Hepatitis B virus vaccination status and associated factors among health care workers in Shashemene Zonal Town, Shashemene, Ethiopia: a cross sectional study

Study design

Institution based cross-sectional study was conducted to assess HBV vaccination status and associated factors among HCWs in Shashemene Zonal Town.

Study area and period

The study was conducted in Shashemene Zonal Town, Oromia Region, South Ethiopia from March 16, 2015 to March 31, 2015. The town is located at 250 kms away from the capital city, Addis Ababa due south with a total area of 12,960 sq km and a total population of 153,162. The city administration is divided into eight sub-cities. According to the city administration, Health office report: the public and private organizations which are involved in health care delivery include 1 referral hospital having 171 HCWs, 1 district hospital having 94 HCWs and 3 governmental health centers having 42 HCWs which makes the total number of HCWs in governmental institutions 307 and 1 health center having 19 HCWs, 4 special clinics having 15 HCWs, 6 higher clinics having 55 HCWs, 16 middle clinics having 97 HCWs and 8 primary clinics having 28 HCWs which collectively give a total of 214 non-governmental institution HCWs. Therefore there are a total of 521 HCWs in Shashemene Zonal Town.

Source and study population

The source population was all health care workers working in health care institutions found in Shashemene Zonal Town, during the data collection period and study population were those health care workers selected by using simple random sampling technique.

Sample size

The study employed single population proportion sample size determination formula. Fifty percent vaccination prevalence was considered with 95% CI, and 5% marginal error (where n is desired sample size, Z is value of standard normal variable at 95% confidence interval and, p is proportion of vaccination status of HBV among HCWs and d is marginal error which is 5%) was considered to calculate the sample size. Considering a non-response rate of 10%, the final sample size was calculated to be 423.

Sampling procedure

Samples were taken from all health care institutions in the town, and the samples were allocated proportionally based on the number of HCWs in the institutions by using a simple random sampling technique. Workers payroll was used as a sampling frame. To determine how many samples to take from a single institution, the total sample size needed (423) was multiplied by the total number of HCWs in the institution and divided by the total number of HCWs in the town (521).

Inclusion and exclusion criteria

Variables

Operational definitions

Data collection procedures

A structured questionnaire was used to collect the required quantitative information through a face to face interview. The data was collected by nurses who were educated at the graduate level. Four data collectors and one supervisor were recruited to conduct the study. Those HCWs on the night or weekend shifts were addressed according to their working shifts. The investigators were responsible for coordination and supervision of the data collection process.

Quality assurance

First the questionnaire was developed in English and then translated into Amharic and Affan Oromo, and back to English by different persons to check for consistency. Five percent of the questionnaires were pre tested in health care workers of Ras Desta Hospital. Based on the pretest, questions were revised, edited, and those questions found to be unclear or confusing were removed prior to the actual study. Knowledge questions were checked for content validity by experts. All the data collectors held graduate degrees, and they were given a half day of training. The investigators collected the completed questionnaires every day and checked each for inconsistencies and omissions. Any format with a defect was rejected from the study.

Data management and data analysis

After being coded, the data was entered using EPI info version 7 statistical software and analyzed using SPSS version 21 statistical package. Data cleaning was performed to check for frequencies, accuracy, consistencies, missed values and variables. Any logical or consistency errors identified during data entry were corrected after revision of the original completed questionnaire. The cleaned and edited data was prepared for appropriate statistical analysis. The mean, standard deviation and the proportion of the variables were computed. Both bivariate and multivariable logistic regression analyses were used to measure how the outcome variable (HBV vaccination status) depends on the covariate variables (knowledge about HBV, perceived price of vaccine, availability of vaccine and socio demographic variables). Variables that showed association in the bivariate analyses were taken into the multivariable regression model to control for confounding. The results of the analysis were presented using tables, charts and graphs. 95% CI and p value of less than 0.05 were considered as indicators of statistical significance.

