Hepatitis B discrimination in everyday life by rural migrant workers in Beijing

ABSTRACT

Background: In China, the hepatitis B virus (HBV) is a particularly challenging public health issue, with an estimated 90 million chronic hepatitis B carriers accounting for almost 7% of the population. Health-related discrimination can serve as a barrier to prevention and care for infectious diseases, such as HBV, degrade the HBV sufferers' quality of life and limit HBV patients' employment opportunities. While rural migrants account for up to 40% of the total urban population in the developed cities in China, there has been no study of the discrimination behavior of rural migrant workers toward HBV carriers.

Objective: This study evaluates the discrimination behavior of rural migrant workers toward HBV carriers and patients and proposes public policy recommendations to address discrimination and stigma.

Methods: The sample comprised 903 rural adults, aged over 18 years old, who migrated to Beijing. Using a face-to-face interview, we surveyed rural migrants' demographic characteristics, knowledge of HBV and discrimination against HBV carriers. Descriptive statistics were used to characterize the study population, HBV stigma and knowledge of HBV. Three discrimination levels (no-mild, medium and severe discrimination) were modeled using multiple logistic regression.

Results: Rural migrants to Beijing had a mean age of 36 years, were overwhelmingly married (91.58%), mostly with a junior high school or lower education (78.05%) and mainly engaged as temporary workers (42.52%) or self-employed (33.78%). Only 30.56% reported that they had been vaccinated against HBV. On the 0–10 discrimination scale, rural migrants rated 6.24, with only 4.54% displaying no sign of HBV-related discrimination. The high discrimination score occurred alongside a low mean knowledge of HBV (7.61 on the 1–22 ranking of HBV knowledge). Multiple logistic regression results suggest an inverse relationship between discrimination levels and HBV knowledge, especially knowledge about treatment and transmission routes. The “fear of being infected with HBV” and being HBV vaccinated was positively associated with HBV-related discrimination. Unemployed rural migrants were more likely to exhibit severe HBV-related discrimination than other occupational groups. Personal attributes, such as gender, age, marital status and education level were not associated with the level of discrimination.

Conclusions: Knowledge of HBV, its transmission and treatment, and the fear of HBV infection were key features in understanding HBV discrimination by rural migrant workers. To reduce discrimination, HBV public health education campaigns need to focus on both knowledge about HBV and the fear of HBV infection. Such campaigns should target rural migrant subgroups, such as unemployed rural migrant workers.

KEYWORDS:

• cognition
• discrimination
• HBV vaccination
• migration

Introduction

The hepatitis B virus (HBV) is a common and prevalent infectious disease of the liver, having affected an estimated 2 billion individuals worldwide.¹ In 2012, there were an estimated 350–400 million carriers of chronic hepatitis B infection.² In China, HBV remains a particularly challenging public health issue, with an estimated 90 million chronic hepatitis B virus carriers, which accounts for almost 7% of the Chinese population (WHO, 2015). Every year, an estimated 300000 people die from HBV-related diseases in China.³

Besides the direct health effects of HBV infection, HBV patients also suffer from discrimination.⁴ HBV discrimination has negative health outcomes, especially related to not seeking treatment and mental health impacts, and non-health poor quality of their life effects.⁵ Perceived HBV stigma leads to high levels of anxiety and exaggerated fear of transmission, and it can be a major cause of social isolation and reduced intimacy in relationships.⁶ The economic impacts from HBV discrimination occurs when HBV sufferers are denied employment, with costs incurred by both the individual and the wider economy. Since 2007 the Chinese Government has taken steps to eradicate discrimination against HBV carriers in employment and education,^7-9 but HBV-related discrimination remains common. Due to discrimination and the stigma of HBV, HBV sufferers frequently hide their health status, failing to seek HBV health care. A survey of HBV-infected Asian immigrants in Canada found that 53% were unwilling to discuss their illness with family or friends ¹⁰ and 70% of university students in Taiwan expressed fear of revealing their HBV status to friends.¹¹ Discrimination and HBV stigma underpinned the reluctance of HBV carriers to access HBV screening, diagnosis, care and treatment services and also a barrier to adopting prevention behaviors.^2,12

