Analysis of the association between areal socioeconomic deprivation levels and viral hepatitis B and C infections in Japanese municipalities

Hepatitis B virus (HBV) and hepatitis C virus (HCV) infections are major global public health issues, with an estimated a few hundred million infected people for each virus [1, 2]. HBV and HCV are major causes of liver diseases, such as cirrhosis and cancer [3]. The prevalence of HBV declined from the 1990s to the 2000s, possibly because of increased immunization [4]. However, there are regional differences in HBV and HCV infection worldwide [1, 4]. In Japan, administrative measures, such as HBV screening of all pregnant women and national hepatitis screening of the general population, have been implemented in recent decades for eliminating hepatitis virus infection [5]. As a result, the prevalence of HBV and HCV has decreased over Japanese birth cohorts [5]. Mortality related to HCV has also decreased in recent decades [6]. On the other hand, there are still a few million estimated carriers of HBV or HCV in Japan [5]. The early detection of undiagnosed patients and prevention of new infections are warranted.

Individual and regional socioeconomic status are related to HBV and HCV infection [7‐9], and the prevalence of the viruses is particularly high in lower- and middle-income countries [1, 10]. A study investigating factors associated with the spatial distribution of HCV prevalence in the Netherlands, found that one-person households, non-Western immigrants, and divorced persons comprised the high-risk population [11]. In addition, a study in India found that a lower socioeconomic status and illiteracy were associated with HCV positivity [12]. Moreover, a study in Turkey found that self-employed persons, business owners, and public sector workers were less likely to be infected with HBV compared with labourers [13].

In Japan, in contrast, differences in the prevalence of HBV and HCV depending on individual or regional socioeconomic status have not been investigated, and it is meaningful to verify the association also in Japan. One way to determining the association between socioeconomic status and virus prevalence is statistically analyzing the association between the prevalence and regional socioeconomic status. In Japan, there are studies investigating regional differences in the prevalence of HBV and HCV [5, 14], and high-prevalence regions or prefectures have been shown. On the other hand, differences in prevalence by municipality across Japan have not been investigated. We could more accurately analyze regional socioeconomic positions and disease burdens by revealing municipal differences in disease prevalence or mortality, as in previous studies [15, 16]. Socioeconomic deprivation level is often calculated using multiple socioeconomic characteristics of regions in ecological studies [17, 18]. If a disparity in prevalence related to areal socioeconomic level exists in Japan, effective administrative measures in more deprived areas will help reduce HCV and HBV prevalence. For example, more encouragement of hepatitis screening or hepatic function test in those regions is meaningful and recommended in this case. In Japan, although data on HBV prevalence are not publicly available, data on positivity of hepatitis B surface antigen (HBsAg) are publicly available.

In this study, we investigated the association between areal socioeconomic deprivation levels and the positivity of HBsAg and the prevalence of HCV using national screening data from Japanese municipalities.

Data from 1,665 for HBV screening and data from 1,668 municipalities for HCV screening were available, and the data from 1,665 municipalities were available for both the results of HBV and HCV screening from 2013 to 2017. We used the data from 1,660 municipalities in the analysis after removing municipalities where number of residents was zero or extremely small because of an evacuation following the Great East Japan Earthquake. The total number of participants in the data was 4,233,819 for HBV and 4,216,720 for HCV screening.

Figure 1 shows the geographic differences in the standardized HBsAg positive and HCV prevalence ratios in Japan. A base map of Japan indicating the name of prefectures is shown in the Supplementary Fig. 1. Municipalities with a high standardized HBsAg positive ratio tended to be seen often in Hokkaido and Okinawa. Many municipalities with a low standardized HCV prevalence ratio were seen, probably because the number of HCV carriers tended to be zero in many municipalities. Spearman’s correlation coefficient of the standardized HBsAg positive ratio and the standardized HCV prevalence ratio was 0.079 (p-value = 0.001), and there was a positive correlation between them.

Table 2 shows the basic characteristics of the data used in this study.

Table 3 shows the results of the principal component analysis, and the principal components 1–4 are shown. The eigenvalues of principal components 1 and 2 were relatively large. Principal component 1 can be interpreted as a component related to level of rurality because coefficients of variables related to urbanization tended to be below zero and large. In addition, principal component 2 can be interpreted as a component related to low socioeconomic status among individuals because only the coefficient of taxable income per capita is below zero. We used the first two principal components for deriving the socioeconomic deprivation level because the eigenvalues were higher than 1 for those principal components.

Table 4 shows the posterior means of relative risk and the 95% CIs obtained by spatial Poisson regression analysis according to type of socioeconomic deprivation level. The deprivation level derived from the sum of the scores of principal components 1 and 2 was significantly and positively associated with HBsAg positivity and HCV prevalence. The deprivation level derived from the score of principal component 2 was also significantly and positively associated with the outcomes, whereas that derived from the score of principal component 1 was not associated with the outcomes. Population density was significantly and positively associated with HBsAg positivity and HCV prevalence when using the deprivation level derived from the score of principal component 1 and that derived from the sum of the scores of principal components 1 and 2.

This study revealed that risks for HBsAg positivity and HCV prevalence were associated with the municipal socioeconomic deprivation level among individuals who have never participated in hepatitis screening. There was an association between socioeconomic deprivation level and both viruses. Although it is known that the risk of liver cancer mortality differed depending on municipal socioeconomic deprivation level in Japan [16], this is the first time the association between the results of hepatitis virus screening and socioeconomic deprivation level has been shown in Japan. In addition, the socioeconomic deprivation level based on the principal component that is interpreted as low socioeconomic status among individuals was associated with HBsAg positivity and HCV prevalence. Thus, it is considered that the socioeconomic status of individuals living in a municipality, rather than the municipal level of rurality, is positively associated with the outcomes.

