Assessment of adherence to pre-vaccination precautions and AEFI reporting practices during BCG vaccination in 4 hospitals in Ghana

ABSTRACT

The BCG vaccine, like all other vaccines, is associated with adverse events following immunization (AEFI). Reducing the incidence of AEFI is crucial in reposing confidence in BCG vaccination and reducing hesitancy associated with the vaccine. This requires safety precautions before and during vaccinations, as well as reporting AEFIs after vaccination. This study assessed the adherence of health-care professionals to pre-vaccination precautions and adverse events following immunization (AEFI) reporting practices during BCG vaccination in four hospitals in Ghana. It is hoped that the findings of the study will serve as a baseline to identify gaps for further studies to generate a stronger evidence for policy formulation aimed at improving BCG vaccine safety in Ghana and other tuberculosis endemic countries. A cross-sectional study design was employed, and Statistical Package for Social Sciences, IBM® SPSS version 25 (SPSS Inc. USA) software was used for analysis. Chi-square and binary logistic regression tests were used to test the association between categorical variables and predictors of adherence to pre-BCG vaccination precautions, respectively, and a p-value of <.05 was considered statistically significant. The AEFIs commonly reported by mothers included abscess, injection site pain, injection site redness, fever, rash, muscle weakness, diarrhea, vomiting, coughing and rhinitis. Ninety-three participants (73.2%) were adherent to pre-BCG vaccination precautions. Ninety-two participants (72.4%) informed mothers to report all AEFIs encountered. Adherence to pre-BCG vaccination precautions and AEFI reporting were generally good; however, there is still room for improvement.

KEYWORDS:

• Adverse events following immunization
• Severe Combined Immunodeficiency (SCID)
• Bacillus Calmette Guḕrin (BCG) vaccine
• Mycobacterium tuberculosis
• Mycobacterium bovis

Introduction

One of the major vaccines administered to children less than 5 years of age is the Bacillus Calmette Guḕrin (BCG) vaccine and it is given as soon as possible after birth and considered to be the most widely used vaccine globally. The use of the BCG vaccine is particularly high in poverty-stricken regions of the world where the incidence of tuberculosis is relatively high.¹ The BCG vaccine has been used for the immunoprophylaxis of tuberculosis since 1921 and it is prepared from live-attenuated Mycobacterium bovis, a bacterium closely related to Mycobacterium tuberculosis which causes tuberculosis.² The BCG vaccine induces T-cells and innate immune response against Mycobacterium tuberculosis and other organisms such as Candida albicans and Staphylococcus aureus.³

Several studies have demonstrated that the BCG vaccine is associated with serious adverse events following immunization (AEFI) such as erythema, keloid formation, non-suppurative lymphadenitis and skin ulcers and minor AEFI such as swelling, pruritus, pain, fever, weakness, arthralgia, headache, myalgia, nausea, extranodal injection site abscess and suppurative lymphadenitis.^4–6 Aside from these AEFI, one other rare but serious unexpected adverse event associated with the BCG vaccine is anaphylaxis. Anaphylaxis requires immediate management because it is very life-threatening.⁷ Reducing the incidence of AEFI is crucial in reposing confidence in BCG vaccination and reducing hesitancy associated with the vaccine. This requires safety precautions before and during vaccinations, as well as reporting AEFIs after vaccination.

