First sexual experience and current sexual behavior among older Thai men and women

Abstract

The focus of most of studies of sexual behavior has been on younger adults and adolescents because they are perceived as having higher levels of sexual activity than older persons and a consequently higher risk of HIV infection. Much less attention has been paid to the sexual behavior of men and women in their forties and fifties. Using a life course framework, the objective of this study is to examine AIDS knowledge, HIV testing and sexual risk behaviors among Thai men and women age 40–59 in the 2006 Thai National Survey of Sexual Behavior. The study also examines the influence of initial sexual experiences on later sexual behavior. The results indicate that older Thai men and women were well informed about methods of HIV transmission, but many were unaware of antiretroviral medications (ARV). Older adults were also less accepting of HIV positive persons than were younger adults. Fewer than half of the older adults had undergone HIV testing, with testing associated with medical checkups or undertaken before operations. Reported condom use was very low with regular partners, moderate with casual partners, and high with sex workers. While the age at first sex has not changed markedly, the type of partner has changed over time. Fewer men have their first sex with a commercial partner and . In general, first sex with someone with strong ties to the respondent was related to lower levels of risk behavior in late adulthood. Areas of concern for AIDS prevention programs include condom use with casual partners and paid partners, knowledge of ARV, and attitudes toward persons living with AIDS.

Introduction

Since the beginning of the AIDS epidemic in Asia, much research has been conducted on sexual behavior. The focus of most of these studies has been on younger adults and adolescents because they are perceived as having higher levels of sexual activity than older persons and a higher risk of HIV infection. Much less attention has been paid to the sexual behavior of men and women in their forties and fifties.

Data from a number of studies in Thailand and other countries indicate that many persons remain sexually active well into middle and old age (Lindau, Schumm, Laumann, Levinson, Muircheartaigh, & Waite, 2007; Nicolosi, Laumann, Glazer et al., 2004; Knodel & Chayovan, 2001). However, changes in physiology can affect the sexual response of men and women as people age (Bachman & Leiblum, 2004: Araujo, Mohr & McKinley, 2004). In addition, many people, particularly women, loose their sexual partner as they age (Lindau et al., 2007). A nationally representative sample of persons age 50 and over conducted in Thailand in 1995 found that among married persons, at age 50 more than 90 percent of men and women had sexual activity within the last year and at age 60 this figure was closer to 70 percent (Knodel & Chayovan, 2001). National level data on non-marital partners of men and women in this age group have not been available.

Thailand has been regarded as a showcase in the struggle against AIDS (United Nations Development Programme, 2004). The strong “100% Condom Use” program resulted in decreased transmission of HIV through commercial sexual contacts (Hanenberg et al., 1994). The annual number of new HIV infections peaked at 143,000 in 1991 and declined to less than 20,000 in 2004. While the earlier program effort focused on the commercial sex population, the epidemic has matured and the spread of HIV is much more varied than a decade ago. Levels of HIV infection are now unacceptably high among injecting drug users, men who have sex with men, mobile populations, and seafarers. The level of HIV transmission among young people is also of particular concern.

With the changes in sexual behavior that occurred in response to the AIDS epidemic and subsequent prevention programs, research has noted some changes in attitudes, norms, and behavior among young Thai men. Traditional patterns of brothel visitation may be replaced, in some part, by sexual relationships with peers (Chamratrithitrong et al., 2007; Van Landingham & Trujillo, 2002). Noncommercial relationships may now play a more important part in HIV risk behavior.

The life course perspective suggests that long term effects of sexual experiences during adolescence and young adulthood may affect later patterns of sexual behaviors (Elder, 1985). Consistent with social learning theory, sexual experiences early in life may lead to reinforcement of specific behaviors and interaction preferences (Browning, 2002). Due to historical circumstances, men and women in their forties and fifties in Thailand came of age in an era when patronage of commercial sexual partners was normative for Thai men and condoms were not used (Hanenberg & Rojanapithayakorn, 1998; VanLandingham & Knodel, 2007; VanLandingham & Trujillo, 2002). In addition, most women who were not involved in commercial sex typically began their sexual experience within marriage.

