Content uploaded by Miao He
Author content
All content in this area was uploaded by Miao He on Jul 25, 2020
Content may be subject to copyright.
ORIGINAL RESEARCH
Current risk factors for HIV infection among blood donors in seven
Chinese regions
Peibin Zeng ,
1
Jing Liu,
2
Chunhong Zhang,
3
Bowei Zhang,
4
Wei Liu,
5
Mei Huang,
6
Hongli Ma,
7
Yuwei Zhao,
8
Rong Guo,
9
Miao He,
10
Yu Liu,
10
Dan Liao,
11
Marian Sullivan,
11
Jingxing Wang,
10
Wei Cai,
12
Paul Ness,
2
Hua Shan,
12
for the International Component of the NHLBI Recipient Epidemiology and Donor Evaluation Study-III
(REDS-III)
BACKGROUND: In China, there is a rising concern on
the increasing trends of HIV infections in high-risk
groups, who make blood donations that might potentially
challenge the blood safety. Analyses on current risk
factors for HIV infection among Chinese blood donors
are urgently needed for developing effective strategies to
defer high-risk donors and to warrant the safety of the
blood supply.
STUDY DESIGN AND METHODS: We recruited
313 HIV-positive and 762 HIV-negative donors from
seven study sites in China and evaluated donor
demographic characteristics, current medical and
behavioral risk factors associated with HIV infection in a
case–control survey. Univariable analyses examined the
relationship between HIV infection and donor and
donation characteristics, medical and behavioral risks,
living conditions, and lifestyles. Multivariable logistic
regression analyses evaluated the association between
selected individual risks and HIV infection. Regression
tree analysis was used to select covariates correlated
with both HIV infection and individual risks and thus
need to be controlled for in logistic regression models.
RESULTS: Being a man who has sex with men was
associated with the highest odds of HIV infection. Not
using a condom, having sex with HIV-infected
individuals, having sex partners with sexually transmitted
diseases (STDs), having more than two concurrent sex
partners, or having an STD were all associated with
more than five times higher odds of having HIV. Having
remunerated sex was associated with a 2.4 increased
odds of having HIV infection.
CONCLUSION: High-risk sexual behaviors were among
the major risks for HIV infection among Chinese blood
donors.
Since 1998, the implementation of the Blood Dona-
tion Law and mandatory donation screening for
anti-HIV, anti–hepatitis C virus, hepatitis B surface
antigen, syphilis, and elevated alanine aminotrans-
ferase level has greatly enhanced the safety of the blood
ABBREVIATIONS: CDCC = China Data Coordinating Center;
GX-CDC = Guangxi Provincial Center for Disease Prevention and
Control; IM = intramuscular; IV = intravenous; MSM = men who
have sex with men; REDS-III = Recipient Epidemiology
and Donor Evaluation Study-III; RFQ = risk factor questionnaire;
STDs = sexually transmitted diseases; WB = western blot.
From the
1
West China School of Public Health and West China
Fourth Hospital, Sichuan University; the
8
Chengdu Blood Center;
the
10
Institute of Blood Transfusion, Chinese Academic of Medical
Science, Chengdu;
6
Mianyang Blood Center, Mianyang, Sichuan;
and the
3
Chongqing Blood Center, Chongqing; and the
4
Henan
Provincial Blood Center, Zhenzhou;
7
Luoyang Blood Center,
Luoyang, Henan; and the
5
Guangxi Center for Disease Prevention
and Control, Nanning, Guangxi; and the
9
Xinjiang Provincial Blood
Center, Urumqi, Xinjiang, China; the
2
Department of Pathology,
Johns Hopkins Medical Institutions, Baltimore, Maryland; the
11
RTI
International, Research Triangle Park, North Carolina; and the
12
Department of Pathology, Stanford University, Palo Alto,
California.
Address reprint requests to: Jing Liu PhD, Johns Hopkins
School of Medicine, North Wolfe Street, Baltimore, MD 21287, USA,
e-mail: jliu84@jhmi.edu; Miao He PhD, Institute of Blood Transfu-
sion, Chinese Academy of Medical Sciences, 26 Huacai Road,
Chengdu 610052, Sichuan, P. R. China, e-mail:
messa1022@126.com
Funding supported by US National Heart, Lung, and Blood
Institute under contract HHSN2682011000081.
Received for publication July 12, 2019; revision received
November 26, 2019, and accepted November 26, 2019.
doi:10.1111/trf.15659
© 2020 AABB
TRANSFUSION 2020;9999;1–8
TRANSFUSION 1
supply in China.
1
Transfusion-transmitted HIV infections
decreased from 17% in 2006 to 5.5% in 2009, to less than 1%
in 2013 among all newly diagnosed HIV infections.
1,2
Between
2014 and 2017, pilot testing and final implementation of
nucleic acid testing at blood centers further minimized the
residual risks of HIV infection via blood transfusion.
2
Yet the
threat of the HIV epidemic on the Chinese blood supply is
still ongoing. Between 2009 and 2011, the HIV confirmed-
positive rate among 34 million donations from 357 blood cen-
ters increased from 0.014% (2009) to 0.016% (2010), to 0.018%
(2011).