For this reason, the WHO has set AEFI reporting targets for countries to help in generating vaccine safety signals.⁸ Risk-benefit analyses following signal generation from AEFI reports could inform decisions aimed at protecting vaccine recipients such as the withdrawal of vaccines or vaccine batches with serious causal AEFI associations. Furthermore, records of vaccine recipients with previous serious AEFIs such as erythema multiforme, Stevens-Johnson syndrome and toxic epidermal necrolysis (TEN) kept for future reference can minimize the probability of recurrence of AEFI.^9–13

It has been shown that complications of BCG vaccine especially disseminated infection and its most severe forms are known to occur in immunodeficient vaccine recipients, particularly in Severe Combined Immunodeficiency (SCID). Thus, a carefully taken family history before BCG vaccination as well as delaying vaccination if primary immunodeficiency is suspected could be a simple and effective method to avoid inappropriate vaccination of an immunodeficient baby in some cases until the prospect of newborn screening for SCID has been fully developed.¹⁴ A systematic review conducted by Yamoah et al. in 2020 on the safety of vaccines used in routine immunization in Africa showed that there was a high strength of association between BCG vaccine-related AEFI and the BCG vaccine using the Agency for Healthcare Research and Quality guidelines.¹⁵ HCPs involved in BCG vaccination could be instrumental in minimizing the incidence of BCG vaccine-related AEFI and treating life threatening outcomes like anaphylaxis while reporting any observed AEFI in a timely manner for causality assessment. This study therefore assessed the adherence of HCPs to pre-vaccination precautions and AEFI reporting practices during BCG vaccination in four hospitals in Ghana. The specific objectives of the study were to determine the level of adherence of HCPs to pre-vaccination precautions, the AEFIs commonly reported after BCG vaccination, the association between study participant characteristics and adherence to pre-vaccination precautions as well as predictors of adherence to pre-BCG vaccination precautions.

It is hoped that the findings of the study will serve as a baseline to identify gaps for further studies to generate a stronger evidence for policy formulation aimed at improving BCG vaccine safety in Ghana and other tuberculosis endemic countries.

Methods

Study design

This study was conducted from 1 July, 2017 to 31 December, 2017 and was cross-sectional in design using a paper-based structured questionnaire.

Study settings

The study was conducted in four hospitals in Ghana with a blend of the levels of healthcare (i.e. primary, secondary and tertiary) located in the middle and southern sectors of the country. The hospitals in the middle sector were the Komfo Anokye Teaching Hospital (KATH), a 1200 bed tertiary health-care facility in Kumasi and the Agogo Presbyterian Hospital (APH), a 200 bed secondary health-care facility, both in the Ashanti region, whereas those in the southern belt were the Tema General Hospital (TGH), a 300 bed secondary health-care facility in the Greater Accra region and the Nagel Hospital (NH), a 70 bed primary health-care facility in the Sekondi-Takoradi metropolis of the western region. KATH serves as a referral center for hospitals in the Ashanti region, the entire northern belt and other neighboring West African countries. The APH is located in the capital town of Ashanti-Akim North Municipal and also serves as the major health-care facility in Agogo as well as a referral center for clinics located in the adjoining towns and villages. It is also one of the vaccine clinical trial sites of Ghana. NH is a mission facility that has been adopted by the government to support the other hospitals where minor medical and surgical cases in the Western region are treated. TGH serves the Tema communities and other surrounding towns in the Greater Accra and Volta regions of Ghana. These sites were selected purposively because of established adverse drug reaction (ADR) and AEFI reporting systems and collaboration with the National Pharmacovigilance Center of the Food and Drugs Authority of Ghana.

Participants

Participants were HCPs at vaccination centers of the selected hospitals were the participants in this study.

Inclusion/exclusion criteria

Vaccination center staff and interns who had been at post for at least 6 months in the various hospitals and consented to partake in the study were recruited. All other HCPs apart from these categories were excluded from the study.

Study sample

The study sample was drawn from four districts in four regions of Ghana. The total population of HCPs involved in vaccination in these four districts at the time of the study were 342. Assuming a margin of error of 5%, a confidence level of 95% and a response distribution of 50% for best responses, a sample size of 182 was estimated for the 368 study population using the online raosoft sample size calculator.¹⁶ There were 133 eligible participants for the study based on records from the human resource departments of the various study sites. All these participants were contacted for informed consent.