It is possible that these early experiences with non-marital partners influence expectations about sexual behavior and actual behavior later in life. Due to experience with these interactions and behaviors, persons with experience with commercial and other non marital partners in their youth may be more likely to have non marital partners in later life. Although older adults were in Thailand during the 100% condom use campaign, prevention efforts were not been directed toward them. We hypothesize that persons with relatively early initiation of sexual experience and sexual initiation with paid partners or less known partners will have higher levels of risk behavior later in life. Using data from a large national survey, the objectives of this paper are to describe the recent and early sexual experiences of Thai men and women in their forties and fifties compared to younger Thais and to examine the association of early sexual experiences with sexual activity in later life.

Materials and methods

Sample

Data for the study were drawn from the 2006 National Survey of Sexual Behavior in Thailand. The study was conducted by the Institute for Population and Spcial Research at Mahidol University in Thailand and funded by the Joint United Nations Program on AIDS (UNAIDS). The sample was designed to allow comparison between of differences in behavior between three residence categories: Bangkok, other urban, and rursl. When properly weighted, this sample provides nationally representative estimates of risk behaviors. Data was collected in Bangkok and in urban and rural areas of 14 provinces, selected from the 75 provinces of Thailand (excluding Bangkok) with probability of selection proportional to the size of the province. Further details of the sampling strategy can be found in Chamratrithirong et al, (2007). In total, 6,048 individuals were interviewed through a household probability sample, including 4,041 adults age 18–39, 923 adults age 40–49 and 644 adults age 50–59. Overall, response rates for the survey exceeded 80%, with most losses due to not being able to find the person at home during the field period.

Interviews were conducted face-to-face using paper questionnaires. Respondents and interviewers were matched by gender. Interviews were conducted in the respondent’s home or a nearby area where privacy could be assured. The instrument was developed and pre-tested by IPSR and included questions on knowledge of HIV/AIDS and antiretroviral therapy, attitudes and values about sexual behavior, risk of different partner types, and HIV/AIDS, prevention program exposure, HIV testing, STI treatment seeking behavior, and medical injections, martial status, sexual orientation, and sexual behaviors, sexual history (last 5 partners), adolescent sexual experience, and substance use.

Measures

Education was coded as (1) primary or lower, (2) lower secondary, (3) upper secondary, and (4) university.

Marital Status was coded as (1) unmarried and never had sex, (2) unmarried and had sex, (3) registered marriage, (4) unregistered marriage after ceremony, (5) widowed, divorced, or separated.

Living arrangements were coded as (1) with registered spouse, (2) with unregistered spouse after ceremony, (3) with cohabiter, (4) live apart from spouse, and (5) not living with a partner.

AIDS knowledge. An AIDS knowledge scale was created by summing the correct responses to 12 questions about AIDS transmission and prevention. This knowledge scale was developed for the 2006 National Survey. The questions included items on the appearance of HIV positive persons and methods of transmission. The reliability of the scale was evaluated with coefficient alpha (score = 0.98). Scores ranged from zero to 12.

Knowledge of Antiretroviral (ARV). Respondents were asked if they know about drugs that fight AIDS or drugs that would prevent someone with HIV infection from developing AIDS. The answer was coded 1=knew of ARV, 0=did not know.

AIDS stigma. An AIDS stigma scale was created by summing seven items on willingness to care for a relative with AIDS, to be friends with a person with AIDS, and to work with someone with AIDS. The scale was develop for the 2006 National Suvey. A high value indicates a high level of acceptance of persons living with AIDS. The reliability of the scale was evaluated with coefficient alpha (score=0.98). Scores ranged from zero to six.

HIV Test. Respondents were asked if they ever had an HIV test. Responses were yes (1) or no (0). Respondents were also asked the reason for the HIV test. These responses were coded 1=had a test and 0=did not have a test.

Sexual Behavior. Data were collected on overall experience with sexual behavior as well as experience with individual partners. About 98% of the sexual relationships that were reported were for opposite sex partners. For this reason, we have limited the analysis of sexual behavior to heterosexual relationships. Furthermore, due to the very strong differences in patterns of sexual behavior between men and women, data are presented for men and women separately.

Age at first sex and type of partner. Respondents were asked about the age that they first had sex and their relationship with their partner at the time of first sex. Relationships were coded into a) spouse or fiancé, b) boyfriend/girlfriend, c)acquaintance or friend, d) sex worker, or e) stranger/other partner. They were also asked if they used a condom with this first partner.