3
In the general population, sexual transmission
accounts for more than 90% of new HIV cases.
2,4
Increasing
trends of HIV infections in high-risk groups like men who
have sex with men (MSM), sex workers,
5–7
and migrant
workers have contributed greatly to the growing HIV epi-
demic in China.
8–10
However, the current Health Inquiry
Questionnaire used in blood donor screening do not include
questions addressing these high-risk groups.
More recently, an upsurge of HIV incidence rate as high
as 6.7 per 100 person-years among 18- to 25-year-old MSM
has aroused public health concern.
11–13
Young donors have
always been important sources of Chinaʼs blood supply.
Based on the donation databases of five participating blood
centers in our study, about 48.5% of 648,607 first-time donors
from five regional blood centers were 18 to 25 years old, and
16.5% of 627,937 repeat donations came from donors of the
same age group.
14
The increasing HIV infection among the
youth in general poses a threat to the safety of the current as
well as the future blood supply in China. A better under-
standing of new and changing risk factors associated with
HIV infection is critical in developing effective donor screen-
ing strategies that target specific risk factors among blood
donors and further reduce HIV residual risks in the context
of the evolving HIV epidemic in China. In the present study,
using a case–control survey design, we examined the current
risk factors associated with HIV infection among Chinese
blood donors. The risk factors we investigated include high-
risk sexual behaviors, living conditions, and lifestyles; medi-
cal history; and medical events in the 6 months before the
study donation as well as donorsʼlifetime.
MATERIALS AND METHODS
The study protocol was reviewed and approved by research
ethic committees at all participating institutions: Chinese
Institute of Blood Transfusion, Guangxi Provincial Center for
Disease Prevention and Control (GX-CDC), Johns Hopkins
Medical Institutions, RTI International, and FEI Systems.
Study population and questionnaire survey
All study sites were selected from regions with high HIV
prevalence where over 10,000 cumulative cases of individ-
uals living with HIV/AIDS in each region were reported.
1
The total reported HIV/AIDS cases in these regions
accounted for 83.5% of the total number of cases of people
living with HIV/AIDS in China.
1
We recruited anti-HIV
screening reactive blood donors from seven study sites,
including six regional blood centers located at the following
sites: Luoyang, Zhengzhou, Chongqing (Chongqing autono-
mous municipality), Urumqi (Xinjiang province), Mianyang,
and Chengdu. Participants from these regional blood centers
were donors recruited from their local blood donor pools. In
addition, we also recruited donors through GX-CDC.
We planned to enroll cases and controls at the desired
ratio of 1:2 to 1:3 from anti-HIV screening reactive donors.
The majority of donors’western blot (WB) confirmed HIV
status was unknown or not available to either the study per-
sonnel or the participants at the time of enrollment. With
an average of 29.6% WB confirmatory positive rate that var-
ies across participating blood centers, we expected moder-
ate to large case–control ratio variations across study sites.
Donors whose anti-HIV screening reactive donations were
WB confirmed positive were defined as “cases.”Donors
whose anti-HIV screening reactive donations were WB con-
firmed negative were included as “controls.”
Survey packages were mailed to all participants who
gave oral agreement for participation, including 1) study
information sheet; 2) risk factor questionnaire (RFQ); 3)
consent form; and 4) prestamped envelope. The RFQ
includes 44 questions with multiple-choice responses on
the following components mainly about lifetime behavior,
lifestyle, or medical events. Detailed information for survey
package and RFQ is provided in the Supporting Information
(S1, S2 and S3, available as supporting information in the
online version of this paper).
Data entry and data transfer
Online surveys were converted into electronic data and stored
at a secured FTP (File Transfer Protocol) site, accessible only
by the Recipient Epidemiology and Donor Evaluation Study-III
(REDS-III) China Data Coordinating Center under FEI Systems
(CDCC, located at Xiʼan, China). Data from paper surveys were
entered into tables at the FTP site by trained blood center staff
andsubmittedtotheCDCC.BothHIVconfirmatory testing
results and risk factor survey data were compiled, underwent
quality control for accuracy and logical errors at the CDCC and
then at RTI, the REDS-III Data Coordinating Center.
Statistical analysis
Responses to the survey questionnaire were first compared
between HIV-positive (cases) and -negative donors (con-
trols) in univariable analyses that used chi-square tests.
Pvalue less than 0.05 was defined as the statistical signifi-
cance level. For responses to each selected question
that was significantly associated with HIV infection in
univariable analysis, multivariable logistic regression analy-
sis was used to further assess its association with HIV infec-
tion while controlling for four covariates: study site and
2 TRANSFUSION
ZENG ET AL.
