A Study of Sexual Behavior Among Rural Residents of China
HONGJIE LIU1, ROGER DETELS1, JIE XIE2, WENZHOU YU3,
WEISHENG SONG2, ZHENYA GAO2, ZHIXIN MA3
1 Department of Epidemiology, UCLA School of Public Health, 2 Yingshang County Anti-Epidemic Station, 3 Anhui Provincial Anti-Epidemic Station
- Abstract
- Intruduction
- Method
- Result
- Discussion
- References
- Picture
Objective: Although the spread of HIV/AIDS and other STDs in China have been associated with sexual activities, there is little information about sexual behavior and variations that determine the spread of HIV in rural areas where the vast majority of the Chinese population lives and where recent reports suggest that HIV prevalence is highest. Hence, studies of human sexual behavior aimed at identifying high risk behaviors are needed to formulate effective prevention programs.
Method: A cross-sectional study design was used to measure sexual activities retrospectively and currently. The survey sample was drawn by a two-stage cluster method. A two-part anonymous questionnaire was used in this survey. The sensitive questions related to sexual behavior were administered using a tape recorder, ear phones and an answer sheet which included the number of the question, but not the text of the question.
Results: A total of 1,057 subjects were interviewed. Among 886 sexually active individuals, 7.8% reported having had more than one sexual partner, 22.8% reported having had premarital sex, 2.4% having had anal intercourse, 4.1% having had oral intercourse and 2.3% having had both anal and oral intercourse. Less than 2 percent reported past or current sexually transmitted diseases. 10.4% used condoms, only 11.2% for every sexual act. History of pre/extra marital sex was higher in recent cohort. Logistic regression analysis revealed the following associations: young age at first sexual intercourse (OR=2.40) and older age at marriage (OR=1.68), desire to have multiple partners (OR = 2.40), engaging in more than 2 coital acts per week (OR = 1.96), men (OR = 1.92), seen pornographic material (OR=1.77) and higher income (OR=1.75).
Conclusions: Sexual norms and behaviors in rural China are changing rapidly. The prevalence of high risk sexual behavior among young rural residents is increasing. Strategies to prevent HIV/AIDS should include extensive education to promote delayed onset of sexual activity, safer sexual behavior and condom use, especially among those who temporarily migrate to cities.
WHO estimates indicate that globally there were 27.9 million persons infected with HIV by mid-1996. Of these, 5 million live in Asia, where sexual contact among heterosexuals and drug use are the major modes of transmission [1]. In the absence of available effective treatments and vaccines, the best strategy to prevent infection is convincing people to modify their behavior and to use condoms. Hence, studies of human sexual behavior to identify the prevalence and types of risk behaviors in a community are needed to implement effective behavioral intervention programs.
The first AIDS case in China was reported in June 1985 [2]. Since then, HIV infected individuals have been documented by a growing network of testing sites and sero-epidemiologic studies. There is now sufficient serologic evidence available from China to document that the nation has a firmly established and growing HIV epidemic [3]. By March1998, 9,970 HIV-positive people had been reported from 29 of China's 30 provinces, municipalities and autonomous regions, including 290 AIDS cases [4]. The majority of these infected individuals came from the rural areas of China [5].
Although 52-75 % of AIDS cases and HIV infections in China are believed to have been acquired through injecting drug use, transmission via injecting drug use is concentrated in the southern border provinces, and heterosexual transmission is now increasing there as well [5]. Sexual transmission will probably become the dominant mode of transmission in China as it has in Thailand and Myanmar.
Sexually transmitted diseases (STDs) had been eliminated for several decades since the 1960's in China. After an absence of 20 years, STDs have been reappeared and been recognized as a major public-health problem. The prevalence has continued to rise over the past few years in urban areas as well as rural areas. The number of STDs cases reported in China increased from 13 in 1977 to 5,838 in 1985, 153,305 in 1990, and 300,466 in 1994. The reported incidence is highest in the age group of 20-39 years [6].