Types of sexual partners. Types of partners were identified by the respondent as married/cohabiting, regular partners they were not living with, casual partners, and paid partners or sex workers.

Condom use with different types of partners. Condom use frequency with each type of partner was coded as (1) always used condoms and (0) other. This coding was used because most respondents reported always use or never use.

Frequency of sex in the last three months with a partner they were living with was coded as (0) not at all, (1) less than once a month, (2) two to three times a month, and (4) once a week or more. This coding was chosen due to the distribution of responses.

Statistical methods

Chi-square and t tests were used to test the significance of differences in Tables 1–4. Logistic regression methods were used to estimate the models in Table 5. Sampling weights were used in the descriptive tables 1–4. In the logistic regressions of Table 5, sampling weights were not used but age and place of residence (Bangkok, other urban, rural) were controlled for.

Table 1.

Education, marital status, and living arrangements of adults by age group.

	Male			Female
	Adults 18–39	Adults 40–49	Adults 50–59	Adults 18–39	Adults 40–49	Adults 50–59
Education
Primary/lower	33.8	59.4	75.1	55.6	76.2	92.5
Lower secondary	26.2	11.0	11.0	16.5	39.3	32.7
Upper secondary	25.8	16.5	6.9	17.0	8.1	1.8
University	14.2	13.1	7.1	10.9	6.4	2.9
N	2,241	492	289	2,272	480	272
Working	84.7	96.5	89.2	68.0	58.7	73.2
Marital Status
Unmarried, never had sex	10.9	1.3	0.3	9.8	3.9	3.8
Unmarried, had sex	34.4	4.8	2.9	12.3	0.5	0.3
Married (registered)	25.1	66.0	77.2	47.1	66.5	71.2
Married (not registered)	267	20.6	11.7	22.7	14.9	5.7
Widowed/divorced/separated	2.9	7.3	7.9	8.0	13.8	19.0
N	2,243	492	289	2,272	480	272
Living arrangements (married or had sex)
With registered spouse	27.3	65.4	75.9	42.8	65.1	72.7
With unregistered spouse (had ceremony)	28.1	20.2	10.8	22.0	14.3	5.3
With cohabiter (no registration or ceremony)	4.0	0.7	0.2	3.5	0.4	0.5
Live apart from spouse	3.1	2.7	2.5	12.5	4.7	1.9
Not living with a partner	37.6	11.0	10.6	19.2	15.5	19.5
N	1,847	485	288	1,762	461	256
Religion
Buddhism	95.5	97.6	95.8	97.3	96.4	94.0
Islam	3.2	2.1	4.2	1.4	2.2	4.3
Christianity	1.3	0.3	0.0	1.3	1.3	1.7
N	2,272	491	288	2,272	480	272

^*Differences by age group are significant for all measures p<.05

Table 4.

Sexual experience with different types of partners in the last year

	Male			Female
	Adults 18–39	Adults 40–49	Adults 50–59	Adults 18–39	Adults 40–49	Adults 50–59
Had a cohabiting partner***	59.6	86.3	87.0	80.7	81.3	74.7
N	1,847	485	287	1,501	432	356
Frequency of sex with cohabiting partner***
In last 3 months
Once a week or more	38.5	54.1	66.3	47.5	65.8	56.5
Had sex less than once a week	55.7	40.3	15.5	39.0	19.3	3.7
Not at all	5.8	5.5	18.1	13.5	15.0	39.9
Always use condom with regular partner***	1.8	0.4	0.7	1.6	3.3	0.0
Had sex with a casual partner	17.6	4.4	2.1	0.6	0.2	0.0
Always used condom with casual partner**	42.6	28.6	14.2	-	-	-
Had sex with a sex worker***	8.7	5.0	1.6	0.1	0.0	0.0
Always used condom with sex worker*	93.9	98.6	66.7	-	-	-

Note: Data include only respondents who ever had sex with opposite sex partners.

^***p<.01 for males and females.

^**p<.01 for males.

^*p=.06 for males.

Table 5.

Logistic regressions for factors related to sex with a casual partner in the last year among persons age 40–59.