Data collection tool

A paper-based structured questionnaire was the data collection tool for the study. The tool was prepared by the principal investigator, co-investigators and a public health physician at the vaccination center of one of the study sites, KATH. This tool was prepared after an extensive search of the literature for a standardized data collection tool for collecting information on knowledge, perceptions and practice of AEFI reporting from HCPs. After the search, no standardized data collection tool was found and therefore, reference was made to the portion of the data collection tool assessing HCPs on BCG vaccination practices in a similar study conducted by Masika et al., 2016, in the design of the questionnaire for the current study.¹⁷ A seven-stage process was used to validate the questionnaire as follows: initial design of tool, face validity testing, content validity testing, criterion validity testing, development of a yardstick for the tool, assessment of inter- and intra-rater reliability of tool and assessment of inter-rater reliability in situ. A 2 member panel that included a public health physician and a public health nurse assessed the questionnaire for face validity. They were asked to comment in writing on the suitability of the questionnaire while providing any alterations needed to improve its quality. The same panel was asked to assess the content and criterion of the questionnaire using the Delphi techniques^18,19 to avoid the need of panel members to have a face-to-face meeting during the assessment process. Furthermore, another panel of three eminent academic researchers were sought to develop a yardstick to assist in the assessment of reliability of the questionnaire after which two knowledgeable and independent pharmacovigilance professionals were asked to rate the questionnaire by filling a form. After filling the form, the intra-rater reliability indicated a 94% agreement in the reliability of the questionnaire. Lastly, three public health nurses with some basic knowledge in vaccine pharmacovigilance were randomly selected from each study site to try the questionnaire in situ after receiving a 15 minute briefing about the content of the questionnaire by the principal investigator. To minimize bias, total independence was maintained between these raters. An inter-rater questionnaire reliability of 96% was obtained after the exercise. Altman provides a scale of acceptance levels of reliability scores for medical research and reports that an agreement of 80% or more represents a high level of reliability.²⁰ With reference to this scale, the data collection tool was deemed to be suitable for the study.

Data collection

Data was collected by one assigned research assistant with no laboral relationships with the HCPs in each of the hospitals. Prospective participants were contacted by the research assistants at the vaccination centers during working hours (8am–5pm), from Monday to Friday throughout the study period. The questionnaires were then administered to the participants to complete immediately after informed consent. All questionnaires were received on the same day of obtaining consent from participants.

Details of collected data

The questionnaire consisted of 19 questions and was divided into 3 sections. The first section consisted of five questions and was designed to collect socio-demographic data. The second portion which ascertained the adherence to pre-BCG vaccination practices consisted of eight questions, whereas the third section consisted of six questions on AEFI encounter and AEFI training received within the past year.

Data analysis

Collected data were entered periodically into Statistical Package for Social Sciences, IBM® SPSS version 25 (SPSS Inc., USA) software for cleaning and analysis. Descriptive statistical analysis was conducted to generate frequencies, mean and standard deviation. A mark of 1 was allotted for each “yes” response to the eight questions assessing the pre-vaccination precautions taken by staff of the vaccination centers. Thus, a total maximum score of 8 was expected for participants who adhered to all the precautions and 0 for participants who adhered to none of the precautions. The average score was computed, and all participants scoring below the average score were deemed to exhibit an overall non-adherence to pre-vaccination precautions.

A chi-square test was used for tests of association between adherence to pre-BCG vaccination precautions and categorical variables and a p-value of <.05 was considered statistically significant. Binary logistic regression was used to determine the predictors of adherence to pre-BCG vaccination practices using odds ratio at a confidence interval of 95% and a p-value of <.05 was considered significant. All results are summarized in tables.

Ethical considerations

Ethical approval was sought from the Committee on Human Research Publication and Ethics (CHRPE) of the School of Medical Sciences, Kwame Nkrumah University of Science and Technology (KNUST) before the commencement of the study (Reference number: CHRPE/AP/220/17). Before administering the questionnaires, each prospective study participant was given a participant information and consent (PIC) form to complete.