There have been limited studies of discrimination against HBV patients. Cultural values and social-cognitive factors are known to affect HBV-related discrimination,¹³ and discrimination studies have identified education as an important stigma-reduction intervention.¹ For China's rural population, knowledge deficits about HBV have been identified as a key factor in HBV discrimination.¹⁴ Surveys of Taiwanese college students ¹⁵ and HBV-infected Asian immigrants in Canada ¹⁰ also found that a lack of knowledge about transmission contributed to the prevalence of the disease among unimmunized immigrants. Surprisingly, we know little of the attitudes of urban residents in China toward HBV patients and carriers.

China's urban population can be bifurcated between urban-born residents and rural migrants. Over the past 15 years, China has experienced a growing rate of rural to urban migration, with 53% of the population urban in 2013 compared to just 36% in 2000.¹⁶ Today, there are an estimated 270 million rural migrants in China's cities, with forecasts that rural-to-urban migration will increase to 300 million by 2025. Ranked third in China as a rural worker destination, 38% of Beijing's 2013 population of 21.1 million is accounted for by migrant workers.¹⁶ With higher rates of chronic hepatitis B infection than the urban-born population,² and different age, gender, education and attitudinal characteristics than urban-born residents, rural migrants form a unique urban subgroup. For this rural migrant subgroup in Beijing, our study evaluates the factors accounting for discrimination against HBV patients and carriers in their everyday life, and suggests policy responses to attenuate such discrimination.

Results

The demographic characteristics of the study population are detailed in the Table 1. The mean age was 36 years, with 72.09% in the 24–44 age group, and given that 91.58% were married, the respondents were evenly balanced between male and female respondents. Roughly 60% of the respondents had between 6 and 9 years education, with 42.52% temporary workers, 33.78% self-employed and 19% unemployed. Only 30.56% reported that they had been vaccinated against HBV.

Table 1. Descriptive statistics for independent variables (N=903).

Variable	Variable definition	n	%
Male	1 if male; 0 otherwise	459	50.83
Female	1 if female; 0 otherwise	444	49.17
Age 18–24	1 if aged 18–23; 0 otherwise	69	7.64
Age 24–45	1 if aged 24–44; 0 otherwise	651	72.09
Age 45-	1 if aged 45-; 0 otherwise	183	20.27
Low education	1 if years of schooling<=6 years; 0 otherwise	179	19.82
Medium education	1 if 6< years of schooling<=9 years; 0 otherwise	526	58.25
High education	1 if years of schooling>9years; 0 otherwise	198	21.93
Unmarried	1 if unmarried; 0 otherwise	65	7.2
Married	1 if married; 0 otherwise	827	91.58
Divorced or widowed	1 if divorced or widowed; 0 otherwise	11	1.22
Temporary workers	1 if temporary workers; 0 otherwise	384	42.52
Steady workers	1 if village doctor, teachers or public officers; 0 otherwise	42	4.65
Self-employed people	1 if self-employed people; 0 otherwise	305	33.78
Unemployed	1 if unemployed; 0 otherwise	172	19.05
Household income group 1	1 if household income in the bottom quintile; 0 otherwise	175	19.38
Household income group 2	1 if household income in the second lowest quintile; 0 otherwise	167	18.49
Household income group 3	1 if household income in the middle quintile; 0 otherwise	165	18.27
Household income group 4	1 if household income in the second lowest quintile; 0 otherwise	207	22.92
Household income group 5	1 if household income in the top quintile; 0 otherwise	189	20.93
No fear of being infected with HBV	1 if no fear of being infected; 0 otherwise	485	53.71
Fear of being infected with HBV	1 if fear of being infected; 0 otherwise	418	46.29
Without vaccination history	1 if received 0 dose of HBV vaccine; 0 otherwise	627	69.44
Received vaccination	1 if received at least 1 dose of HBV vaccine; 0 otherwise	276	30.56
Knowledge of symptoms 1	1 if identify 0 true symptoms; 0 otherwise	440	48.73
Knowledge of symptoms 2	1 if identify 1/2true symptoms; 0 otherwise	236	26.14
Knowledge of symptoms 3	1 if identify >=3 true symptoms; 0 otherwise	227	25.14
Knowledge of consequence 1	1 if identify 0 true knowledge of consequence; 0 otherwise	184	20.38
Knowledge of consequence 2	1 if identify 1/2 true knowledge of consequence; 0 otherwise	584	54.49
Knowledge of consequence 3	1 if identify 3 true knowledge of consequence; 0 otherwise	135	14.95
Knowledge of treatment route 1	1 if identify 0 true route of treatment; 0 otherwise	290	32.12
Knowledge of treatment route2	1 if identify 1 true route of treatment; 0 otherwise	398	44.08
Knowledge of treatment route3	1 if identify>=2true route of treatment; 0 otherwise	215	23.81
Knowledge of transmission route 1	1 if identify 0 true route of transmission; 0 otherwise	210	23.26
Knowledge of transmision route 2	1 if identify 1/2 true route of transmission; 0 otherwise	278	30.79
Knowledge of transmission route 3	1 if identify>=3 true route of transmission; 0 otherwise	415	45.96