Dependent Variable	Sex with a casual partner	Always use a condom with casual partner	Sex with a sex worker
	Odds ratio (95% CI)	Odds ratio (95% CI)	Odds ratio (95% CI)
AIDS Knowledge	0.97 (0.88,1.06)	0.96 (0.83,1.12)	1.01 (0.82,1.24)
Not living with a partner	1.47 (0.82,2.65)	0.78 (0.33,34.15)	5.98*** (2.59,13.80)
Frequency of sex with live in partner At least once a week	0.97 (0.66,1.40)	0.91 (0.42,1.88)	0.44* (0.17,1.14)
Less than once a week (reference)	1.00	1.00	1.00
Residence
Bangkok	0.88 (0.57,1.37)	1.35 (0.70,2.64)	2.49 (0.92,6.80)
Other urban	1.06 (0.70,1.60)	0.60 (0.30,1.17)	2.01 (0.74,5.43)
Rural (reference)	1.00	1.00	1.00
Education University	2.70*** (1.54,4.71)	2.66** (1.24,5.70)	2.69* (0.95,7.60)
Upper secondary	1.78** (1.10,2.88)	4.40*** (2.16, 8.95)	2.26* (0.87,5.91)
Lower secondary	1.55* (0.92,2.60)	1.66 (0.76,3.64)	1.88 (0.69,5.16)
Primary (reference)	1.00	1.00	1.00
Age first sex	0.86*** (0.81,0.91)	0.89** (0.80,0.98)	0.92 (0.81,1.04)
Type partner first sex
Spouse	0.17*** (0.09,0.36)	6.55** (1.47, 29.11)	0.48 (0.05,4.20)
Sex worker	0.43*** (0.29,0.66)	3.37*** (1.56,7.26)	1.62 (0.63,4.21)
Friend	3.42*** (1.82,6.45)	3.02*** (1.34,6.81)	1.61 (0.45,5.76)
Other	0.03*** (4.48,18.21)	1.95* (0.91,4.17)	2.96** (1.00,8.79)
Boy/girl friend (reference)	1.00	1.00	1.00
N	758	328	758
Chi-Square	248.53***	48.33***	56.15***

^***p<.01,

^**p<.05,

^*p<.10

Results

Demographics

Table 1 shows the demographics of the study sample by age group. The majority of Thai men and women age 40 or more have just primary education, while many more of the adults under 40 (60% of females and 66% of males) have at least some secondary education. The data document the increase in the educational level of the Thai population over the last twenty years. Most of the respondents were working though there was some decline in labor force participation between the forties and fifties (96% to 89% for men, 74% to 67% for women).

The majority of men and women were married and living with a partner. The proportion of women not living with a partner in their fifties (22%) was higher than that for men (11%).

AIDS knowledge, testing and stigma

The AIDS knowledge scores for older adults in the sample were high, indicating a good understanding of how AIDS is transmitted and how it can be prevented (Table 2). Knowledge of antiretroviral therapy (ARV) however was lower, with less than half of male and female adults reporting that they know about ARV. Indeed, only 28% of women age 50–59 reported knowing about ARV.

Table 2.

AIDS knowledge and testing

	Male			Female
	Adults 18–39	Adults 40–49	Adults 50–59	Adults 18–39	Adults 40–49	Adults 50–59
AIDS knowledge score (Range 0–12)	11.9	11.8	11.5	11.7	11.3	10.6
Knowledge of ARV therapy	50.6	48.5	39.1	44.0	44.5	43.7
Stigma scale	4.4	4.2	3.9	4.5	4.4	4.1
Ever had HIV test	42.9	47.2	38.7	62.6	40.5	15.7
N	2,238	491	288	2,271	480	271
Reason for last test
Pregnant	-	-	-	67.0	26.4	8.2
Job application	10.9	6.9	0.6	1.7	0.6	1.8
Marriage	6.2	4.3	2.0	0.3	0.6	0.0
Military recruitment	5.0	0.1	0.0	-	-	-
Partner has risk behavior	0.8	0.7	0.0	0.5	19.9	0.5
Respondent has risk behaviour	2.2	1.5	0.7	0.0	0.0	0.0
Partner HIV positive	0.0	0.0	0.0	0.0	0.0	0.0
Health check up	13.6	33.0	45.9	6.6	30.2	50.8
Operation/ sick	11.4	15.3	17.6	10.8	9.8	31.4
Wanted to know	19.5	13.3	15.3	11.4	4.6	2.3
Blood donation	13.4	17.0	11.4	1.4	2.6	2.8
Other	17.0	7.8	6.2	0.6	5.2	1.8
N	868	245	120	1,152	231	49