TABLE 1. Donor demographic and donation characteristics by case and control status*
Variables
Case/WB HIV+
†
,
total = 313 (n, %) Control/WB HIV−, total = 762 (n, %) p value
Site Chengdu 54 (17.2) 81 (10.6) <0.0001
Chongqing 64 (20.4) 154 (20.2)
Guangxi 116 (37.0) 246 (32.3)
Luoyang 13 (4.2) 95 (12.5)
Mianyang 23 (7.4) 33 (4.3)
Urumqi 15 (4.8) 71 (9.3)
Zhengzhou 28 (9.0) 82 (10.8)
Age, y 18-25 96 (30.7) 294 (38.8) 0.090
26-35 85 (27.2) 197 (26.0)
36-45 76 (24.2) 166 (21.9)
46-55 55 (17.6) 98 (13.0)
56+ 1 (0.3) 2 (0.3)
Sex Male 272 (87.2) 445 (58.5) <0.0001
Female 40 (12.8) 316 (41.5)
Education Middle school or less 129 (41.2) 214 (28.3) <0.0001
High school 76 (24.3) 165 (21.9)
Some college/technical college 58 (18.5) 142 (18.8)
Bachelor and above 50 (16.0) 234 (31.0)
Ethnicity Han 261 (83.4) 645 (85.6) 0.433
Zhuang 35 (11.2) 74 (9.8)
Hui 2 (0.6) 10 (1.3)
Other 15 (4.8) 25 (3.3)
Occupation Farming/Fishing/Forestry 16 (5.1) 43 (5.6) <0.0001
Migrant Workers 39 (12.5) 51 (6.7)
Students 45 (14.4) 186 (24.4)
Military/Government/Health
care employees
15 (4.8) 107 (14.0)
Commercial/Service 64 (20.5) 127 (16.7)
Self-employed 42 (13.4) 108 (14.2)
Other 92 (29.3) 140 (18.4)
Donor status First-time donors 175 (55.9) 400 (52.5) 0.308
Repeat donors 138 (44.1) 362 (47.5)
*Data are reported as numbers.
†WB = western blot.
TABLE 2. Responses to questions about major medical risks by case and control status*
Questions Responses
Case/WB HIV+,
total = 313 (n, %)
Control/WB HIV−,
total = 762 (n, %) p value
In the past 6 months, did you have any
injection (including intravenous [IV] and
intramuscular [IM] injections)?
1 = Yes 77 (24.6%) 106 (13.9) <0.0001
2 = No 228 (72.8%) 645 (84.7)
99 = Unknown/Missing 8 (2.6%) 11 (1.4)
How many times did you have injection(s)? 1 28 (36.4%) 64 (60.4) 0.002
2 12 (15.6%) 18 (17.0)
3 13 (16.9%) 11 (10.4)
4+ 24 (31.1%) 13 (12.2)
Have you ever received a blood transfusion? 1 = Yes 12 (3.8%) 22 (2.9) 0.025
2 = No 290 (92.7%) 731 (95.9)
99 = Unknown/Missing 11 (3.5%) 9 (1.2)
Have you ever had any dental cleaning? 1 = Yes 76 (24.2%) 133 (17.5) 0.036
2 = No 234 (74.8%) 622 (81.6)
99 = Unknown/Missing 3 (1.0%) 7 (0.9)
Have you ever been previously diagnosed with
hepatitis?
1 = Yes 6 (1.9%) 11 (1.5) 0.002
2 = No 292 (93.3%) 741 (97.2)
99 = Unknown/Missing 15 (4.8%) 10 (1.3)
Have you ever been previously diagnosed with
syphilis, gonorrhea, or any other sexually
transmitted disease (STD)?
1 = Yes 18 (5.7%) 3 (0.4) <0.0001
2 = No 286 (91.4%) 750 (98.4)
99 = Unknown/Missing 9 (2.9%) (1.2)
*Data are reported as numbers.
WB = Western blot.
TRANSFUSION 3
RISK OF HIV INFECTION AMONG CHINESE BLOOD DONORS
donorʼs sex, education, and occupation. The selection pro-
cedure involved a series of regression tree analyses and
backward variable selection procedures in logistic regres-
sion. Donorʼs age, first time versus repeat donor status,
donorʼs awareness of possible HIV infection (received possi-
ble infectious donation notification from blood center or
anti-HIV WB confirmed-positive status from local Center for
Disease Prevention and Control) were also evaluated as
candidate covariates but not selected based on results of
regression tree and logistic regression analyses after
adjusting for the four selected covariates.
All analyses were performed using computer software
(SAS Windows version 9.4, SAS Institute; https://www.sas.
com/en_us/legal/editorial-guidelines.html).
RESULTS
Between January 1, 2015, and December 31, 2017, we
contacted 3222 anti-HIV screening reactive donors from
all study sites. A total of 1835 (57%) of all contacted
donors gave oral consent to participate in the study, to
whom survey packets were mailed. Among them, 1248
(68.0%) completed the survey. We excluded donors with
WB indeterminate (n = 79; 6.3% of consented donors) and
missing results (n = 94; 7.5%) due to sample shipping and
handling errors and/or inadequate sample quantity.
A total of 1075 donors were included in the final analyses,
of which 313 (29.1%) were cases and 762 (70.9%) were
controls.
TABLE 3. Responses to questions about major behavioral risks by case control status*
Questions Responses
Case/WB HIV+,
total = 313 (n, %)
Control/WB HIV-,
total = 762 (n, %) p value
Have you ever used illegal oral or intranasal
drugs without doctorʼs prescription?