Socio demographic characteristics

Vaccination status of HCWs

The overall fully vaccinated respondents were found to be 53 (12.9%) and those who had ever been vaccinated were 66 (16.1%) from the total 410 respondents. Among the 66 ever vaccinated respondents, 53 (80.3%) were fully vaccinated or took the vaccine three times and the remaining 5 (7.6%) and 8 (12.1%) took two and one time respectively or partially vaccinated. Only one (1.5%) had checked status of immunity after taking the vaccine.

Reasons for not being vaccinated

There were 344 unvaccinated respondents. The reasons for not being vaccinated varied from the highest 165 (48%), because the vaccine was not found in nearby institution, to the lowest 1 (0.3%), thinking it was not important, and 1 (0.3%) because of always forgetting to get the vaccine. The reasons for not being vaccinated are shown in Fig. 1.

None of the health care institutions have made available the vaccine and among the ever-vaccinated respondents, 10 (15.2%) had to pay to get the vaccine. The price of the vaccine was rated as expensive by 137 (33.4%), followed by 119 (29%) as very expensive and 36 (8.8%) as affordable, while 118 (28.8%) of respondents did not know the price. The majority of the respondents were found to be knowledgeable 257 (62.7%).

Factors associated with vaccination status

Then, multivariable logistic regression analysis was applied to control for confounders and identify the actual factors that are significantly associated with vaccination status of HCWs. After analysis, three variables were found to be significantly associated with vaccination status: sex, work experience and type of institution. Other socio demographic variables like age, occupation and marital status of respondents lost their association during the multivariable logistic regression model.

Female respondents were about four times as likely (AOR 3.84 [95% CI 1.66, 8.92]) (p = 0.002) to be fully vaccinated as compared to the male HCWs. Those who had a work experience 5–9 years were about six times more likely (AOR 5.88 [95% CI 2.24, 15.42]), and those who had ≥10 years were twelve times more likely (AOR 12.51 [95% CI 3.10, 50.41]) (p < 0.001) to be fully vaccinated than those who had a work experience of <5 years. The other variable that showed association was type of health institution. Those who work at governmental health care institutions are two times more likely (AOR 2.45 [95% CI 1.10, 5.46]) (p = 0.029) to be fully vaccinated than that of respondents from non-governmental health care institutions.

Vaccination status

The World Health Organization has estimated that mean HBV vaccination rate among HCWs ranges from 18 to 39% in developing countries to 67–79% in developed countries [13]. However, the overall fully vaccinated respondents in our study was found to be 53 (12.9%) from a total of 410 respondents. Therefore, the percentage of fully vaccinated HCWs in Shashemene Zonal Town (12.9%) was found to be lower when compared to WHO’s estimation.

The result in this study was found to be higher when compared to other studies done in Ethiopia. A study conducted among HCWs of Bulle Hora Woreda, Southern Oromia, revealed that only 2 (1.2%) of the HCWs were vaccinated [14], and a study in St. Paul’s and Zewditu hospital, Addis Ababa showed 9 (3.5%) had history of vaccination even though none of them were fully vaccinated [15]. A study in Harar also showed only 15 (4.7%) of the respondents were fully vaccinated against HBV [16]. Another study conducted in Bahir Dar City Administration, North West Ethiopia showed low vaccination status among HCWs. According to the study 20 (5.4%) reported that they were fully vaccinated [11]. The disparity in vaccination rates in our study compared to the others conducted in Ethiopia could be due to the time they were conducted. With the exception of the study in Bahir Dar, all studies mentioned were conducted 3–5 years prior to our study. Another reason could be the vaccination campaign done by the Zonal Health Bureau among Shashemene Referral Hospital stuffs in the year 2012.