Figure 1 summarizes the participants' attitudes toward HBV patients and carriers. The mean score on the stigma scale was 6.24 (range 0-no to 10-highest discrimination), with higher scores indicating greater stigma. Taking all the participants, only 43.41%, were willing to accept gifts, 47.73% willing to hug and shake hands and 31.78% willing to have dinner with HBV patients or carriers. Even fewer parents (22.81%) were willing to let their children play with hepatitis B-infected children, and only 6.42% parents thought their children should marry HBV patients or carriers.

Figure 1. Attitude toward hepatitis B patients and carriers.

Table 2 shows the HBV-related discrimination distribution. Based on 5 discrimination questions, the median discrimination score was 6, with only 4.54% of the respondents displaying no sign of HBV-related discrimination. Almost one third (30.12%) of participants scored at the highest discrimination level, indicating that they were unwilling to have any contact with hepatitis B patients or carriers. According to their HBV-related discrimination index scores, the participants were divided into 3 groups, as shown in Table 3. Roughly one quarter (24.03%) of the participants displayed no-mild discrimination; 32.34% medium discrimination; and 43.63% severe discrimination.

Table 2. Hepatitis B Discrimination Index Score.

Score	Freq.	Percent	Cum.
0	41	4.54	4.54
1	57	6.31	10.85
2	75	8.31	19.16
3	44	4.87	24.03
4	85	9.41	33.44
5	70	7.75	41.20
6	87	9.63	50.83
7	50	5.54	56.37
8	79	8.75	65.12
9	43	4.76	69.88
10	272	30.12	100.00
Total	903	100.00	100.00

Table 3. Hepatitis B Discrimination Level of Participants with Different Characteristics (N=903).