1 = Yes 4 (1.3) 5 (0.7) 0.025
2 = No 305 (97.4) 756 (99.2)
99 = Unknown/Missing 4 (1.3) 1 (0.1)
Have you ever lived with a person who was an
intravenous (IV) drug user?
1 = Yes 14 (4.5) 26 (3.4) 0.001
2 = No 250 (79.8) 673 (88.3)
99 = Unknown/Missing 49 (15.7) 63 (8.3)
Have you had more than 2 concurrent sex
partners of the opposite sex?
1 = Yes 90 (28.8) 45 (5.9) <0.0001
2 = No 221 (70.6) 716 (94.0)
99 = Unknown/Missing 2 (0.6) 1 (0.1)
(FOR MALE RESPONDENTS ONLY) In your
lifetime, have you ever had sex with another
male?
1 = Yes 82 (30.1) 5 (1.1) N/A
2 = No 184 (67.7) 439 (98.9)
99 = Unknown/Missing 6 (2.2) 0
(FOR MALE RESPONDENTS ONLY) In your
lifetime, how many times did you have sex
with males?
1-2 14 (17.1) 1 (20.0) N/A
3-5 18 (21.9) 2 (40.0)
6-10 15 (18.3%) 0
>10 35 (42.7) 2 (40.0)
(FOR MALE RESPONDENTS ONLY) In your
lifetime, how many male partners have you
had sex with?
1-2 22 (26.8) 1 (20.0) N/A
3-5 33 (40.2) 4 (80.0)
6-10 14 (17.1) 0
>10 13 (15.9) 0
(FOR MALE RESPONDENTS ONLY) In your
lifetime, how often do you or your sex
partner use a condom when you have sex
with male partner?
Never 7 (8.5) 1 (20.0) N/A
Sometimes 25 (30.5) 0
Half of time 6 (7.2) 1 (20.0)
Most of time 31 (37.8) 2 (40.0)
Always 13 (15.85) 1 (20.0)
(FOR MALE RESPONDENTS ONLY) In the
past 6 months, how many times did you
have sex with males?
1-2 32 (41.0) 2 (50.0) N/A
3-5 22 (28.2) 1 (25.0)
6-10 12 (15.4) 0
>10 12 (15.4) 1 (25.0)
Have you ever paid or received money or other
forms of remuneration for having sex
1 = Yes 54 (17.3) 47 (6.2) <0.0001
2 = No 252 (80.5) 713 (93.6)
99 = Unknown/Missing 7 (2.2) 2 (0.2)
In your lifetime, have you ever had a sex
partner who had a positive test for syphilis,
gonorrhea, or any other sexually transmitted
disease (STD)?
1 = Yes 8 (2.6) 6 (0.8) <0.0001
2 = No 194 (62.0) 716 (94.0)
99 = Unknown/Missing 111 (35.4) 40 (5.2)
In your lifetime, have you ever had a sex
partner who had been diagnosed with
HIV/AIDS?
1 = Yes 10 (3.2) 1 (0.1) <0.0001
2 = No 176 (56.2) 718 (94.2)
99 = Unknown/Missing 127 (40.6) 43 (5.7)
In the past 6 months, did you have a sex
partner who had been diagnosed with
HIV/AIDS?
1 = Yes 6 (60) 0 N/A
2 = No 1 (10) 0
99 = Unknown/Missing 3 (30) 1 (100)
*Data are reported as numbers.
N/A = not available; WB = Western blot.
4 TRANSFUSION
ZENG ET AL.
Univariable analysis results
Demographic characteristics of the 1075 donors by case
and control status are displayed in Table 1. The majority of
the cases were male (87.2%). A lower percentage of donors
with college-and-above education were found in the cases
as compared to the controls (p < 0.05). A higher proportion
of migrant workers and other unspecified occupations
were cases than controls respectively (p < 0.05).Conversely,
a lower proportion of students and military/government/
health care employees were cases than controls (p < 0.05).
Medical risk questions were about donorʼs lifetime and
past-6-month medical history and medical events, including
acupuncture, injection, surgery, dental cleaning, and endos-
copies (Table 2). Having an intramuscular (IM) or intrave-
nous (IV) injection in the past 6 months as well as having
four or more IM or IV injections were associated with being
HIV positive (p < 0.01). Meanwhile, 5.8% of cases versus
0.4% of controls had a previous diagnosis of STDs such as
syphilis or gonorrhea, suggesting an association between
other sexually transmitted infections and HIV-positive sta-
tus. Additionally, a higher percentage of cases than controls
reported “Yes”or “Unknown,”or did not respond to the
question about household contact with someone with either
HIV/AIDS or an unknown HIV/AIDS status (p < 0.01).
For donor responses to questions about high-risk
behaviors that were reportedly associated with HIV infection
(Table 3), significantly higher proportions of cases than con-
trols responded “Yes”to the following lifetime behavioral
risk questions: Having concurrent sex partners; paid or
received money or other forms of remuneration for having
sex; and had a sex partner who had other sexually transmit-
ted diseases (STDs) (p < 0.05).