Results

Socio-demographic characteristics of study participants

Out of 133 prospective participants, 127 (95.5%) consented to partake in the study. The mean age of recruited subjects was 28.6 years (SD = 4.20 years) with minimum and maximum ages of 20 years and 51 years, respectively. Moreover, the mean number of years of practicing was 7.9 (SD = 1.4 years) with minimum and maximum years of practice of 1 and 28 years, respectively. There were 18 (14.2%) males and 109 (85.2%) females. Also, there were 76 (59.8%) public health nurses, 28 (22.1%) disease control officers and 23 (18.1%) interns. Table 1 summarizes the socio-demographic characteristics of study participants.

Table 1. Socio-demographic characteristics of study participants.

Variable, n = 127	Frequency (n)	Percentage (%)
Study site
KATH	49	38.6
APH	46	36.2
TGH	24	18.9
NH	8	6.3
Age (years)
20–29	73	57.5
30–39	37	29.1
40–49	13	10.2
≥50	4	3.1
Gender
Male	18	14.2
Female	109	85.9
Profession
Public health nurse	76	59.8
Disease control officer	28	22.1
Intern	23	18.1
Work experience (years)
1–5	76	59.8
6–10	32	25.2
11–15	15	11.8
16–20	3	2.4
≥21	1	0.8

Table 2. AEFI reporting practices.

Variable, n = 127	Frequency (n)	Percentage (%)
Encountered AEFI
Yes	91	71.7
No	36	28.3
Types of AEFI reported by parents
Injection site pain	78	20.3
Injection site redness	46	12.0
Abscess	34	8.9
Fever	89	23.1
Rash	18	4.7
Muscle weakness	54	14.1
Diarrhea	8	2.1
Nausea	16	4.2
Vomiting	29	7.6
Coughing	7	1.8
Rhinitis	5	1.3
Reported AEFI to authorities?
Yes	64	70.3
No	27	29.7
Told guardians to report all AEFIs?
Yes	92	72.4
No	35	27.6

AEFI reporting practices of HCPs

Ninety-one (71.7%) study participants had ever encountered an AEFI in their practice. However, only 64 out of this number (70.3%) had ever reported an AEFI. A total of 11 types of AEFIs yielding a cumulative number of 384 were encountered out of which the highest number of AEFIs was attributed to fever, whereas the lowest was rhinitis. Table 2 is a detailed description of AEFI reporting practices among study participants.

Table 3. “Yes” responses to pre-BCG vaccination precautions.

Variable, n = 127	Frequency	Percentage
Ensure that refrigerator in which vaccines are kept is functional	87	68.5
Ensure that vaccines are maintained at a temperature of 2–8${ }^{\circ }\mathrm{C}$	52	40.9
Check the batch number of vaccine	123	96.9
Check expiry date of vaccine	121	95.3
Ensure that the vaccine solution is clear in its ampoule with no color change	62	48.8
Rule out contra-indication of a vaccine in a child prior to administration	74	58.3
Ensure there is anaphylactic pack with adrenaline in the injection room	43	33.9
Inform care takers about common adverse events to be expected	39	30.7
Adherence to pre-vaccination precautions
Adherent	93	73.2
Non-adherent	34	26.8

Table 4. Relationship between participant characteristics and adherence to pre-vaccination precautions.