Because the spread of HIV/AIDS and other STDs in China have been strongly associated with sexual activities, several studies on HIV/AIDS-relative knowledge, attitudes and HIV risk behaviors have been conducted among medical students [7], health professionals [8], taxi drivers and hotel attendants [9], and drug users [10, 11]. There is little information, however, about sexual behavior among residents of rural areas where the vast majority of Chinese population live and where recent reports indicate that HIV is occurring. The potential of the HIV epidemic to spread beyond the current risk groups depends upon the prevalence of high risk behaviors in the general population. Therefore, we have conducted a study of sexual behaviors in a rural population of Anhui Province in the People's Republic of China in June 1997.
Setting
Anhui province is located in eastern China, where there is a high population density (436 people per Km2). More than 82.1% of the population are farmers, most of whom are of Han ethnicity. The major industry is agriculture, including rice, wheat, and tobacco production [12]. The first AIDS case in Anhui was reported in December 1994. Two AIDS cases and 6 HIV infected persons had been reported in the province by 1995 [13, 14]. Yingshang is a typical rural county in Anhui province. The total population is 1,146,900, of whom, more than 90 % are farmers [15]. For economic reasons, farmers often temporarily migrate to urban areas during the slack farming season. This survey was undertaken in this rural county.
Sample selection
A cross-sectional study design was used to measure past and current sexual activities. Residents of this county who were 15-49 years were eligible to participate in the study. A two-stage cluster sample was conducted in June 1997. Using a Csurvey program [16], 30 clusters (villages) were selected as study frames with probability proportionate to size (PPS) from 636 villages [17, 18]. At the second stage, interviewers randomly selected the first household and then proceeded to the next nearest household until 35 eligible subjects were sampled. All eligible residents who were home at the time of the visit were interviewed. Most of the subjects who were not at home had migrated to other cities for temporary jobs.
Questionnaire
An anonymous questionnaire was administered to each interviewee. The questionnaire consisted of two parts. The first part included less sensitive questions, such as sociodemographic characteristics, knowledge and attitude to AIDS. The sensitive questions relating to sexual behavior were included in the second part. The questionnaire was piloted before the field study to select and tailor the items to the study population. A copy of the final questionnaire is included in the appendix.
Tape recorder technique
A tape recorder with ear phones was used to ask the sensitive questions in part two. The questions were first recorded on tapes in the local dialect by two health workers. The participants listened to the questions and recorded their responses on a coded answer sheet listing the possible responses, but not the questions. In order to make the recording more interesting, the questions were asked by a male voice while the possible answers were given by a female voice. For example, the male voice asked: "Question number 1. Who was your first sex partner?". The female voice answered: " Your answer can be: 1, My spouse. 2, My friend. 3, A prostitute. 4, Other." The answer sheet showed 4 numbers, one of which the respondent was to mark. Thus, no one knew what their responses were. Their names and identifying information were not recorded on any of the forms including the answer sheet.
Interview
Trained interviewers were paired with respondents of the same sex. Individuals were interviewed in a separate room. Subjects were trained by the health workers to use the tape recorder. Once the respondents knew how to use the recorder and how to choose the appropriate answers, the interviewers left them alone to complete the questionnaire. No one except the respondent was allowed to be present. If respondents were illiterate, interviewers assisted them in completing the answer sheet, but did not state the questions. Before the study began, a five-day training course was conducted. Interviewers reviewed the proposal and questionnaire and were lectured on the purpose of the study, the method and procedure of interview. Emphasis was put on communication skills with respondents. Interviewers practiced conducting interviews using their counterparts as interviewees. Two pilot studies were conducted in two clusters, in which 65 subjects were interviewed.
Re-interviews for data validation
One cluster was randomly selected for re-interview to verify demographic (sex, age, education) and sexual behavior (such as age at first intercourse, age at marriage, number of sex partners and coital frequency per week) questions. The distributions of the results were compared with the responses in the main survey.