	Mild/without (217, 24.03%)		Medium (292, 32.34%)		Severe (394, 43.63%)
Variable	n	%	n	%	n	%
Male	115	25.05	145	31.59	199	43.36
Female	102	22.97	147	33.11	195	43.92
Age 18–24	24	34.78	22	31.88	23	33.33
Age 24–45	149	22.89	255	34.56	277	42.55
Age 45-	44	24.04	45	24.59	94	51.37
Low education	40	23.35	43	24.02	96	53.63
Medium education	113	21.48	181	34.41	232	44.11
High education	64	32.32	68	34.34	66	33.33
Unmarried	23	35.68	19	29.23	23	35.38
Married	191	23.10	268	32.41	368	44.50
Divorced or widowed	3	27.27	5	45.45	3	27.27
Temporary workers	110	28.65	117	30.47	157	40.89
Steady workers	12	28.57	12	28.57	18	42.86
Self-employed people	67	21.97	118	38.69	120	39.34
Unemployed	28	16.28	45	26.16	99	57,56
Household income group 1	39	22.29	59	33.71	77	44.00
Household income group 2	50	29.94	57	34.13	60	35.93
Household income group 3	28	16.97	55	33.33	82	49.70
Household income group 4	44	21.26	72	34.78	91	43.96
Household income group 5	56	29.63	49	32.34	84	44.44
No fear of being infected with HBV	195	40.21	192	39.59	98	20.21
Fear of being infected with HBV	22	5.26	100	23.92	296	70.81
Without vaccination history	159	25.36	223	35.57	245	39.07
Received vaccination	58	21.01	69	25.00	149	53.99
Knowledge of symptoms 1	100	22.73	131	29.77	209	47.50
Knowledge of symptoms 2	70	29.66	79	33.47	87	36.86
Knowledge of symptoms 3	47	20.70	82	36.12	98	43.17
Knowledge of consequence 1	42	22.83	53	28.80	89	48.37
Knowledge of consequence 2	148	25.34	188	32.19	248	42.47
Knowledge of consequence 3	27	24.03	292	32.34	394	43.63
Knowledge of treatment route 1	46	15.86	82	28.28	162	55.86
Knowledge of t treatment route 2	118	29.65	135	33.92	145	36.43
Knowledge of treatment route 3	53	24.65	75	34.88	87	40.47
Knowledge of transmission route 1	40	19.05	45	21.43	125	59.52
Knowledge of transmission route 2	78	28.06	91	32.73	109	39.21
Knowledge of transmission route 3	99	23.86	156	37.59	160	38.55

Table 3 also displays the characteristics of the HBV discriminators by the characteristics of the participants. Married, unemployed, older participants and those with less than a senior high school education were most likely to be severe discriminators. Surprisingly, high education participants were evenly spread across the 3 discriminator groups. There was no difference in the severe discriminators by household income group. Participants who were vaccinated (53.99%) and had a fear of HBV infection (70.81%) were over-represented in the server discriminator group. Participants will the lowest level of knowledge of HBV symptoms (47.50%), consequences (48.37%), treatment (55.86%) and transmission (59.52%) were in the severe discrimination group.

Table 4 shows the results of the multiple logistic regressions of medium and severe HBV discrimination levels against the explanatory factors, relative to the no-mild discrimination base.

Table 4. Multiple Logistic Regression of Hepatitis B Discrimination Level against Explanatory Factors (N=903).