^*Differences by age group are significant for all measures except the stigma scale for women, p<.05

Many of the older adults in the sample had been tested for HIV. About 47% of male 40–49 and 39% of males age 50–59 had an HIV test. For these two groups of men, the tests were done as part of a health check up or for an operation or blood test. However, 17% of men age 40–49 and 11% of men age 50–50 had the test because they wanted to know their HIV status.

Compared to men, smaller proportions of women reported an HIV test. Among women age 40–49, 41% reported a test while 21% of women age 50–59 reported a test. Among women age 40–49, many women (31%) had the test because they were pregnant, while another 46% had the test for a health checkup or an operation. Among women in their fifties, only 5% had the test because they were pregnant, while 74% had the test for a checkup or operation. Only 6–7% of women in their forties and fifties reported testing because they wanted to know their HIV status.

Sexual experience

First sexual experience

Ninety-five percent or more of adults in their forties and fifties reported that they had ever had sex (Table 3). Among respondents in these age groups, the median age at first sex was 18 to 19 for men and 20 for women. Data on the first partner show the changes that have taken place over the last 20 years in sexual behavior of young Thai adults. The older men in the study were more likely to have their first sex with their spouse or a sex worker than were younger men. Younger men were more likely than older men to have had their first sex with a girlfriend. The first partner for about one third of the male adults in their forties and fifties was a girlfriend. In contrast, among male Thai adults under age 40, more than half (54%) reported that their first partner was a girlfriend. Older men were also more likely to have their first partner be a spouse (24% for men in their fifties). A large difference can also be seen by age group in men whose first partner was a sex worker (22% men age 50–59, 25% men age 40–49, men age 18–39 11%). Condom use at first sex was only 10% to 12% for older men, but was at a higher lever (31%) for men under 40. Condom use at first sex was very low for women in their 40s and 50s (6% and 2% respectively), although most of these women had their first sex with their husbands.

Table 3.

First sexual experience and lifetime partners

	Male			Female
	Adults 18–39	Adults 40–49	Adults 50–59	Adults 18–39	Adults 40–49	Adults 50–59
Ever had sex***	89.4	98.8	99.7	85.2	95.6	95.7
N	2,202	491	288	2,092	457	375
For respondents who ever had sex:
Median age first sex***	18	18	19	19	20	20
First Partner***
Boy/girlfriend	54.5	36.5	34.6	36.5	16.3	11.3
Friend	17.3	10.1	7.7	2.2	1.2	0.5
Acquaintance	10.3	13.5	10.7	1.6	1.8	1.4
Fiancée	0.0	0.0	0.0	0.9	0.2	1.0
Spouse - registered	2.8	9.7	17.0	20.5	39.1	41.5
Spouse – not registered	3.4	4.4	6.7	37.9	40.8	44.0
Sex worker	11.2	25.3	21.8	0.3	0.6	0.2
Other	0.4	0.5	1.5
Condom use at first sex***	30.7	12.2	10.1	15.2	6.3	2.3
Lifetime partners
Had a regular partner**	85.5	96.9	97.8	99.7	99.4	100.0
Had a casual partner*	56.6	48.8	36.4	3.5	2.1	0.7
Had a paid partner**	37.4	45.9	42.1	0.1	0.2	0.0
N	1,847	485	287	1,501	432	356

^***p<.01 male and female.

^**p<.01 male only

Lifetime partners

In terms of lifetime partners, almost all of the older adults reported that they had a regular partner in their lifetime. Lifetime casual partners were also often reported by men in their forties (49%) and fifties (36%), but were more often reported by younger men (57%). In contrast, paid partners were more often reported by men in their forties (46%) and fifties (42%) than by men in under 40 (37%). Only a small number of women reported having a casual partner, although the proportion was higher in the younger age group (3% women under 40, 2% women 40–49, 1% women 50–59). Only a few women reported paid partners.