Statistical analysis
A x² test was used to compare proportions and a "t" test to compare means. Logistic regression was performed to evaluate the association between possible factors determining high risk sexual behavior and reported high risk sexual behavior. We defined premarital sex , multiple partnership, oral and anal sexual intercourse and sexually transmitted diseases as high risk sexual behaviors. We dichotomized and compared those who had one or more high risk sexual behaviors versus those who had no high risk sexual behaviors. Odds ratios (ORs) were estimated to measure the association using Epi-Info and SAS. There were 13 AIDS knowledge questions about routes of AIDS transmission, severity of AIDS and methods of prevention. A summary score was developed from these questions. One point was assigned for each correct answer and 0 points for each incorrect or unsure answer. Persons who had not heard of AIDS were given the score of 0. Thus, the possible score ranged from 0 to 13.
Study population
A total of 1,269 people were invited to participate. One hundred and eight-five (15%) people were not at home , most of whom (62.2%) had temporarily migrated to other cities for employment. Among the 1,084 who were at home, 27 (2.5%) refused to participation. A significantly higher proportion of men were not at home. There was no difference in marriage status and age distributions between those who temporarily migrated and the population (Table 1). Hence, 526 men and 531 women were finally interviewed. Of 1,057 respondents, 98.7% were Han which is the main ethnic group in China. The others (1.3%) belonged to the Hui minority. Twelve people (1.1%) reported that they had donated blood for pay. Table 2 presents and compares the sociodemographic characteristics and HIV/AIDS knowledge and perception by gender. Eighty three percent of subjects were married. Men were more educated and had more other jobs than women. The level of HIV/AIDS knowledge was low, especially among women. Both genders had the same pattern of perception of HIV infection.
Profile of sexual behavior
Age at first marriage and age at first sexual intercourse
Eight hundred and seventy eight respondents were married. In addition 8 never-married people reported sexual experience. Thus, there were 886 sexual active respondents, 421 men and 465 women.
The mean age at marriage was 23 years for men, and 22 for women and did not vary by calendar year of marriage. The earliest age at first sexual intercourse was 17 years. The mean age at the first sexual intercourse was 23 years for men, and 22 for women. Fig. 1 shows the declining age at first sexual intercourse by year of birth. From 1952 - 59 to 1975-82, age at first intercourse decreased from 25.7 to 19.1 years for men and from 23.2 to 19.8 years for women.
Premarital sexual partners
There were 202 (22.8 %) sexually active respondents who reported having had premarital sex, 27.1 % (114/421) for men, 18.9% (88/465) for women. Fig. 2 presents the proportion of married persons who reported having had premarital sex by year of birth. For both men and women, the proportion increased steadily from 1952-61 to 1972-82 birth year.
Coital frequency
The median coital frequency was 2 acts per week on average in the previous four weeks. The frequency varied according to age. Old people reported less frequent sexual intercourse while the young had more (Fig. 3). In addition, 4.4 % (39/886) of sexually active people claimed that their average sexual intercourse per week was zero.
First sexual partner
Most of the respondents, 94.9% (41/886), reported that their first sexual partner was their spouse while 4.2 % (37/886) reported first intercourse with a friend or other individual. More men (5.9%) than women (2.6%) reported that their first sexual partner was a friend or other individual.
Oral and anal sexual intercourse
Among sexually active respondents, 2.4% (21/886) reported having had anal intercourse, 4.1% ( 36/886) reported oral-genital intercourse and 2.3% ( 20/886) reported both. A higher proportion of younger respondents reported oral-genital and/or anal sexual intercourse(Fig. 4). No respondents reported having sex with their same gender.