MSM behavior was mostly found in cases (30.2%), with
only five (1.1%) controls responding “Yes”to this question
(Table 3). Among male donors who reported lifetime MSM
behaviors, only 17.1% donors reported having MSM behav-
iors once or twice; 42.7% reported having more than
10 MSM encounters; 40.2% of donors with MSM behaviors
had three to five male sex partners; and 32.9% had more
than five male sex partners. Furthermore, about half of
TABLE 4. Multivariable logistic regression analysis of
factors associated HIV case control status
Variable Covariate subgroups
Odds ratio
(95% CI)
Site Chengdu 1.57 (1.01-2.46)
Chongqing 1.16 (0.77-1.75)
Luoyang 0.30 (0.16-0.58)
Mianyang 1.65 (0.88-3.11)
Urumqi 0.54 (0.29-1.03)
Zhengzhou 0.86 (0.50-1.48)
Guangxi 1
Sex Male 4.90 (3.35-7.18)
Female 1
Education Middle chool or less 1.95 (1.16-3.29)
High school or vocational
school
1.49 (0.90-2.46)
Associate degree 1.57(0.98-2.53)
College and above 1
Occupation Farming/Fishing/Forestry 1.44 (0.63-3.33)
Migrant workers 1.48 (0.74-2.95)
Military/Government/Health
care employee
0.45 (0.23-0.88)
Commercial/Service 1.13 (0.65-1.98)
Self-employed 0.99 (0.55-1.81)
Other/Missing 1.65 (1.00-2.73)
Student 1
TABLE 5. Individual risk factors associated with HIV case control status after adjusting for site, sex, education, and
occupation in multivariate logistic regression analysis (reference group = No response)
Questions about each risk factor Responses
n (%) of
Responses
Odds ratio
(95% CI)
Q1. In the past 6 months, did you have any injection (including intravenous [IV]
and intramuscular [IM] injections)?
Yes 183 (17.46) 2.20 (1.52-3.20)
Q2. Have you ever received a blood transfusion? Yes 34 (3.25) 1.08 (0.49-2.37)
Q3. Have you ever had any dental cleaning? Yes 207 (19.58) 1.70 (1.18-2.45)
Q4. Have you ever been previously diagnosed with syphilis, gonorrhea, or any
other sexually transmitted disease (STD)?
Yes 20 (1.91) 11.35 (3.08-41.8)
Q5. Have you ever had household contact with someone with HIV/AIDS? Yes 42 (3.94) 7.09 (3.35-14.99)
Unknown/Missing 232 (21.74) 3.09 (2.17-4.41)
Q6. Have you ever lived with a person who was an IV drug user? Yes 40 (3.75) 1.42 (0.68-2.99)
Unknown/Missing 111 (10.40) 2.15 (1.38-3.36)
Q7. Have you had more than two concurrent sex partners of the opposite sex? Yes 135 (12.69) 6.01 (3.87-9.34)
Q8. How often do you or your sex partner use a condom when you have sex
with your hetero-sex partner?
Never use condom 38 (3.57) 10.58 (4.65-24.04)
Sometimes or
always use
condom
97 (9.12) 4.91 (3.00-8.03)
Q9. (MALE ONLY) In your lifetime, have you ever had sex with another male? Yes 87 (12.32) 82.10 (29.9-225.4)
Q10. Have you ever paid or received money or other forms of remuneration for
having sex?
Yes 100 (9.45) 2.23 (1.40-3.55)
Q11. In your lifetime, have you ever had a sex partner who had a positive test
for syphilis, gonorrhea, or any other STD?
Yes 14 (1.31) 6.14 (1.91-19.74)
Unknown/Missing 148 (13.88) 9.70 (6.26-15.02)
TRANSFUSION 5
RISK OF HIV INFECTION AMONG CHINESE BLOOD DONORS
these MSM donors reported “always”using a condom or
“most of the time”; and 10 MSM donors reported having an
HIV-positive sex partner.
Multivariable logistic regression analysis
Table 4 displays results of the logistic regression analysis
that examined the association between selected covariates
and HIV infection when risk factors were not included. After
adjusting for site differences and other covariates, male
donors were almost five times more likely than female
donors to be cases (odds ratio [OR], 4.90; 95% confidence
interval [CI], 3.35-7.18). When compared with donors who
had a college or higher level of education, donors with mid-
dle school or lower levels of education were more likely to
be cases (OR, 1.95; 95% CI, 1.16-3.29).
Based on results of the univariable analyses and our con-
ceptual framework for this study, we selected 11 major medi-
cal and behavioral risk factors and evaluated their association
with HIV infection after adjusting for the covariates in a series
of multivariable logistic regression analyses (Table 5). Results
of multivariable analyses suggested that 9 of the 11 selected
risk factors were significantly associated with HIV infection
among blood donors. Previous diagnosis with other STDs was
an important risk factor for HIV infection (OR, 11.35; 95% CI,
3.08-41.80). Having household contact with HIV/AIDS infected
individuals was significantly associated with HIV infection (OR,
7.09; 95% CI, 3.35-14.99). Compared to donors who did not
have multiple concurrent sex partners, those who had more
than two concurrent sex partners of the opposite sex were 6.01
times more likely to be HIV infected. Further, the OR for those
who “Never”used condoms was 10.58 (95% CI, 4.65-24.04).