Sexual experience in the last year

Table 4 shows the number and type of partners that respondents reported in the last year. A cohabiting partner was reported for 86–87% of men in their forties and fifties. Among women, 81% reported a cohabiting partner for respondents in their forties, while 75% of women in their fifties reported a cohabiting partner.

The reported frequency of sex with cohabiting partners varied by age and gender. Compared to respondents under 40, lower frequencies of sex were reported by the older adults. While more than half of the males under age 40 reported sex once a week or more, only 40% of males age 40–49 and 15% of males age 50–50 reported this frequency. Among women, 39% of respondents under 40 while 19% of women in their forties and 4% of women in their fifties reported sex once a week. Only a small number of respondents reported that they always use condoms with their cohabiting partner.

Older men also reported fewer casual partners and fewer visits to sex workers. While 18% of men under age 40 reported sex with a casual partner, only 4% of men in their forties and 2% of men in their fifties reported a casual partner. Condom use with casual partners was also lower for older men. While 43% of men under age 40 reported that they always use a condom with casual partners, only 29% of men in their forties and 14% of men in their fifties reported consistent condom use with casual partners. Very few women reported casual partners

Finally, fewer older men reported having sex with a sex worker in the last year. While nine percent of men under 40 reported a paid partner, five percent of men 40–49 reported a partner and two percent of men in their fifties reported a paid partner. While 99 percent of men in their forties reported always using condoms with a sex worker, only 67 percent of men in their fifties reported always using condoms with paid partners.

Factors related to risk behavior

The most common risk behaviors reported by men in their forties and fifties were sex with a casual partner, condom use with a casual partner, and sex with a sex worker. These behaviors were risky due to the low level of consistent condom use with casual partners for all men and condom use with sex workers for men in their fifties. Models are not shown for women due to the small number of casual partners or other risk behaviors reported. Logistic regressions were estimated for three dependent variables: 1) sex with a casual partner in the last year, 2) always used a condom with a casual partner in the last year, and 3) sex with a sex worker in the last year. Independent variables in the model included AIDS knowledge, urban-rural residence, cohabiting with a partner, frequency of sex with a cohabiting partner, age at first sex, and the type of partner at first sex (Table 5).

The models are shown for men in Table 5. In the first model for men with the dependent variable sex with a casual partner, AIDS knowledge, not living with a partner, and frequency of sex with a live in partner were not significantly related to having a casual partner. Education had a positive significant association with having a casual partner. Men with university, upper secondary and lower secondary education were more likely to have a casual partner than were men with just an elementary school education.

To examine the effects of sexual experience early in life, the model included two added the variables related to first sexual experience, age at first sex and type of partner at first sex. For males, age at first sex was significantly associated with having a casual partner. The older the age at first sex, the lower the likelihood of having a casual partner in the last year, OR 0.86, 95%CI 0.81–0.91, p<.001. The type of partner at first sex was also related to the likelihood of a casual partner. Men whose first partner was a spouse or a sex worker were less likely to have a casual partner than were men whose first partner was a boyfriend or girlfriend (p<.01). Men whose first partner was a friend or an acquaintance were more likely to have a casual partner than men whose first partner was a boyfriend or a girlfriend (p<.01).

Table 5 also shows the factors related to always using a condom with a casual partner for males (model 2). As with the model for having a casual partner, only education and the variables related to the first sexual experience were significantly associated with condom use with casual partners for men. Men with a university or upper seocnday education were more likely to always use a condom with casual partners than were men with a primary education. Men with a higher age at first sex were less likely to use always use a condom with a casual partner (OR 0.89, 95% CI 0.80,0.98). The type of partner at first sex was also related to consistent condom use. First sex with a sex worker, spouse or friend, as compared to a girlfriend was related to more consistent use of condoms with a casual partner.

Models for sex with a sex worker in the last year are shown in Table 5 for men (model 3). AIDS knowledge was not related to having sex with a sex worker. Men who were not living with a partner were more likely to visit a sex worker, OR 5.98, 95%CI 2.59, 13.80, p<.01. Frequency of sex with a cohabiting partner once a week or more had an odds ration of 0.44, CI 0.17, 1.14) significant only at the .10 level. Although the odds ratios for visiting sex workers were greater than 2 for urban men in Bangkok and elsewhere, they were not statistically significant. Finally, education was related to visiting a sex worker. Men with university or upper secondary education were more likely to visit sex workers than were men with only a primary education (p<.10). Most of the variables representing early sexual experience did not have a significant effect of whether the man had visited a sex worker in the last year. Only one new coefficient, first sex with an “other” partner (mainly an acquaintance or stranger) was related to sex with a sex worker (OR 2.96, 95% CI 1.0, 8.79, p<.05).