Sexually transmitted diseases
Fifteen (1.7%) respondents reported having had sexually transmitted diseases. The proportions of sexual active men and women reporting sexually transmitted diseases (STDs) were 2.6%, and 0.87% respectively. The reported sexually transmitted diseases were Syphilis (13.3%), Gonorrhea (26.7%), Condyloma acuminatum (6.6%), Genital ulcer (13.3%), Viginal discharge (33.3%) and others (6.8%).
Multiple partnership
Among sexually active individuals, 7.8 % (68/886) reported more than one sexual partner in their lifetime, 7.2% (63/878) among ever married and 62.5% (5/8) among never married respondents with sexual experience. The proportions of those declaring 2, 3, and more than 4 sexual partners among sexually active individuals were 5.9%, 1.0%, and less than 1% respectively. Younger men were more likely to have had multiple partners than older men, as were older women than younger women (Table 3). Also, 22.2% (106/421) of men, and 7.4% (34/465) of women reported they would have more then one sexual partners if permitted.
Contraceptive practices.
Among sexual active respondents, 68% (602/886) reported using contraceptives, 46.6% (196/421) among men and 87.3% (406/465) among women. The proportions of those reporting condom use was 10.4%. Among persons who used condoms, only 11.9% used them for every sexual act, 30.5% used them only during ovulation, and 57.6% used them occasionally.
Univariate Analysis of the factors related to high risk sexual behavior
Table 4 summarizes factors associated with high risk sexual behavior. Factors significantly associated with high risk sexual behavior included men, younger age, more education, higher annual per capita income, younger age at first sex, currently living with spouse, greater coital frequency, intention to have multiple partners, and saw pornographic items. Factors not associated included occupation, drinking alcohol, contraceptive practices, HIV/AIDS knowledge, perception of AIDS risk, and age at marriage.
Logistic regression analysis identifying factors related to high risk sexual behavior
To identify factors associated with high risk sexual behavior, we compared the group engaged in high risk sexual behavior with the group not reporting this behavior using logistic regression analysis (Table 5). All factors in the univariate analyses were also used in the multivariate analysis. Controlling for other variables, high risk sexual behavior was associated with younger age at first sexual intercourse (OR=2.40) and older age at marriage (OR=1.68), people who desired to have multiple partners (OR = 2.40), people who had more than 2 coital acts per week (OR = 1.94), men (OR = 1.92), people who saw pornographic items (OR=1.77) and persons with higher income (OR=1.75).
Results of re-interview of sample
One of the clusters was selected randomly for interview. Since the original survey was anonymous, we were not able to re-interview the same people as in the original sample although there was probably some overlap between the two samples. The distribution of responses for several key variables is compared in table 6. There were no significant differences in the distribution of responses on these variables. In fact, the distributions were quite comparable. Also, there are no significant differences among other sexual activities, such as pre/extra marital sex, anal or oral sexual intercourse.
This study of a random sample of residents 15-49 years of a rural area of China documented the profile of sexual behavior and the proportion of residents engaging in high risk sexual behavior that may carry an elevated risk of HIV infection in an area where HIV has recently been reported. Information about sexual behavior is of obvious value in defining the potential exposure of a population to risk of sexually transmitted diseases, such as HIV/STDs. This study provides evidence that age at first intercourse occurs at a relatively young age among both men and women. A trend toward earlier age at first intercourse has been recorded in studies of sexual behavior in other counties as well as in this study[19, 20] . The trend may reflect improved nutrition, an increased economic level, declining parental and governmental influence, and/or a earlier sexual maturity. Age at marriage has remained quite stable, but the age at first intercourse has declined. The interval between first sexual intercourse and marriage for the younger respondents is, therefore, longer, providing greater opportunities for pre-marital sex. High risk sexual behavior was, in fact, most strongly associated with earlier age at sex and later age at marriage in the logistic regression analyses.