Notably high risk for HIV infection was observed in
male donors having ever had MSM behavior (OR, 82.10;
95% CI, 29.90-225.4) than other male donors. In addition,
having remunerated sex was significantly associated with
HIV case status (OR, 2.23; 95% CI, 1.40-3.55) and having a
sex partner with an STD was associated with a 6.14 odds of
being HIV positive. Overall, 210 of 313 (67%) cases reported
having at least one of the eight selected risks that were sig-
nificantly associated with HIV infection in multivariable
analysis (excluding dental cleaning), whereas only 23% of
controls reported having at least one of these risks
(p < 0.001).
DISCUSSION
The present study evaluated medical and behavioral risk
factors associated with HIV infection among HIV-positive
and -negative donors through a case control survey study.
After adjusting for site and donor demographic characteris-
tics in multivariable logistic regression models, we identified
eight major risks that were associated with 67% of HIV
infected donors. All results underscored high-risk sexual
behaviors as important factors for current HIV infection
among Chinese blood donors. Each of the top six high-risk
sexual behaviors was associated with five times or higher
odds of being HIV infected even after adjusting for differ-
ences in study site, sex, education, and occupation. Being
MSM, having a previous diagnosis of STDs, never using a
condom during sex, having sex with HIV-infected individ-
uals, having sex partners with other STDs, and having more
than two concurrent sex partners were major contributors
to HIV infections among Chinese blood donors.
It appears that being MSM as well as having unsafe sex
practices might have exacerbated the spread of HIV infec-
tion among this subpopulation. The high number of sex
partners reported by donors who had MSM behaviors sug-
gests that, without effective intervention, increasing spread
of HIV infection may occur not only among MSM but also
among female sex partners because most MSM in China
appear to have active female sex partners.
15
These findings
are consistent with the reported increase of new HIV cases
among MSM from 2.5% in 2006 to 26% in 2014.
2,16
The
MSM donor deferral policies are developed and used in sev-
eral countries and areas, like Canada, where the blood
donation waiting period for MSM is 3 months.
17
In China,
such policy is still lacking and, if implemented, might be
helpful to elevate the threshold of blood safety, regarding
the notably high risk of HIV infection in potential MSM
donors observed in our study.
Only a small number of donors reported living with
someone who had HIV/AIDS or used illicit drugs, having
household contact with someone with HIV/AIDS, or having
household contact with someone without knowing their
HIV/AIDS status. Yet our data suggest that these were high-
risk factors for HIV infection. However, due to the ambiguity
of the phrases in these questions, “living with someone”
and “having household contact”could be synonymous with
having sexual contact with HIV-infected individuals that led
to HIV infection. In other words, such lifestyles or living
conditions were also suggestive of the risks of unsafe sex
practice for HIV infection. Furthermore, known as a typical
unsafe sex practice, inconsistent condom use is recognized
as one of the high-risk behaviors resulting in HIV infec-
tion.
14,18
It is no surprise that we found those who “Never”
used condoms were 10.58 times more likely to be HIV
infected.
Previous diagnosis of other STDs and lack of knowl-
edge of oneʼs own hepatitis status were also associated with
risks of HIV infection among a small group of donors. While
the latter might indicate lack of routine health care, previ-
ous diagnosis of other STDs is reported by multiple studies
as a good indicator of high-risk sexual behaviors.
19–21
These
results, combined with our findings about significant risks
associated with MSM, multiple concurrent sex partners,
having remunerated sex, and inconsistent use of condoms,
highlight the importance of enhanced effort in educating
the public on HIV transmission risks and the importance of
safe sex practice in China. The health information
6 TRANSFUSION
ZENG ET AL.
questionnaire currently used by blood centers in donor
screening does not include any specific questions about safe
sex practice due to cultural sensitivity and donor privacy
considerations. Our findings emphasize the importance of
including specific sexual behavior questions in donor
screening to help identify and defer donors highly vulnera-
ble to HIV infection due to their high-risk sexual behaviors
and lifestyles.
Interestingly, a substantial number of donors who
skipped or responded “Unknown”to questions about their
sex partnersʼSTD or hepatitis infection status or illicit drug
use displayed significantly higher odds of being HIV infected
than those who gave negative responses. Apparently, know-
ing the health status of oneʼs sex partners is a reliable indica-
tor of long-term, stable, and likely healthy sex partnerships.
From a different perspective, such results also underscore the
importance of safe sex practice education as a protective fac-
tor in the battle against the HIV epidemic in China.