Summary and Discussion

Data from the 2006 Thai National Survey of Sexual Behavior show that older Thai men and women are well informed about methods of HIV transmission, but many are unaware of ARV. Older adults were also less accepting of HIV positive persons than were younger adults. Less than half of adults age 40 or more have undergone HIV testing, mainly as part of medical checkups or before operations. Only a small percent of these older adults had a test because they wanted to know their status. Overall, the frequency of sexual activity decreased with age. Sexual frequency with marital/cohabiting partners was lower for older than for younger adults. The proportion of men with casual and/or paid partners was also lower for older adults. Among men, factors related to having sex with a casual partner were higher levels of education, early age at first sex, and first sex with someone who was not a boyfriend or girlfriend. Reported condom use was very low with regular partners, moderate with casual partners, and high with sex workers.

The study has some limitations. First, the study is based on reports of personal behaviors that cannot be verified. Second, the association of early sexual experiences with current behavior may also be due to differences between respondents in personality characteristics such as sensation seeking behaviors (Zuckerman, 1990) and the data set does not include measures of these characteristics. Third, this is a cross sectional survey, which limits the inferences about causality that can be drawn.

Consistent with the life course model, the age and type of partner at first sex had some relationship with later sexual behavior for men, suggesting that early sexual experiences have some influence on later behavior. Age at first sex and the type of partner at first sex was also related to having a casual partner as an older adult. Compared to sex with a boy/girlfriend, having first sex with a spouse reduced the likelihood of a casual partner. The first partner being a sex worker also reduced the likelihood of a casual partner. In contrast, if the first partner was a friend, men were more likely to report a casual partner compared to those whose first partner was a boyfriend or girlfriend. Hence, having a friend, who could be considered a casual partner at first intercourse may be related to later sexual activity. Furthermore, having visited a sex worker in the last year was related to first sex with someone with no personal relationship to the respondent. In general, first sexual relationships with persons with stronger ties to the respondent was related to lower levels of risk taking in later life.

While about a quarter of the older men reported that their first partner was a sex worker, only a small proportion of older men reported sex with paid partners in the last year. This may be due to the fact that most of them have regular partners, as well as norms for older men and declining sexual desire with age. This may also reflect changing norms, attitudes, and behaviors of young Thai men in response to the AIDS epidemic and accompanying prevention programs. A qualitative study of older men noted that while the men reported that going to the brothel was common when they were young, they have different priorities as they age (VanLandingham & Knodel, 2006). They became less focused on sex and having children and more on the well-being of their families. In addition, brothels were seen as a place for younger men where older men would be out of place. Finally, many of the older men did not want the health risks of brothels.

The proportion of men with casual partners was also smaller for older men. This may also reflect the presence of a regular partner and perceived health and financial risks of these partners. Among older men with casual partners, the level of condom use was low, suggesting a need for health promotion among these men.

The older men and women in the study had a good level of knowledge of transmission and prevention of AIDS. An earlier study in Thailand (Im-em, VanLandingham, Knodel & Saengtienchai, 2002) noted that older persons had inaccurate knowledge of casual transmission, but that was not confirmed here. On the other hand, the low levels of knowledge of ARV among older adults is of concern because it may reduce their interest in HIV testing.

Older adults also had lower levels of acceptance of persons living with AIDS. Attention should be given to this issue because persons in this age group are likely to be working with persons with AIDS and a small proportion may be involved in care of friends or relative who have periods of illness due to AIDS.

In summary, lower levels of risk behaviors were observed in older Thai adults compared to younger adults. Although older men were less likely to visit sex workers, when they did they were less likely to use condoms. The characteristics of early sexual experiences were also associated with later sexual behavior. Areas of concern for AIDS prevention programs include condom use with casual partners and paid partners, knowledge of ARV, and attitudes toward persons living with AIDS.