Popular belief holds that premarital intercourse among young people is a fairly recent development in China. Many hospitals and private clinics now report that unmarried women account for most of their abortion patients [22]. Jinlin reported that 14.3% of university students in his study had premarital sex and 48% of male and 25% of female students approve of sexual activities before marriage [23]. Premarital sex as a risk behavior is closely related to multiple partnerships. In our study, 60.3 % (41/68) of people who engaged in sex with multiple partners reported having premarital sex. Young respondents were more likely to report premarital sex than older respondents.
Coital frequency per week was also positively associated with high risk behavior; people who reported multiple partners tended to also have a high frequency of sexual intercourse. These individuals put themselves and their partners at higher risk of HIV/STDs infection. The proportion of people reporting anal sexual intercourse was not very high among farmers. This kind of sexual intercourse has apparently been not accepted by society in China. Anal intercourse was reported more frequently among the youth in this study, suggesting an increasing prevalence of this high risk behavior in China.
Having multiple sexual partners is associated with an increased risk of HIV infection. However, 95.8% of women reported only a single lifetime sexual partner in this study. This does not mean that women in rural areas are not at high risk group. Risk of HIV infection in rural area may be influenced more by their male partner's behavior than their own. Other studies have reported a similar tendency to report more partners among men than women [24]. A possible explanation for this difference is that more men than women reported temporary migration to urban areas. We also interviewed 210 subjects from 5 randomly selected clusters and investigated the migration to cities among farmers during slack seasons. Among those, 25.7% (34.2% for men and 15.5% for women) had migrated to big cities, such as Shanghai, Beijing or Guangdong for temporary jobs. Most wives (79.7%) did not migrate out to cities with their husbands. However, 71.7% of women who migrated to cities did so with their husbands. Thus, men have more opportunities to have extramarital sexual partners in the cities. The second possible reason is a different attitude to having multiple partners among men than women. More men than women in this study reported a desire to have extramarital sexual partners. Logistic regression analysis also indicated that high risk sexual behavior is associated with a desire to have extramarital sexual partners.
Interestingly, our study shows that the proportion of women who ever have had multiple partners is much lower among those who were 15-25 years old than among those 36-45 years old. This may, however, reflect the greater opportunities to accumulate multiple partners with increasing age. Thus, older people would be expected to have more partners unless the patterns of acquiring sexual partners is changing. The reverse pattern observed among men in this study suggests that the rate of acquire multiple partners, although probably stable among women, is increasing among younger men and is a cause for concern. Only 13.2% (9/68) of people reporting multiple partners ever used condoms, and only 1.5% used them for every act of intercourse. Hence, men with multiple partners and their spouses are at high risk of HIV/STD infection. It is now generally accepted that the majority of women living with HIV infection in the world have acquired HIV heterosexually from their partners [25]. For this reason, when we talk about women' risk of HIV infection, we must take men's sexual behavior into consideration.
China has had a strong Family Planning Policy for many years. As a result, contraceptive practice is quite common. Some of the contraceptive techniques may reduce the risk of HIV/STDs infection while others may increase it or have no effect. It is reported that the use of oral contraceptives and intrauterine devices (IUD) may increase the risk of HIV/STDs [26], whereas barrier contraceptives, such as condoms and spermicides can reduce the risk of HIV infection [27, 28]. In this rural area, the use of condoms and spermicide is very low. Most sexually active persons have had a sterilization operation or use an IUD. Among the farmers who used condoms, most of them used them only occasionally or only during ovulation. Thus, current contraceptive practices in this rural area will not slow the transmission of HIV and STDs.
The number of STD patients has been increasing rapidly among workers, farmers, private business persons and the unemployed in China [29]. The most common diseases are syphilis, gonorrhea and condyloma acuminatum. In our study, the prevalence of STDs reported among rural residents was 1.7%. This result is lower than other studies among high risk groups. Shuan did an STD survey among rural residents returning home from cities where they worked in construction sites and found a prevalence of 4.3% [30]. Han reported that the prevalence of STDs among truck drivers, however, is 6.8% [31]. Our study relied on reported STDs only and, therefore, would not identify asymptomatic STDs, a particular problem among women.