Variable	Adherent (%)	Non-adherent (%)	df	χ2	p-Value
Work location
KATH	35 (71.4)	14 (28.6)	6	12.953	.007
APH	39 (84.8)	7 (15.2)
TGH	14 (58.3)	10 (41.7)
NH	5 (62.5)	3 (37.5)
Age
20–29	54 (74.0)	19 (26.0)	9	11.739	.957
30–39	27 (73.0)	10 (27.0)
40–49	9 (69.2)	4 (30.8)
≥50	3 (75.0	1 (25.0)
Gender
Male	12 (66.7)	6 (33.3)	12	16.273	.642
Female	81 (81.2)	28 (18.8)
Profession
Public Health Nurse	68 (89.5)	8 (10.5)	9	12.286	.269
Disease control officer	17 (60.7)	11 (39.3)
Intern	8 (34.8)	15 (65.2)
Work Experience (years)
1–5	53 (69.7)	23 (30.3)	15	18.391	.046
6–10	24 (75.0)	8 (25.0)
11–15	12 (80.0)	3 (20.0)
16–20	3 (100.0)	0 (0)
≥21	1 (100.0)	0 (0)
Number of trainings on vaccination within the past year
None	1 (33.3)	2 (66.7)	12	14.058	<.0001
1	3 (37.5)	5 (62.5)
2	5 (38.5)	8 (61.5)
3	5 (100.0)	0 (0)
More than 3	3 (100.0)	0 (0)

Pre-BCG vaccination precautions

The highest pre-vaccination precaution taken was recording of vaccine batch numbers (96.9%) and expiry dates before administration, whereas the lowest was informing parents about the common AEFIs to expect (30.7%). Fifty-two (40.9%) ensured that BCG vaccines were maintained within the cold chain temperature range of 2–8 degree Celsius, 121 (95.3%) checked the expiry dates of BCG vaccines before administration, 62 (48.8%) ensured that the BCG vaccine solution was clear in its ampoule with no color change before administration, 74 (58.3%) ruled out contraindications of the BCG vaccine before administration and 43 (33.9%) ensured the availability of adrenaline anaphylactic pack before BCG vaccine administration. Over 70% of study participants were adherent to pre-vaccination precautions. The highest and lowest pre-vaccination precaution scores were 8 and 3, respectively, with a mean score of 4.7 (SD = 1.6). Details of pre-vaccination precautions are presented in Table 3.

Chi-square test of associations

From the chi-square test of associations, the relationship between work location, work experience and the number of trainings on vaccines per year and adherence to pre-vaccination precautions were statistically significant. However, the relationship between age, gender and profession were not statistically significant. Detailed results of the chi-square test of association between participant characteristics and pre-vaccination precautions are as shown in Table 4.

Bivariate analysis

Binary logistic regression revealed that work location, age, work experience and the number of vaccination trainings in a year were independent predictors of adherence to pre-BCG vaccination precautions, whereas gender and profession were not. Results of the bivariate analysis are presented in Table 5.

Table 5. Bivariate analysis of independent predictors of pre-BCG vaccination precautions.

Variables	Wald	Bivariate p	Odds ratio (95% Confidence interval)
Work location
TGH	Reference	0.003	4.86 (4.11–5.48)
APH	5.81
KATH	7.57
NH	2.31
Age
20–29	Reference	0.018	4.27 (3.83–4.97)
30–39	1.79
40–49	8.94
≥50	12.62
Gender
Male	Reference	0.571	3.62 (3.76–5.62)
Female	1.86
Profession
Public Health Nurse	3.79	0.962	4.92 (4.33–6.17)
Disease control officer	1.22
Intern	Reference
Work Experience (years)
1–5	Reference	0.037	4.84 (3.92–5.69)
6–10	1.03
11–15	3.76
16–20	2.48
≥ 21	4.92
Number of trainings on vaccination within the past year
None	Reference	0.001	4.96 (4.62–5.83)
1	1.05
2	1.94
3	2.87
More than 3	4.63