Although the geographical distribution of the outcomes differed, there was a positive correlation between them among municipalities. This positive correlation is considered to be related to the association of both the outcomes with the municipal socioeconomic deprivation level.

We briefly discuss possible reasons for the association shown in this study. Firstly, it is considered that HCV or HBV infection tends to occur among persons with low socioeconomic status or living in deprived areas. In Japan, blood transfusion was a major infection route for HBV and HCV in the twentieth century [30, 31], and continuous use of injection equipments is also known to be related to infection of HBV [32]. The sanitary environment of areas is related to the infection rate of these viruses, and infection by blood transfusion and injection equipments are still a major concern in developing countries [33]. Therefore, there is a possibility the infection rate from blood transfusion or injections may have varied depending on areal socioeconomic deprivation levels in Japan in the twentieth century, whereas an evidence supporting this hypothesis does not exist. Sexually transmitted infection is another HBV infection route in Japan [30]. There is a socioeconomic disparity in diagnosing sexually transmitted infections [34, 35], and a similar disparity might also exist for HBV in Japan. Moreover, awareness or perception of the HBV vaccine may differ depending on socioeconomic status. In China and Korea, socioeconomic inequality in hepatitis B vaccination has been shown [36‐38], possibly due to income status and awareness of hepatitis B [36, 37]. However, infants of pregnant women whose HBsAg and Hepatitis B e antigen are positive began to be vaccinated by public expenses from 1986 in Japan [39]. Therefore, a disparity in vaccination rate depending on areal socioeconomic deprivation level might not be true for Japan.

An another explanation of the disparity, the treatment rate or awareness of infection for hepatitis viruses may differ depending on the socioeconomic deprivation level. In Japan, screening for hepatitis viruses is also conducted in workplaces [40], and implementation of the screening depends on the workplaces. Opportunities for participation in this screening may differ according to the type of occupation or workplace, and employees in large companies or of high socioeconomic status might have more opportunities to be screened. New methods of HCV treatment, such as direct-acting antivirals, have been developed in recent decades [41]. Therefore, if the hepatitis virus is detected early in workplaces, patients can receive treatment early. In addition, according to a study in Korea, self-awareness and family awareness of hepatitis virus infection differed depending on socioeconomic status [42]. Persons with higher socioeconomic status might know of the infection risk of themselves or their family members and have an early examination. An increase in treatment and awareness of infection in persons with high socioeconomic status may have reduced the prevalence of HCV and HBV in less deprived areas in Japan.

Population density was also positively associated with HBsAg positivity and HCV prevalence. Although the reason is uncertain, an association between urbanization level and the prevalence of viruses differs across countries. A study in India found that anti-HCV prevalence was associated with living in a rural area [43]. HBsAg prevalence was also higher in rural areas in a study conducted in Madagascar [44]. In contrast, a study in the Netherlands indicated that HCV prevalence was clustered in urban areas [11].

This study showed that the deprivation level increase relative risk of HBsAg positivity and HCV prevalence among individuals who have not participated in the hepatitis screening. Hepatitis virus screening on a larger scale in deprived areas will increase the detection of undiagnosed carriers, and early detection and treatment of virus carriers can reduce liver cancer and cirrhosis in Japan. Nevertheless, people in those deprived areas might be less likely to participate in testing for the hepatitis virus. In the United States, it was shown that people with less education, lower income, and private health insurance were significantly less likely to be screened for HCV [45]. It is meaningful to conduct an epidemiological study investigating whether the participation rate for hepatitis virus screening differs depending on socioeconomic deprivation levels. If the participation rate is low in highly deprived areas, an awareness program for informing the population of the merits of participation in screening is also needed. In addition, as discussed, the rate of HBV vaccination or awareness of the hepatitis virus may differ depending on the deprivation level. It will be useful to investigate those issues in the future.

There are some limitations for this study. First, we could not obtain data on gender in this study. The HBsAg positivity and HCV carrier rates are higher in middle-aged or older men than in women in Japan [5]. Geographic differences might differ by gender. Secondly, this was an ecological study. A study investigating the individual socioeconomic status and seropositive rates is needed to verify the results of this study. Thirdly, the study subjects had never participated in hepatitis virus screening. Those who have participated in screening before or who have been treated for the hepatitis virus are not included in the data. In addition, participation in hepatitis screening is not mandatory, and individuals voluntarily apply for the screening, and the attributes of participants and nonparticipants may be different. Moreover, number of target persons of the screening (sum of the participants and persons who have not participated in the screening) for each municipality is uncertain, and we could not calculate the participation rate in each municipality. A similar epidemiological study using randomly chosen individuals from each municipality will also be meaningful in the future. Fourthly, socioeconomic deprivation level of a municipality can change based on methods for deriving the level. Another method for deriving areal deprivation index using the Census data has been proposed in a previous study [46]. We confirmed that a similar socioeconomic disparity in the results of hepatitis screening was also observed when using the deprivation index proposed in the previous study [46]. Fifthly, participants aged 40 years old is quite high in the national screening because the main target of the screening is those aged 40 years old.

We investigated the association between areal socioeconomic deprivation level and the HBsAg positivity and HCV prevalence using national screening data from Japanese municipalities. This ecological study, conducted using a spatial model, found that the municipal socioeconomic deprivation level was significantly and positively associated with HBsAg positivity and HCV prevalence among individuals who have never participated in hepatitis screening. Thus, participation in hepatitis virus screening is important and meaningful, particularly for areas with a high lower socioeconomic level in Japan.