Several limitations in our study need to be mentioned. First, all variables were self-reported. Respondents may have underreported some sexual behaviors and diseases, due to conservative social standards. Second, since some information required respondents to recall events which happened several years ago. Thus, some recall bias might have occurred. If these limitations existed, our results might be biased in both directions. ORs may have been either decreased or increased. In addition, we calculated the confidence interval by treating the data as simple random sample (SRS). Actually, this was a cluster survey. As a results, the confidence intervals we calculated here might be smaller than they should be.
We did a re-interview for data reliability in a sample of villages. There were no significant differences between the initial and second responses. We also analyzed the consistency between reporting of sexual frequency and types of sexual activities of married couples. The results indicated a high degree of reliability of answers regarding premarital sex, oral or anal sex and coital frequency [32]. The reliability analysis suggests that our data, at least on these parameters were valid.
These results provide information which can be used to make preventive programs on HIV/AIDS control in rural areas more effective and also underscores the need for HIV/STD programs in rural areas. First, sexual norms and behaviors in rural China are changing rapidly. Second, it is important to promote delayed onset of sexual activity and to promote safe sexual behavior, such as the use of condoms. Third, extensive education on HIV/AIDS prevention should be implemented among farmers, especially those who will migrate to cities for temporary jobs. Even though AIDS-related information is not effective alone, it is needed to increase personal skill in the protection against HIV infection. Fourth, migrants are a high risk group and HIV surveillance should be carried out among them to monitor the epidemic trend of HIV. Lastly, preventive programs should strongly emphasize condom use, especially with sex workers and casual partners.
Acknowledgments:
We wish to thank Prof. Ralph Frerich, Prof. Virginia Li and Dr. Zunyou Wu for their assistance on this study proposal and for reviewing the manuscript.
1. WHO: WER 1996. 71: 205-212.
2. Dai Zhi-cheng: AIDS prevention and control in China. Chinese Medical
Journal 1991, 104 : 795-798.
3. Vincent E. Gil: Sinic Conundrum: A history of HIV/AIDS in the People's Republic
of China. Journal of sex research 1994, 31: 211-217.
4. China Ministry of Health, Joint United Nations Programme on HIV/STD: China
Responds to AIDS. World Bank November 1997.
5. Sun XH, Nan JH, et al: AIDS and HIV infection in China. AIDS 1994, 8 (supp 2):
555-559.
6. Zheng zhongbe: Analysis on epidemic of STDS in China. Chinese Journal of
Prevention and Control of STD and AIDS 1996, 2: 49-53.
7. Li, V.C., Cole, B.L., Zheng, S. Z., et al: HIV-related knowledge and attitudes
among medical students in China. AIDS Care 1993, 5: 305-312.
8. Li, V.C., Clayton S, et al: AIDS and sexual practices of health professionals in the
People's Republic of China. AIDS Education and Prevention 1992, 4: 1-5.
9. Zhang, X., Luo, B., et al: A KABP study about AIDS among taxi drivers and hotel
attendants in Beijing. Chinese Journal of Epidemiology 1994 15: 323-327.
10. Wu ZY, Zhang J, Detels R, et al: Characteristics of Risk-Taking Behaviors,
HIV/AIDS knowledge and Risk Perception among Young Males in Southwest
China. (submitted).
11. Zhang, X., Tian, C., et al: HIV risk behavior for HIV infection among drug users
in Yunnan province in China. AIDS 1994, 8:1701-1706.
12. Anhui Statistics Bureau: Anhui statistical yearbook, China Statistical Publishing
House, Beijing, 1997.
13. Su Bing: A survey report of first case of AIDS patient an HIV infection in Anhui
province. Disease Surveillance 1997, 3: 91-92.
