Factors predicting the level of vaccine protection against hepatitis B virus infection among physicians and nurses in Šabac, Serbia

This descriptive cross-sectional study of 100 physicians and 100 hundred nurses in primary and secondary health care in Šabac, Serbia took place from September 2020 to February 2021. The minimum sample size of 178 participants was based on statistical assessment described by Yanase et al. (20), with defined alpha error of 0.05, and confidence interval of 95 %. Our sample consists of 200 out of 255 healthcare workers from our previous pilot study who agreed to have their blood anti-HBs titre determined. At the time of the study, 100 worked at the Dr Draga Ljočić Healthcare Centre, consisting of nine services, and the other 100 at the Dr Laza K. Lazarević General Hospital, consisting of 18 units (nine surgical and nine internal medicine).

The inclusion criteria were direct contact with and treatment of patients, work experience of at least one year, and voluntary informed consent to participate in this study.

Table 1 shows detailed sociodemographic data and significant differences between groups of participants. Their overall mean age was 44.52±0.73 years and mean work experience 19.66±0.69 years. There are significant differences in some categorical variables, namely sex (dominantly female, p<0.001), workplace (dominantly outpatient clinic, p<0.001), and occupation (dominantly nurses p<0.001).

As far as continuous variables are concerned, they all significantly differ (Table 2).

All participants reported to have been vaccinated (200 of 200) in the deltoid muscle, which correlates with the data obtained from epidemiological services and the electronic database of both health institutions. One hundred and seventy-nine (89.5 %) received the complete vaccination series of three shots (Figure 1), of whom 41.5 % according to the 0–1–6 schedule (day zero, one month after the first dose, and six months after the first dose). Most participants (65 %) had been immunised more than 10 years before the study.

Figure 1

Distribution of participants according to the number of received vaccine doses (three doses denote completed vaccination)

Figure 2 shows the distribution of chronic diseases reported by the participants: 16.5 % had a chronic disease, whereas the rest reported none.

Figure 2

Distribution of participants by self-reported chronic diseases. HBV – hepatitis B virus; COPD – chronic obstructive pulmonary disease

The calculated BMI of our respondents shows that 32 % were overweight (BMI 25–29.9) and 9.5 % obese (BMI >30).

Thirty-five participants reported abnormal lipid status, of whom nine used and 26 did not use therapy. Tobacco smoking was reported by 66 participants.

Blood samples showed that 15 % (30 participants) did not develop seroconversion, i.e., did not have the protective antibody titre of over ≥10 mIU/mL (Figure 3).

Figure 3

Distribution of participants by anti-HBs titre threshold and number of received vaccine doses

Four participants had positive total anti-HBc, which is equal to the number of participants who reported being infected with the hepatitis B virus during their lifetime. In all other participants, total anti-HBc was negative.

Fisher’s test shows a significant association between the anti-HBs titre and the number of vaccine doses received (p=0.015). Participants who were fully vaccinated had a higher antibody titre than those who received one or two doses (Table 3).

ANOVA has also established a highly significant association between low anti-HBs titre (<10 mIU/mL) and the reported autoimmune disease (p=0.000) and cancer (p=0.015). Previous jaundice is significantly associated with the high titre (≥10 mIU/mL; p=0.034) (Table 4).

In turn, we found no significant association between the anti-HBs titre and participants’ affiliation, sex, field of medicine, workplace, occupation (ANOVA, Table 5), or any of the continuous variables (year of age, years of service, body mass, body height, and BMI) (Fisher’s test, Table 6).

All our participants received at least one dose of HB vaccine, which translates to 100 % coverage. Studies across the world report healthcare worker coverage ranging from 10–81 % in Africa to over 80 % in the USA and Australia ( 24, 25, 26, 27, 28, 29, 30, 31, 32). In Asia, surveys of healthcare workers providing secondary and tertiary care report 65 % and 67 % coverage, respectively (33, 34). In Serbia, recent data indicate that the coverage ranges between 31 and 96 % ( 35, 36, 37).

As for our findings that 89.5 % of our participants from Šabac completed the immunisation (received three shots), these figures excel reports of 50 % from three university hospitals in Belgrade (38) or 39.2 % from the public sector and 27.8 % from the private sector in the province of Vojvodina (39).

