorder characterized by peripheral blood cell reduction and bone marrow failure, with the core mechanism of pathogenesis being immune-mediated bone marrow failure driven by T cell hyperactivity²⁶. While most cases of AA are acquired, there are also rare genetic forms. The case reported in this study developed AA following hepatitis B vaccination, which suggests an acquired form. Therefore, genetic forms of AA are not discussed here.

Acquired AA is largely caused by autoimmune lymphocyte-mediated destruction of hematopoietic stem cells, often presenting as life-threatening pancytopenia. Environmental exposures, such as drugs, viruses, and toxins, are believed to trigger abnormal immune responses in some patients²⁷, and vaccination may also be a risk factor for inducing acquired AA^28–31. We reviewed 16 aplastic anaemia adverse events and found that the mean age of these patients was 14.50 years, including 11 male patients and 5 female patients. Their average duration of illness was 175.23 days. In addition, we examined other adverse events that occurred in these patients around the same time as aplastic anaemia, and the top three with the highest frequency were bruising (5 cases), pancytopenia (5 cases), and petechiae (4 cases). It is worth noting that petechiae were detected as positive signals in this study. As early as 2005, Brodsky et al. identified peripheral blood cytopenia, dyspnea when forced, fatigue, easy bruising and petechial as clinical features of aplastic anaemia, which may explain the phenomenon of aplastic anaemia accompanied by bruising, pancytopenia and petechial symptoms after hepatitis B vaccine vaccination³².

The role of HBV as a predisposing factor for AA has never been demonstrated, but such cases have been reported to occur after hepatitis B vaccination^33,34, Among them, Viallard et al. speculated that immunization might stimulate a latent autoimmune genetic predisposition. An underlying immune predisposition (HLA-DR3) may have indirectly enabled the vaccine to trigger a hepatitis B virus-specific cytotoxic T-lymphocyte response. It is therefore possible that the pancytopenia was induced by a dysregulation of the CD8 T-cell compartment via increased IFN-γ production³⁴.

This is consistent with the findings of this study, highlighting the importance of understanding the medical history of individuals before administering hepatitis B vaccination in the real world. Killick et al. suggested that there is a potential risk of AA recurrence after vaccination in patients with AA responding to immunosuppressive therapy, and vaccination, including influenza vaccination, should be avoided if possible³⁵. Additionally, patients with a genetic predisposition to autoimmune diseases should seek immediate medical assistance if they experience symptoms such as anaemia, pallor, high fever, palpitations, etc., after vaccination, and closely monitor their vital signs.

Dermatitis exfoliative

Dermatitis exfoliative, also known as "red skin disease", was first described by Von Hebra in 1868, and later characterized by inflammation of the skin with impaired skin barrier and metabolic function³⁶. Multiple diseases can lead to Dermatitis exfoliative, broadly classified into congenital, infectious, immune-mediated, neoplastic, iatrogenic, and idiopathic causes, with psoriasis and eczema being the main ones³⁷. However, the pathogenesis of the disease remains poorly understood, with some theories suggesting increased expression of adhesion factors, stimulation of dermal inflammation by pro-inflammatory mediators and increased epidermal turnover rate leading to shedding of skin cells³⁶.

Cases of psoriasis developing after hepatitis B vaccination have been reported, with one subject withdrawing from a study by Morris et al. investigating the efficacy and feasibility of intradermal yeast recombinant HBV due to severe adverse event of psoriasis³⁸. Furthermore, Júnior et al³⁹. found a close association between psoriasis risk and hepatitis B vaccination (p = 0.034) by studying environmental and personal factors related to protection and susceptibility to autoimmune diseases. This may be due to the inclusion of thimerosal, a mercury-containing compound, in vaccine formulations as a preservative to prevent microbial contamination. In mice, inorganic mercury causes polyclonal T-cell dependent B-cell activation, hypogammaglobulinemia, and production of autoantibodies against fibrillarin^40,41.

Therefore, individuals with skin diseases or chronic skin conditions who experience itching, measles-like rashes, mossy-like lesions, or hives-like eruptions after hepatitis B vaccination should be closely monitored for potential worsening of their condition.