Discussion

To the best of our knowledge, this is the first study being conducted in Ghana to ascertain whether HCPs involved in BCG vaccination adhere to precautions related to BCG vaccination. The study also assessed BCG vaccine-related AEFI reporting in the same population. According to the WHO, AEFI could be caused by vaccine products.²¹ Vaccine product integrity after regulatory approval largely depends on cold chain storage since vaccine products are biologic in nature. Vaccine storage outside the optimal range of 2–8 degree Celsius results in irreversible loss to potency and efficiency.²² For this reason, the availability of functional refrigerators and proper cold chain storage systems must always be ensured by HCPs in vaccination centers. In most LMICs, rampant electrical power outages can cause refrigerators to malfunction which may affect cold chain storage of vaccines. As a result, stand-by generators are of the essence in such jurisdictions. Moreover, thermometers in refrigerators must be ensured to be functional at all times. It was observed in this study that about two-thirds of study participants ensured that refrigerators in which vaccines are kept are functional, whereas two-fifths checked whether cold chain temperatures were maintained before administering vaccines. These findings are similar to those of a similar study conducted at primary health-care facilities in Dalocha District in Ethiopia, East Africa, which concluded that there was a gap in vaccine cold chain management among the primary HCPs.²³

Checking the expiry dates as well as the color of vaccines before administration are critical activities that safeguard the integrity and potency of vaccines and reduce the incidence of AEFI. Expired or unwholesome vaccines may have a color change that must be carefully examined by HCP administering them. Inadvertently administering expired vaccines could lead to life-threatening AEFI. Moreover, the vaccine recipient risks getting infected with the disease being vaccinated against and spreading it due to lack of protection resulting from impotent vaccine. These effects are not only limited to the BCG vaccine. A July 2018-March 2019 report from the United States Vaccine Adverse Event Reporting System (VAERS) observed that expired injectable influenza vaccines were inadvertently administered to 192 vaccine recipients, which led to 125 AEFI highlighting the need to check vaccine expiry dates before administration.²⁴ In the current study, however, it was observed that more than 95% of the study participants checked the expiry dates of the BCG vaccine before administration which is commendable.

Checking the batch number of BCG vaccines before administration is a very important precaution in identifying batches that may have defects in manufacturing and contamination leading to adverse events. About 97% of the participants in this study checked the batch numbers of the BCG vaccine before administration. This high figure could be explained by the fact that recording the batch number is mandatory in vaccination schedule booklet for infants.

Assessing the immune function of infants before BCG vaccine administration is very crucial because the vaccine strain could replicate abnormally to cause tuberculosis in immuno-compromised infants such as low-birth weight infants, pre-term infants and infants with SCID.^14,25 This assessment requires collaboration with the mothers and their gynecologists. It must be emphasized that ruling out contraindications of vaccines could sometimes be laborious as not all past untoward events associated with vaccines may prove beneficial. For example, in infants who are prone to developing an auto-immune disease after vaccination based on a history of BCG vaccination-related AEFI of the parents, past records of its occurrence alone may not be sufficient as autoimmunity may have resulted from a family history of autoimmunity or autoimmune antibodies resulting from a genetic predisposition.²⁶ This shows that genetics has a link with vaccine adverse events and therefore experts in genetics could play a significant role in ruling out possible contraindications of vaccines before administration. There is a growing vaccine pharmacovigilance field called Adversomics which studies the linkage between vaccine adverse events and genetics and this may mark a new paradigm for vaccine safety and design. Developing this area could lead to personalized vaccine design through genetic profiling of the population. Such products are currently being used in the field of oncology and are safer and cheaper.^27,28 In the current study, almost two-thirds of the participants assessed the immune function of infants before BCG vaccine administration, which requires significant improvement.