	Medium discrimination				Severe discrimination
	Coef.	SE	RRR	P	Coef.	SE	RRR	P
Female	0.112	0.199	1.127	0.550	−0.024	0.221	0.976	0.913
Age 24–45	0.208	0.407	1.231	0.609	0.753	0.463	2.125	0.104
Age45-	−0.305	0.473	0.737	0.519	0.635	0.528	1.888	0.229
Medium education	0.507	0.281	1.661	0.071	0.075	0.295	1.079	0.797
High education	0.107	0.332	1.111	0.750	−0.446	0.359	0.640	0.214
Married	0.202	0.424	1.224	0.634	0.183	0.479	1.201	0.703
Divorced or widowed	0.366	0.899	1.443	0.684	−1.152	1.101	0.316	0.296
Steady workers	0.101	0.488	1.106	0.836	−0.824	0.534	1.086	0.877
Self-employed people	0.272	0.220	1.941	0.003	0.381	0.247	1.464	0.124
Unemployed	0.657	0.304	1.619	0.114	0.865	0.317	2.377	0.006
Household income group 2	−0.355	0.312	0.701	0.254	−0.488	0.353	0.613	0.166
Household income group 3	0.077	0.336	1.080	0.819	0.110	0.371	1.116	0.768
Household income group 4	0.007	0.306	1.198	0.555	0.505	0.337	1.658	0.133
Household income group 5	−0.708	0.318	0.492	0.026	−0.234	0.343	0.791	0.494
Fear of being infected with HBV	1.57	0.266	4.851	0.000	3.453	0.272	31.623	0.000
Received vaccination	−0.211	0.234	0.809	0.364	0.753	0.248	2.123	0.002
Knowledge of symptoms 2	−0.045	0.240	0.956	0.850	−0.239	0.265	0.788	0.365
Knowledge of symptoms 3	0.381	0.282	1.465	0.176	0.501	0.312	1.651	0.108
Knowledge of consequence 2	0.001	0.274	1.002	0.995	0.988	0.291	1.104	0.734
Knowledge of consequence 3	0.176	0.369	1.193	0.632	0.038	0.400	1.039	0.923
Knowledge of treatment route 2	−0.589	0.262	0.555	0.024	−1.044	0.278	0.352	0.000
Knowledge of treatment route 3	−0.576	0.321	0.562	0.073	−0.787	0.341	0.455	0.021
Knowledge of transmission route 2	−063	0.322	0.938	0.843	−0.905	0.328	0.404	0.006
Knowledge of transmission route 3	−0.269	0.323	1.309	0.405	−1.013	0.337	0.363	0.003

Note. The basis outcome is mild or without discrimination.

Discussion

Our data on migrant workers in Beijing were significantly different from the characteristics of China's rural workers. According to Yu et al.,⁴ 46.65% of rural workers were in the 28–48 age group; 43.87% had between 6 and 9 years education and 41.8% had less than 6 years education; and 65.80% were farmers and 13.48% temporary workers. Only 19.01% of rural workers reported that they had been vaccinated against HBV. Therefore, our rural migrant population was younger, better educated, in different occupations and had higher HBV vaccination rates than rural workers. Given these different characteristics between rural workers and rural migrant workers, and the significant proportion of the urban population comprised of rural migrants, it is important to understand the behavior of rural migrant workers toward HBV patients and carriers.

Compared with Yu et al.'s data on China's rural population, our participants' rated higher levels of HBV stigma across all questions in Figure 1, except for parents allowing their children to marry hepatitis B patients or carriers.⁴ The main findings were: (1) HBV-related discrimination by rural migrants was serious; (2) the fear of HBV infection was significantly associated with migrants' HBV-related discrimination; (3) the knowledge of the treatment and transmission routes were also main factors associated with rural migrant workers' HBV-related discrimination; and (4) unemployed rural migrants were more likely to evidence severe HBV-related discrimination than other occupational groups. Since discrimination degrades individual's quality of life, our results mean that HBV patients and carriers require social support and care. Specifically, our results show that discrimination against HBV patients and carriers in Beijing by rural migrants was widespread. Given the substantial proportion of migrants in China's urban population, HBV discrimination is an important health and social issue in China.

Previous studies ^4,6,13 found that knowledge about HBV was low in rural Chinese populations, Asian immigrants ^18,19 and Chinese immigrants.^12,18,20 In our study, HBV-related knowledge consisted of knowledge about HBV symptoms, consequences if untreated, treatment regimes and transmission routes. On the total HBV knowledge scale (1–22), the mean score was 7.61, with higher scores indicating greater HBV knowledge levels. Roughly one quarter of our participants knew no HBV transmission routes, and only 15.9% of participants could answer all transmission routes. For example, 64% of participants knew the mother to child route, 60.02% unclean medical or dental equipment, 40.86% unprotected sex, 32.56% sharing shaving equipment with an infected person and 29.67% unhygienic tattooing or ear-piercing. Importantly, 36% of the participants did not know the mother to child route, but mother-to-child transmission is the leading cause of HBV infection in China.¹⁷