14. Zunyou Wu, Zhiyuan Liu, Roger Detels: HIV-1 infection in commercial plasma
donors in China. The Lancet 1995, 346: 61-62.
15. Yingshang county statistic department: Yingshang statistical yearbook, Yingshang,
1997.
16. Iwan Ariawan, Frerichs RR: User's Manual for Csurvey, Department of
Biostatistics and Population studies, Faculty of Public Health, University of
Indonesia, Indonesia, Fogart International HIV/AIDS Training Program,
University of California, Los Angeles
17. Frerichs RR: Simple analytic procedures for rapid microcomputer-assisted cluster
surveys in developing countries. Public Health Reports 1989, 104: 24-35.
18. Freichs RR: Lecture notes for Epi 418.
19. Wellings K., Field, J., Wadsworth, J. et al: Sexual lifestyles under scrutiny. Nature
1990, 348: 276-278.
20. ACSF investigators: AIDS and sexual behavior in France. Nature 1992, 360: 407-
409.
21. Michel Carael: Sexual behavior. In AIDS in the developing world. Edited by
Cleland J, Ferry B. London: Taylor and Francis; 1995: 79-96.
22. Li, V.C., Clayton S, et al: AIDS and sexual practices of health professionals in the
People's Republic of China. AIDS Education and Prevention 1992, 4: 1-5.
23. He Jinglin, Wang Hongyuan, Cui Gang et al: Study on sexual issues among
university students. Chinese Journal of Prevention and Control of STD & AIDS
1997, 3: 103-105.
24. Laumann EO., Gagnon JH., Michael RT., Micheals S.: The organization of
sexuality. The university of Chicago Press; 1994:172-220.
25. Catherine Hankins: Sexual transmission of HIV to women in industrialized
countries. Wld hlth statist. Quart. 1996, 49:106-113.
26. New IUD prompts new look at age-old device. Contraceptive technology update,
1989, 10: 71.
27. World Health Organization: Contraceptive method mix, Guidelines for policy and
service delivery. WHO, 1996: 30-50.
28. Feidlum PJ., Weir SS.: The protective of nonoxynol-9 against HIV infection.
American Journal of Public Health 1994, 84: 1032-34.
29. Xu WY: The epidemic of STDs in China: Situation, Features and Problems.
Nanjing: National Center for STD Control and Research, 1992.
30. Sun Baowang: STD investigation among 5,892 return-home building workers.
Chinese Journal of Prevention and Control of STD & AIDS 1997, 3: 99-98.
31. Han Fuzhen, Qiao Hua, et al: STDs surveillance report of 3,628 drivers. Chinese
Journal of Prevention and Control of STD & AIDS 1997, 3: 101.
32. Liu HJ, Detels R: An Approach to Improve Validity of Responses in a Sexual
Behavior Study in a Rural Area of China, Submitted.
Picture 1: Interviewers reviewed the proposal and questionnaire and given lecture on the purpose of the study
Picture 2: Interviewers practiced interviewing using their counterparts as interviewees
Picture 3: Questions were recorded by two trained health workers in the local dialect
Picture 4: Interview team left for selected villages in the early morning
Picture 5: Randomly selected the first household by using the last two digital numbers on Chinese Currency
Picture 6: In this village, the number 28 was randomly selected. We started our survey from this household, then proceeded to the next nearest household
Picture 6,7: Interviewers were paired with respondents of the same sex. Participants were interviewed for the first part of the quewtionnaire which include less senditive questions, such as sociodemographic characteristics, knowledge and attitude to AIDS. Interview was delivered in a separeate room. No others were permitted to show up.
Picture 8,9: For confidential concern, tape recorders and earphones were used to administer the sensitive questions in the second part related to sexual behavior. Once the respondent know how to use recorder and how to select the appropriate answer, interviewers left them alone
Picture 10: In a rapid survey , a car is needed. But, sometimes you have to drive your car like this
Picture 11: This was my Interview team. Thank all of you