In our study, no significant association was detected
between a history of inpatient or outpatient surgery, dental
or cosmetic surgery, acupuncture, or endoscopy with HIV
infection. Although more HIV-positive donors reported hav-
ing blood transfusion that was possibly a remote history,
transfusion was not significantly associated with HIV infec-
tion after controlling for the covariates. Similarly, having IV
or IM injections in the past 6 months, although significantly
associated with HIV infection in multivariable analyses, did
not seem to be risks for HIV infection. The association
between dental cleaning and HIV infection, although signifi-
cant in both univariable and multivariable analysis, is not
supported by any other research literature. The significant
relationship might be driven by some undetected behavior
patterns in some unknown subgroups or geographic areas.
More detailed analysis is needed in future research.
As expected, a higher proportion of migrant workers
(12.46% vs. 6.69%) and lower proportion of government/
military/health care employees (4.79% vs. 14.04%) were
found in HIV-positive donors, whereas more cases than
controls (29.39% vs. 18.37%) reported “Other”as an occupa-
tion. The high risks associated with migrant workers for HIV
infection was not significant after adjusting for other factors,
perhaps confounded by the effect of education since
migrant workers were often associated with having less edu-
cation.
22
Compared to middle school or lower levels of edu-
cation, having a college or higher level of education was a
protective factor against HIV infection. Moreover, a recent
multisite cross-sectional study in seven Chinese cities
described greater risk of HIV acquisition for migrant
MSM,
23
which might increase the complexity aspect to pre-
cisely explore the influence of migration on HIV infection
risk, as well as its impact on blood safety. Meanwhile, cer-
tain occupations such as military/government/health care
employees seemed to be associated with lower risks for HIV
infection than other occupations. Government and health
care jobs usually require a college or higher level of
education, and individuals in military service might have
strict rules and regulations that prevent them from con-
ducting high-risk behaviors associated with HIV infection,
making these occupations a protective factor and highlight-
ing the importance of education, knowledge, and self-
discipline in curbing the spread of HIV infection.
We did not find significant differences by age, ethnicity,
or first-time and repeat donor status. Yet with 36.3% and
62.5% of all participants within the ages of 18 to 25 and
26 to 35 years, respectively, our study participants were a
coherently young cohort, especially given that 30.7% of the
cases were 18 to 25 years old—sad yet supportive evidence
for the increasing proportion of HIV-infected individuals
among the youth, especially MSM.
24
Among other limitations of this study, we used a conve-
nience sample. In an attempt to catch all possible cases that
could be lost to follow-up, we attempted to recruit all anti-
HIV screening reactive blood donors, knowing that about
two-thirds of the screening reactive donors were likely to be
false positives with WB HIV-negative results. Although the
study was designed to be a case–control study at a 1:2 to 1:3
case:control ratio, large variations in the WB confirmed-
positive rate among the anti-HIV screening reactive dona-
tions in different study sites resulted in variant case:control
ratios at study sites. To maximize the number of cases, we
included GX-CDC, which followed a slightly different study
protocol and were not able to control donorsʼawareness of
their possible infectious status notified either by blood cen-
ters or local Center for Disease Prevention and Control. Due
to the scope of this study, the sensitivity and confidentiality
of this study, and the social stigma and discrimination
against MSM and individuals with HIV/AIDS, we could not
include subtle and specific questions about sexual orienta-
tion and sex practices in the questionnaire. Nor can we link
the surveys to donation data for further retrospective analy-
sis to investigate prior and follow-up donations from the
confirmed HIV-positive donors.
In summary, our findings indicate that high-risk sexual
behaviors are critical risks for HIV infection among Chinese
blood donors. Although an updated blood donor screening
strategy for HIV (including both serologic and nucleic acid
test screening) has been put into effect in China, HIV
transfusion-transmitted risk, especially that from window-
period infections, still persists. Lack of HIV/AIDS knowledge
and safe sex practice education among Chinese youth in gen-
eral, particularly those from remote and undeveloped regions,
combined with social stigma and discrimination against peo-
ple living with HIV/AIDS and MSM continue to be critical
barriers to HIV treatment, prevention, and early detec-
tion
12,25,26
and challenge the safety of Chinaʼs blood supply.
Therefore, it is important to further enhance the effectiveness
of donation testing strategies and pretesting healthy history
screening by incorporating more specific questions on poten-
tial donorsʼsexual practice to defer high-risk donors to war-
rant the safety of the blood supply in the future.
TRANSFUSION 7
RISK OF HIV INFECTION AMONG CHINESE BLOOD DONORS
CONFLICT OF INTEREST
The authors have disclosed no conflicts of interest.
REFERENCES
1. Li L, Li KY, Yan K, et al. The history and challenges of blood
donor screening in China. Transfus Med Rev 2017;31:89-93.
2. National Health and Family Planning Commission of the
Peopleʼs Republic of China. 2015 China AIDS response pro-
gress report. [cited 2019 Mar 12]. Available from: http://www.
unaids.org/sites/default/files/country/documents/CHN_
narrative_report_2015.pdf.
3. Song M, Ren FR, Gong XY, et al. Comparative clinical evalua-
tion of mini-pool NAT and individual donor NAT by PROCLEIX
ULTRIO assay for donation screening. Chin J Blood Transfus
2010;23:173.