HBV stigma was greatest in the fear of contagion domain.¹² We found that fear of being infected with HBV was significantly associated with rural migrant worker HBV-related discrimination (See Table 4). Most previous qualitative studies showed that discrimination stemmed from lack of knowledge about HBV and the way it is transmitted rather than fear of HBV infection. However, Shen found that the fear of HBV infection risk was the primary cause of HBV-related discrimination,²³ which was consistent with our findings. In our quantitative study, the coefficient of the “fear of being infected with HBV” was 3.45 (see Table 4), with a high relative risk ratio (31.63) compared to the reference group of “no fear of being infected.” Therefore, the “fear of being infected with HBV” was the most important cause of HBV-related discrimination by rural migrant workers. Based on previous research,⁶ an effective way to reduce the fear and misunderstanding of HBV that leads to discrimination against HBV patients and carriers is through public education. Public education needs to provide not only scientific knowledge about HBV, but also address the fear of HBV infection.

The evidence shows that rural migrants' knowledge of HBV transmission routes were poorly understood. Previous studies have noted that ignorance about HBV, especially misunderstandings about its transmission, is the main cause of discrimination.^7,21 Similarly, our study confirmed that migrants' knowledge about transmission was significantly associated with decreased HBV-related discrimination (See Table 4).

We also found knowledge about treatment significantly decreased rural migrants' HBV-related severe discrimination. We identified 5 treatment routes: medication (55.15%), injections (46.73%), fumigation (3.77%), dietary changes (20.93%) and herbal medicine (24.03%). Importantly, 31.78% of the participants had no knowledge about any treatment. Specifically, 51.03% of participants had fear of infection in the no-knowledge-of-treatment group compared with 44.65% in high-knowledge-of-treatment group. We posit that more correct knowledge about treatment can reduce the fear of being infected with HBV.

We also found that the level of discrimination was not associated with knowledge about symptoms and HBV consequences if there was no treatment. Participants who had more correct knowledge about HBV symptoms had the similar fear of being infected with HBV (47.58%) as those without knowledge of HBV symptoms (47.27%). Further, we found participants who had more knowledge about HBV consequences if there was no treatment had a higher fear of infection (51.11%) than those without knowledge of HBV consequences (43.58%).

The evidence on improving HBV knowledge as a way of decreasing HBV discrimination is mixed. Increasing migrants' knowledge about transmission routes may significantly reduce the probability of HBV-related discrimination and more correct knowledge about treatment can reduce the fear of being infected with HBV. This suggests that appropriate public policy interventions, such as education campaigns, can reduce the stigma of being an HBV patient and the discrimination that goes with it. However, those with more knowledge of HBV symptoms and HBV consequences if untreated had a higher fear of HBV infection. This suggests that public health campaigns need to be carefully targeted, providing not only greater knowledge about HBV, but also reducing the fear of HBV infection.

Gender was not associated with discrimination by rural migrant workers against HBV patients in Beijing. This finding is consistent with a previous study in Japan that also found no association between gender and prejudice toward HIV, HBV or HCV-infected people.²² No previous study has assessed the influence of education on HBV-related discrimination, although a previous study found Chinese providers with a higher medical education tended to show high levels of discrimination against HIV patients.^21,23 We hypothesized that rural migrant workers' education status was associated with HBV-related discrimination, but our results did not confirm this hypothesis. Education status was not significant in our regressions (see Table 4).

We also investigated the impact of occupation on the level of HBV-related discrimination. Unemployed rural migrants were more likely to exhibit severe HBV-related discrimination compared with rural migrants who were temporary workers. Compared with temporary workers, unemployed rural migrants had a lower level of knowledge about HBV transmission routes and a higher fear of being infected with HBV. Specifically, 27.91% of participants had no knowledge about HBV transmission routes in the unemployed group compared with 22.30% in temporary workers group. Similarly, 55.23% of unemployed participants had a fear of being infected with HBV, but only 42.3% in temporary workers group.