Ensuring the availability of anaphylactic pack including adrenaline during BCG vaccination is a precaution that must not be overlooked by HCPs. In the current study, just a third of the participants took this caution before BCG vaccination. Even though anaphylaxis following BCG vaccination has been found to be rare, its life-threatening potential makes it imperative to prevent it.⁷

AEFI reporting is important in post-market surveillance of vaccines. Reporting by patients and HCPs is a passive AEFI surveillance approach that seeks to generate more safety information on vaccines even though it is associated with underestimation of actual AEFI.²⁹ Mothers of children who receive BCG vaccines must observe their babies closely and report AEFI promptly to HCPs for causality assessment to be undertaken on those that are severe and life-threatening. According to the WHO, the BCG vaccine could cause BCG immune reconstitution syndrome (IRIS) and eye problems including uveitis, optic neuritis and lupus vulgaris particularly in vaccine recipients with primary immune deficiency syndromes.³⁰ These and other serious BCG-related AEFI have been shown to be associated with the switching of the BCG vaccine and the vaccine manufacturing process. Low AEFI reporting is a problem in most LMICs leading to scanty safety information on vaccines that affects the quality of causality assessment reports.³¹ HCPs at vaccination centers have the responsibility of educating mothers about both the common and self-limiting AEFIs as well as rare and unexpected ones while urging them to report. In the current study, about 70% of the participants did not inform mothers of infants who received the BCG vaccine and this requires significant improvement.

The number of respondents who had ever reported an AEFI to authorities for further investigation is comparable to that in similar studies conducted in Nigeria and Albania.^32,33 However, the AEFI reporting in the current study was higher than that of other similar studies conducted in Kenya and USA.^22,34 This shows that poor AEFI reporting is not limited to LMICs. A global systematic review has observed that AEFI surveillance is feasible in both high income and LMICs and can be improved by sharing vital vaccine safety information between these two regions if barriers to surveillance are removed.¹⁸ Several studies have identified the barriers of AEFI reporting and these include HCPs considering most AEFIs as irrelevant or unserious, forgetfulness, unavailability of AEFI reporting forms, unfamiliarity with reporting system and time and resource constraints.^35–37

Overall, about three-quarters of the study participants were adherent to pre-BCG vaccination precautions. It was observed in the current study that the relationship between work location, work experience and the number of trainings on vaccines per year and adherence to pre-BCG vaccination precautions were statistically significant. This was corroborated by the bivariate analysis that showed that HCPs who received at least three trainings on vaccine safety and surveillance within a year were about five times more likely to adhere to the pre-BCG vaccination precautions than those who received no training at all. Moreover, HCPs working at the highest level of care (KATH) were about 8 times more likely to adhere to these precautions than those working at TGH, a lower level of care. It is therefore imperative for the Expanded Program on Immunization to intensify training on vaccine safety and surveillance across all vaccination centers, particularly in the lower levels of healthcare across Ghana. The frequency of trainings may have to be increased particularly for HCPs with one to 5 years of work experience, as those within this bracket were least likely to adhere to the precaution.

Study limitation

The main limitation of the study is the relatively small sample size that may have affected the sizes of effects measured in the study. A larger study involving many study sites may portray a clearer picture of adherence to pre-BCG precautions and AEFI reporting habits by HCPs. Additionally, due to the inaccessibility of vaccination data such as recorded cold chain temperatures and reported AEFIs associated with the BCG vaccine, information was obtained directly from HCPs. This may have resulted in response bias leading to false positives as HCPs are more likely to report positive outcomes than negative outcomes.

Conclusions

Adherence to pre-BCG vaccination precautions and AEFI reporting were generally good; however, there is still room for improvement. This improvement could be achieved by providing anaphylactic packs with adrenaline in BCG vaccine injection rooms, ensuring that mothers report AEFI after BCG vaccination while intensifying training of HCPs on BCG vaccination and AEFI surveillance.

Transparency declaration

The authors affirm that the manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as originally planned have been explained.

Acknowledgments

We wish to acknowledge the WHO Collaborating Centre for Advocacy and Training in Pharmacovigilance in Ghana for providing valuable information on vaccine pharmacovigilance in Africa and the world at large.

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability statement

The datasets generated and/or analyzed during the current study are not publicly available due to confidentiality issues but are available from corresponding author upon reasonable request.

Additional information

Funding

The authors declare that no funding was received to conduct this study.