However, immunisation is not the same as having immunity, that is, having the anti-HBs titre >10 mIU/mL (regardless of the dynamic of decline). This is why all healthcare workers are advised to have their titre checked after having received three vaccine doses, especially those who come into contact with patient blood and body fluids (40, 41). Our study shows that 85 % of participants achieved immunity to HBV. These findings should be taken with some reserve, however, as 55 healthcare workers recruited in our previous pilot study refused to take immunity testing during recruitment, and were therefore excluded from this study. Although all of them were vaccinated, they stated that they considered themselves protected by the vaccine (38/55), were not interested in whether they achieved full protection (10/55), or were not exposed to blood-borne infections (7/55). Another report from Serbia speaks of 93 % seroconversion in their sample of 100 healthcare workers (36). Reported immunity among healthcare workers varies widely across the world, from 15–21 % in certain parts of Asia (42), 30.1 % in Bangladesh (43), 36.7 % in Indonesia (44), 47–74 % in Kenya (45, 46), 56.7 % in Congo (47), 77.1 % in Tanzania (48), 83.3 % in Tehran, Iran (49), 83 % in Japan (20) to 90 % in Egypt (50). Variation is particularly visible in India, ranging from 32 % (51), over 60.8 % (52) to 96 % (53). In Europe, a survey conducted in 21 countries of the European Union in 2016 showed that not all healthcare workers were protected after complete vaccination in as many as 18 countries (54). One meta-analysis of 19 Italian studies established that 27 % of healthcare workers did not achieve immunity (55), whereas another Italian study (56) among nurses showed a 85.7 % immunity. In Spanish Catalonia, one study reported 64.4 % (57), while a 2020 study in Turkey showed a 66.7 % protection (58).

As for factors that may prevent achieving immunity to HBV infection, we found a significant association between antibody titre and received vaccine doses or history of jaundice, autoimmune, and cancer. Age, sex, BMI, and smoking, in turn, are not significantly associated with anti-HBs titre. A 1987–1991 study conducted among 595 healthcare workers in Minnesota, USA six months after the third received vaccine dose (59) showed an association between insufficient anti-HBs titre and the type of vaccine, age, sex, BMI, and smoking, depending on which recombinant HB vaccine the participants received. We, however, could not establish which exactly recombinant vaccine our participant received to look deeper into the issue.

Complete vaccination is certainly a significant predictor of immunity to HBV infection, as it allows most recipients to acquire a sufficiently protective antibody titre (60). However, even receiving three vaccination doses may fail. Of the 30 participants (15 %) in our study whose antibody titre was below 10 mIU/mL, as many as 23 were fully vaccinated. Our analysis, however, has not found that age alone could be the reason, despite some indications that the titre drops with age (53, 61, 62). Perhaps the reason lies in the fact that anti-HBs titre remains high enough even 30 years after vaccination (63).

Our findings also fail to confirm earlier reports of association between obesity and non-response to HBV vaccine (21) or between BMI below 25 and seroprotection in healthcare workers (64). Truth be told, only 19 participants in our study were obese (BMI >30), which may not have been a representative sample to support these findings.

In contrast, failure to respond to vaccination is clearly associated with chronic jaundice, autoimmune disease, and cancer. Some studies argue against a link between HBV infection positive status and chronic diseases (65), while others indicate that people with lower kidney function, the elderly with diabetes, and people with inflammatory bowel diseases have a lower likelihood of seroconversion (66, 67).

Our study has certain limitations. Some are inherent to the study design partly relying on self-reported information. Then there is the issue of representation, as significantly more nurses participated than doctors. However, nurses are generally dominating in health care. Furthermore, the study does not include other categories of healthcare workers, such as laboratory and cleaning staff, who also run a risk of exposure to patient blood and other biological material.

Despite its limitations, our study shows total vaccination coverage of healthcare workers in Šabac, Serbia and a high rate (89.5 %) of those who completed the series by taking all three shots. The fact that 15 % did not achieve the protective antibody titre confirms the necessity of its control after immunisation, which is not routinely carried out in most countries, Serbia included. It is, therefore, necessary to develop a detailed strategy for monitoring vaccination and serological status of healthcare workers in order to improve their safety at work.

Perhaps even more important is to educate future health professionals about protection against HBV infection. Such education should continue throughout the working life to ensure the highest possible level of protection.