4. Hong Y, Huang X, Ling H, et al. Prevalence and trend of HIV
infection among voluntary blood donors in China since imple-
mentation of the Blood Donation Law: a systematic review and
meta-analysis. Trop Med Int Health 2012;17:978-88.
5. Huang ZJ, Wang W, Martin MC, et al. "Bridge population": sex
workers or their clients?–STI prevalence and risk behaviors of
clients of female sex workers in China. AIDS Care 2011;23
(Suppl 1):45-53.
6. Ma W, Wu G, Zheng H, et al. Prevalence and risk factors of
HIV and syphilis, and knowledge and risk behaviors related to
HIV/AIDS among men who have sex with men in Chongqing,
China. J Biomed Res 2016;30:101-11.
7. Choi KH, Steward WT, Miege P, et al. Unpacking the influence
of sexual stigma on HIV Risk: results from a prospective study
of men who have sex with men in Beijing, China. J Acquir
Immune Defic Syndr 2017;74:e38-44.
8. Zou X, Chow EP, Zhao P, et al. Rural-to-urban migrants are at
high risk of sexually transmitted and viral hepatitis infections
in China: a systematic review and meta-analysis. BMC Infect
Dis 2014;14:490.
9. Huang W, Dong Y, Chen L, et al. HIV risk and sexual health
among female migrants in China. Health Promot Int 2016;31:
595-605.
10. Hong Y, Stanton B, Li X, et al. Rural-to-urban migrants and the
HIV epidemic in China. AIDS Behav 2006;10:421-30.
11. Dong Z, Xu J, Zhang H, et al. HIV incidence and risk factors in
Chinese young men who have sex with men–a prospective
cohort study. PLoS One 2014;9:e97527.
12. Huachun Zou JDT, Fan S, Junjie X, et al. Learning about HIV
the hard way: HIV among Chinese MSM attending university.
Lancet Infect Dis 2018;18:16-8.
13. Griffiths J, Fang N, Wang S. Chinaʼs lack of sex education
is putting millions of young people at risk. CNN Health,
Dec 6, 2016. [Cited 2020 Jan 4]. Available from: https://
edition.cnn.com/2016/11/30/health/china-sex-education-
world-aids-day/.
14. Ssewanyana D, Mwangala PN, van Baar A, et al. Health risk
behaviour among adolescents living with HIV in sub-Saharan
Africa: a systematic review and meta-analysis. Biomed Res Int
2018;2018:7375831.
15. Chow EP, Wilson DP, Zhang L. Estimating HIV incidence
among female partners of bisexual men in China. Int J Infect
Dis 2012;16:e312-20.
16. Qin Q, Guo W, Tang W, et al. Spatial analysis of the human
immunodeficiency virus epidemic among men who have sex
with men in China, 2006-2015. Clin Infect Dis 2017;64:956-63.
17. Canadian Blood Services. Blood donation waiting period for
men who have sex with men reduced to three months. 2019
[cited 2019 Nov 11]. Available from: https://blood.ca/en/news-
and-events/newsroom/blood-donation-waiting-period-for-
men-who-have-sex-with-men-reduced-to-three-months.
18. Tang H, Wu Z, Mao Y, et al. Risk factor associated with nega-
tive spouse HIV seroconversion among sero-different couples:
a nested case-control retrospective survey study in 30 counties
in rural China. PLoS One 2016;11:e0164761.
19. Mukanyangezi MF, Manzi O, Tobin G, et al. Sexual risk behav-
iour in a cohort of HIV-negative and HIV-positive Rwandan
women. Epidemiol Infect 2019;147:e54.
20. Capasso A, DiClemente RJ, Wingood GM. Pregnancy coercion
as a risk factor for HIV and other sexually transmitted infec-
tions among young African American women. J Acquir
Immune Defic Syndr 2019;82(Suppl 2):S155-61.
21. Shrestha R, Karki P, Copenhaver M. Early sexual debut: a risk
factor for STIs/HIV acquisition among a nationally representa-
tive sample of adults in Nepal. J Community Health 2016;41:
70-7.
22. He N. Sociodemographic characteristics, sexual behavior, and
HIV risks of rural-to-urban migrants in China. Biosci Trends
2007;1:72-80.
23. Yu YQ, Xu JJ, Hu QH, et al. High-risk behaviour and HIV infec-
tion risk among non-local men who have sex with men with
less than a single yearʼs residence in urban centres: a multi-
centre cross-sectional study from China. Sex Transm Infect
2018;94:51-4.
24. Yang Z, Huang Z, Dong Z, et al. Prevalence of high-risky
behaviors in transmission of HIV among high school and col-
lege student MSM in China: a meta-analysis. BMC Public
Health 2015;15:1272.
25. Qiu J. Stigma of HIV imperils hard-won strides in saving lives.
Science 2011;332:1253-4.
26. Burki TK. Discrimination against people with HIV persists in
China. Lancet 2011;377:286-7.
SUPPORTING INFORMATION
Additional Supporting Information may be found in the
online version of this article.
Appendix S1. Supporting Information.
8 TRANSFUSION
ZENG ET AL.