Haemolytic anaemia

Haemolytic anaemia is a group of diseases with different clinical and molecular heterogeneity, characterized by a decrease in circulating red blood cell levels and pathological manifestations of shortened lifespan of red blood cells accompanied by severe anaemia or compensatory hemolysis with increased reticulocytes⁴². Microangiopathic Haemolytic anaemia (MAHA) refers to a subset of Haemolytic anaemia, where red blood cell damage in small blood vessels leads to fragmentation and hemolysis⁴³. MAHA may occur independently but is more commonly associated with thrombocytopenic purpura⁴⁴.

Recombinant HBV-related thrombocytopenic purpura was first reported by Poullin and Gabriel in 1994, where two young girls, aged 15 and 21, developed the condition after receiving the 2nd and 3rd doses of HBV⁴⁵. Subsequent studies reported more cases of thrombocytopenic purpura following hepatitis B vaccination, establishing an association between HBV and thrombocytopenic purpura^46–50.

Geier MR et al. found 283 cases of reported AEs of complete blood cell count reduction/thrombocytopenia following hepatitis B vaccination by studying scientific literature and events reported to VAERS⁵¹. Epidemiological studies concluded that hepatitis B vaccination significantly increased the risk of thrombocytopenia⁵². This may be due to thrombocytopenic purpura being an immune complex syndrome caused by autoantibodies to glycoprotein IIb/IIIa molecules in platelet membranes⁴⁶. HBV contains yeast, aluminum, thiomersal, and hepatitis B surface antigen epitopes, which may contribute to autoimmune-related diseases in susceptible vaccinated individuals⁵³. In 1999, the United States announced the elimination of the use of thimerosal in children’s vaccines in the United States. As the data in this study showed, a total of 23 cases of hemolytic anaemia were reported, of which 21 cases occurred before 1999, which was consistent with the results of this study^53,54.

Furthermore, most cases in this study reported symptoms of Haemolytic anaemia after receiving two or more doses of recombinant HBV, possibly due to repeated doses of the vaccine promoting the formation of autoantibodies^46,47. Therefore, patients with pre-existing autoimmune diseases should be closely monitored after hepatitis B vaccination, and measures should be taken promptly if symptoms such as pallor, jaundice, or tachycardia occur.

Limitation

The VAERS database, as a voluntary reporting system, existing literature clearly documents that AE reports in VAERS represent only a very small subset of the population AEs, and the reports may contain incomplete, inaccurate, coincidental, and unconfirmed information³. VAERS data does not represent all known safety information for a vaccine, and the number of reports itself cannot be interpreted as evidence of a causal relationship between the vaccine and AEs. The data should be interpreted in conjunction with other scientific information.

Furthermore, the disproportionate reporting measure of detecting ADR signals can only indicate a statistical association between vaccines and ADR signals, serving only as a signal for further investigation and evaluation of causal relationships. This study only analyzed reports involving HBV and did not consider cases of concomitant vaccine administration.

Finally, we must acknowledge that the composition of hepatitis B vaccines produced by different manufacturers may cause some degree of bias in our study.

Conclusion

This study evaluated the safety profile of HBV in the real world setting from the perspective of vaccine safety signals using the VAERS database. A total of 254 positive signals were identified, with the most common being General disorders and administration site conditions, and Nervous system disorders. After screening with the DME list, three potential new signals including Aplastic anaemia, Dermatitis exfoliative, and Haemolytic anaemia were discovered. Some subtypes of these disorders are autoimmune diseases, and immunization may trigger potential autoimmune genetic predisposition. Therefore, individuals with autoimmune diseases or a family history of inherited autoimmune diseases should be closely monitored after HBV vaccination, and appropriate assistance should be provided promptly if any adverse events occur.

Author contributions

All authors contributed to the study conception and design. Data collection and analysis were performed by Huting Zhou, Jiale Yang and Jinbo Zhang. The manuscript writing were completed by Huting Zhou. The manuscript was revised by Huting Zhou, Pengcheng Liu and Dongning Yao, and all authors discussed previous versions of the manuscript. All authors read and approved the final manuscript.

Data availability

Data and the material is provided by the VAERS database data, are available from The Vaccine Adverse Event Reporting System (VAERS) Request Form (cdc.gov) free access. Analytical data sets are available from corresponding authors upon reasonable request.

Code availability

Python code has been uploaded to https://github.com/zhangj1nbo/vaers_detection.git.

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval

Anonymized data were collected from a publicly available database and do not require ethics committee approval.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-90135-8.