Similar to other studies, our model found that income was not associated with rural migrant workers' HBV-related discrimination. This result confirms the results from a study of Japan's working population,²² although the Japanese study used individual income, whereas we used household income. In contrast to Yu et al's. finding that age was positively associated with rural workers' severe HBV-related discrimination,⁴ our findings revealed that age was not associated with severe HBV-related discrimination.

HBV vaccination can effectively protect vulnerable people from HBV infection.^24,25 Our study found that 51.81% of participants who received a vaccination were likely to display severe HBV-related discrimination compared to 43.86% that were unimmunized. This result contrasts with other studies,^4,12 where participants who have received HBV vaccination displayed reduced HBV-related discrimination by decreasing their fear of contagion. We speculate that since immunized participants were more aware of the consequences of HBV, they displayed more fear of contagion than unimmunized participants. This again suggests that HBV public health campaigns need to education the population on the fear of HBV infection as well as knowledge about HBV.

The study has a number of limitations. First, discrimination was evaluated in a convenience sample in a single city at one time point. Of course, Beijing was in the top 3 cities in China to receive rural migrants, but additional research will be needed to validate the survey in other cities and to evaluate whether changes in discrimination can be identified over time. Second, the measurement of HBV-related discrimination in the real world is not limited to people's attitudes toward the 5 questions assessed in our questionnaire. Further research will expand the number and types of questions to more extensively investigate attitudes and behaviors toward HBV patients and carriers. Third, we did not assess the factors that may lead to rural migrant workers fear of infection.

In spite of these limitation, our findings show that HBV-related discrimination by rural migrant workers was extensive, with serious quality of life impacts. The fear of HBV infection by rural migrants was the main factor associated with HBV-related discrimination, and the lack of knowledge about treatment and transmission routes were also significant discrimination factors. In addition, we found rural unemployed migrants were more likely to exhibit severe HBV-related discrimination compared with temporary workers. We submit that the government has a key role in public health education to increase the knowledge and reduce the fear about HBV. Knowledge of HBV may be insufficient to address HBV discrimination without an equal emphasis on the fear of HBV infection. Subsets of the migrant population, specifically rural unemployed migrants, are key targets in a public education campaign.

Participants and methods

Study participants and sampling method

Our survey data were collected as part of a larger project on the impact of user fees on HB vaccination coverage rates in China, funded by the Research Council of Norway and implemented by Shandong University and the University of Oslo. Using face-to-face interviews, data were collected on 995 rural migrants' demographic characteristics, knowledge of HBV and discrimination against HBV carriers. A rural migrant was defined as worker who worked at least 6 month in Beijing, with a registered residence location (Hukou) in another province. We excluded 92 rural migrants who had never heard of HBV, leaving a sample of 903 adults, aged over 18 years old, who migrated to Beijing. The participants' characteristics are detailed in Table 1, and all independent variables are explained below. The interview questionnaire consisted of questions on demographics, attitudes toward HBV patients and carriers, knowledge about HBV and individual vaccination history. Following a pilot survey, well-trained staff administered the face-to-face interviews.

Definition and measurement of dependent variables

We defined HBV-related discrimination as negatively judging and unfairly treating hepatitis B patients or HBV carriers and/or such judgments and treatment being a result of the patients' or carriers' HBV infection status. HBV-related discrimination was divided into 2 general types: discrimination in an employment or school enrollment situations and discrimination in everyday life by strangers, neighbors, co-workers, friends or family, and more distant relatives.⁴ In this study, we focused only on HBV-related discrimination in everyday life.