Comparison of our result with a study conducted in Georgia on health professionals showed a proportional result in which, 37 (12%) of the HCWs had been vaccinated against HBV [17]. Otherwise, it is very low when compared to 106 (53%) of HCWs at a tertiary care hospital in Pakistan who had received the vaccine [18] and a study in Iraq in which 45% of the respondents were vaccinated [4]. Another study from Iraq also showed relatively high percentage of HCWs vaccination. It showed 71 (65.7%) of the respondents had been vaccinated, from which 40 (37%) had completed taking the vaccine [19]. A study in India also showed 35% of the laboratory technicians were fully vaccinated [20]. Another study in tertiary teaching hospital in Nigeria revealed 162 (50.6%) of the respondents had ever received the vaccine while 116 (36.25%) had completed [8]. A study in a tertiary care hospital in India showing, 146 (57.5%) of the HCWs were vaccinated [13] and a study in Kuwait showing, 74.7% of the respondents have actually ever received the vaccine while 84.0% of them had completed [21]. The reasons for the lower rate of fully vaccinated HCWs in our study compared with other studies cannot be completely discerned. However, the difference in demographic characteristics of the study population, the difference in availability of the vaccine in these countries, and preliminary benefits due to the initiation of national programs of immunization in other countries might explain these discrepancies.

Only one (1.5%) had checked their status of immunity after taking the vaccine in our study. This is comparable with study in Iraq, in which no one of the respondents had checked for status of immunity after vaccination [19]. This might be due to the absence of a laboratory test in the town and health institutions nearby to detect antibodies against HBV.

Reasons for not being vaccinated

Although a national hepatitis B vaccination policy is in place, and groups covered by this policy include infants and health care workers [22], is in place, remarkably majority (83.9%) of the respondents in our study have not been vaccinated. The reasons for not taking the vaccine by the respondents were, 165 (48%) because the vaccine was not found in nearby institution and 95 (27.6%) not delivered in the institution, which gives an overall 75.6% due to unavailability of the vaccine which is much higher than a study conducted in India showing 31.1% were not vaccinated due to same reason [13], 45 (13.1%) gave the reason ‘never thought about it’ which was lower than 21.7% of respondents of a study in Pakistan who gave same reason and the study also revealed, 24.8% of the respondents were not vaccinated because of no felt need [23] while in our study, only 1 (0.3%) of the respondents gave the reason ‘it was not important’. Studies conducted in India, Pakistan and Kuwait showed 13.3, 5.5 and 19.5% of the respondents respectively were not aware of the vaccine [13, 21, 23] while in our study 34 (9.9%) of the respondents were not aware of the vaccine, which was comparable with the study in India and about two times higher and lower than the studies in Pakistan and Kuwait respectively. Other reasons mentioned in our study were, 3 (0.9%) because it was expensive, and the remaining 1 (0.3%) always forget to get the vaccine which was much lower than a study conducted in India which showed 5.6% of respondents forgot to take vaccine [13].

Factors associated with vaccination status

Female respondents were about four times as likely (AOR 3.84 [95% CI 1.66, 8.92]) to be fully vaccinated as compared to the male HCWs in this study. Similarly, a study conducted at a tertiary care hospital in India showed higher vaccination rate among female HCWs (χ² = 6.41, df = 1, p < 0.02) [13]. The difference in vaccination status may be explained by greater risk behaviors in males than females, and women are more concerned to their health. In addition women have higher risk perception than men. The other possible explanation could be level of awareness since women were relatively more knowledgeable.

With increasing duration of work experience, there was a corresponding increase in full vaccination status. In our study, those who had a work experience 5–9 years were about six times more likely (AOR 5.88 [95% CI 2.24, 15.42]), and those who had ≥10 years were twelve times more likely (AOR 12.51 [95% CI 3.10, 50.41]) to be fully vaccinated than those who had a work experience of <5 years. A study in Nigeria also revealed the same pattern as our study, though the finding was not statistically significant (p = 0.323) [8]. This might be due to the fact that most of the HCWs were unvaccinated when entering health care services, and they become more likely to be vaccinated as they spend more time in the service because the health care institutions have no regular program of vaccination for the HCWs.

In general, this observation is worrisome because the most productive and economically viable age group of the populations were found in the groups with the least full vaccination status.

While none of the health care institutions found in the town have availed the vaccine, the type of institutions in our study that respondents come from have a significant association with vaccination status. Those who came from governmental health care institutions are two times more likely (AOR 2.45 [95% CI 1.10, 5.46) to be fully vaccinated than respondents from non-governmental health care institutions. This may be due to the fact that vaccination campaigns done in the town were only focused on the referral hospital stuff, and out of 241, majority 135 (56%) of the governmental institution respondents were from the referral hospital.