The HBV-related discrimination levels were the dependent variables. Based on 5 behavior questions about HBV, the sample was divided into 3 groups: no-mild discrimination, medium discrimination, and severe discrimination. The behavior questions were: “Are you willing to accept gifts from hepatitis B patients or carriers?;” “Are you willing to shake hands with or hug them?;” “Are you willing to have dinner with them?;” “Do you think parents should let their children play with hepatitis B-infected children?;” and “Do you think parents should accept their child marrying a hepatitis B-infected person?.” Participants were given 3 options, “yes” (0), “it depends” (1) and “no” (2). Their answers were summed over the 5 questions for each participant, with higher scores indicating higher levels of sigma. Finally, participants were then categorized in 3 HBV-related discrimination levels: no-mild discrimination scores 0–3, medium discrimination scores 4–7, and severe discrimination scores 8–10. The higher the discrimination level, the more unwilling participants were to make contact with hepatitis B patients or carriers.

Definition and measurement of independent variables

Table 1 presents descriptive statistics for all independent variables. Based on U.N. categories,² participants were divided into 3 age groups: the adolescence aged 18–23 years; middle 24–45 years; and mature above 45 years. Participants were also divided into 3 groups based on education status: low—primary education and below; medium—junior high school education ; and high—senior high school education. There were 3 marital status groups: unmarried ; married; and divorced or widowed. We created 2 groups based on vaccination status: those without vaccination history and those who had received one or more doses of the HBV vaccine, the “received vaccination” group.

An income variable, measured at the household level, was defined by the average annual income of all household members during the past 5 years. Five dichotomous variables (Income groups 1–5) were defined to indicate how each participant's income was related to the quintiles of income distribution. Five quintile income groups were defined broadly taking the values RMB20000, RMB36000, RMB48000, and RMB65000. Since several household incomes fell exactly on the cutoff points, there were slight adjustments around the cutoff values. Occupations were divided into 4 groups: steady workers (such as village doctors, teachers or public officers); temporary workers; self-employed; and unemployed.

Participants were asked to identify HBV symptoms. Nine symptoms were listed, including general physical discomfort, jaundice(skin becomes yellow), jaundice (whites of the eyes become yellow), fever, anepithymia, nausea, emesis, body aches, and dark urine. The number of symptoms identified was used to create 3 “knowledge of symptoms” variables: no knowledge of symptoms; knowledge of 1–2 symptoms; and knowledge of more than 3 symptoms. A similar process was used to create 3 variables for “knowledge of consequences.” Participants were asked to identify the 5 HBV treatments, comprising medication, injections, fumigation, dietary changes and herbal medicine. We counted the number of treatment routes identified by each participant to create a new variable, “knowledge of treatment routes” based on no, one and more than one treatment route identified.

Participants also identified 5 HBV transmission routes: mother to child, unclean medical or dental equipment, unprotected sex, unhygienic tattooing or ear-piercing, and sharing shaving equipment with an infected person. Three “knowledge of transmission routes” variables were created based on no, 1–2 and more than 2 routes identified. Finally, participants were asked whether they felt fearful and worried about being infected with HBV while spending time with hepatitis B patients and carriers. Two groups were created: “fear of being infected with HBV” and “no fear of being infected with HBV.”

Statistical analyses

Descriptive statistics were used to characterize the study population, HBV stigma and HBV knowledge. HBV-related discrimination levels were divided into 3 groups: no-mild discrimination group, medium discrimination group, and severe discrimination group. Multiple logistic regression analyses were used to assess the associations between each independent variable and HBV-related discrimination. The no-mild discrimination group was the reference group. Relative risk ratios (RRR) were used to measure the strength of association between each independent variable and HBV-related discrimination level. All data were double-input using Microsoft Access and checked for consistency. Statistical analyses were performed using STATA 12.0. A two-tailed p value of 0.05 was considered statistically significant.

Ethics

Participants were informed that they could refuse to answer any question. The questionnaire did not ask about infection status, and no biological samples were collected. The project was approved by the Medical Ethics Committee at the Shandong University School of Medicine (Grant No. 201001052).

Disclosure of potential conflicts of interest

None of the authors has any conflicts of interest.

Acknowledgments

The authors thank the health workers in all the participating provinces for their support and assistance.

Funding

This work was supported by the Norwegian Research Council (Project no. 196400/S50).