Sexual Experiences According to a Federal Recent Report

Counting vaginal intercourse, oral sex, and anal sex, 10 percent of males 15-44 years of age had never had sex in their lives, 6 percent had had sex with a female partner in their lives but not in the last 12 months, 63 percent had one female partner in the last 12 months, and 18 percent had two or more female partners in that period.

Among adult males 25-44 years of age, 97 percent have had sexual contact with an opposite-sex partner in their lives; 97 percent have had vaginal intercourse, 90 percent have had oral sex with a female, and 40 percent have had anal sex with a female. 

Among women, the proportions who have had sexual contact with an opposite-sex partner were similar. 

Among males 15-44 years of age, 5.7 percent have had oral sex with another male at some time in their lives, and 3.7 percent have had anal sex with another male; overall, 6.0 percent have had oral or anal sex with another male.

The proportion of men who had a male sexual partner in the last 12 months was 2.9 percent, or approximately 1.77 million men (table A). 

The proportion of men who had only male sexual partners in the last 12 months was 1.6 percent (table B).

Among women 15-44 years of age in 2002, 11 percent answered yes when asked, “Have you ever had any sexual experience of any kind with another female?” (The wording of this question was less restrictive and less specific than the questions about men having oral or anal sex with men.) 

The proportion of women who had a female sexual partner in the last 12 months was 4.4 percent, or approximately 2.71 million women (table A), and the proportion who had only female sexual partners in the last 12 months was 1.3 percent (table B). 

Percentage of males 15-24 years of age who have had vaginal intercourse, and percent who have had oral sex with a female but not vaginal intercourse, by age: United States, 2002 +

At ages 18-19, about 11 percent of years of age, 8 percent had never had sex, 7 percent had had sex with a male partner in their lives but not in the last 12 months, 68 percent had one male partner in the last 12 months, and 14 percent had two or more (figure 1, tables 1 and 2). 

At the ages 15-17, about 13 percent of males and 11 percent of females had had heterosexual oral sex but not vaginal intercourse. 

males and 9 percent of females had had oral sex but not vaginal intercourse. 

(The male-female percentages are not significantly different.) 

The proportion of males who have had oral sex with a female but not vaginal intercourse is 12 percent at age 15-19, and drops to 3 percent at age 22-24, when most males have already had vaginal intercourse (figure 2). 

The pattern by age is similar for females (figure 3). 

that they had had sexual experiences with members of their own sex may also have had opposite-sex partners in their lives.  

Approximately 1 percent of men and 3 percent of women 15-44 years of age have had both male and female sexual partners in the last 12 months (table B). 

Among females, 5.8 percent of teens and 4.8 percent of females 20-24 years of age had had both male and female partners in the last 12 months; percentages were lower at ages 25-44. 

Among men, about 1 percent had had both male and female partners in the last 12 months at each age. 

Percentage of females 15-24 years of age who have had vaginal intercourse, and percentage who have had oral sex with a male but not vaginal intercourse, by age: United States, 2002 their lifetimes. 

Among females 15-44 years of age, 4 percent of Hispanic women, 10 percent of white women, and 9 percent of black women have had 15 or more male partners in their lifetimes (figure 7). 

In response to a question that asked, "Do you think of yourself as heterosexual, homosexual, bisexual, or something else?" 90 percent of men 18-44 years of age responded that they think of themselves as heterosexual. 

About 2.3 percent of men answered homosexual, 1.8 percent bisexual, 3.9 percent "something else," and 1.8 percent did not give an answer (figure 8). 

Among women 18-44 years of age, 90 percent said they think of themselves as heterosexual, 1.3 percent homosexual, 2.8 percent bisexual, 3.8 percent "something else," and 1.8 percent did not answer the question. 

The category something else (other than heterosexual, homosexual, or bisexual) deserves further study using the NSFG and other surveys; it may reflect a lack of understanding of these terms by some respondents, a preference for other terms to describe sexual orientation, or both (figure 8). 

+ The percentages of men and women who reported that they think of themselves as homosexual or bisexual are roughly equal at 4.1 percent. 

This represents about 2.27 million men and 2.29 million women 18-44 years of age (tables 12 and 13). 

+ Another question was asked about whether the respondent was sexually attracted to males, to females, or to both. 

Among men 18-44 years of age, 92 percent said they were Figure 4. 

Percentage of males and females 25-44 years of age who have had each type of attracted only to females, and sexual contact: United States, 2002 3.9 percent, "mostly" to females. 

About 3.2 percent said they were related to the more specific wording difference in the number of sexual attracted only or mostly to males or of the questions that were asked of partners reported by men and women to both males and females equally; men (table B). 

has been noted in most previous U.S. this is equivalent to 1.8 million men + Males 30-44 years of age reported an studies of sexual behavior in the 18-44 years of age (table 14). 

average (median) of 6-8 female United States. 

+ Among women 18-44 years of age, sexual partners in their lifetimes. 

+ Among males 15-44 years of age, 86 percent said they were attracted Among women 30-44 years of age, 18 percent of Hispanic men, only to males, 10 percent said the median number of male sexual 22 percent of white men, and ‘‘mostly to males,’’ and 3.4 percent partners in their lifetimes was about 34 percent of black men have had 15 said only or mostly to females or 4, as shown in figure 6. 

This or more female sexual partners in equally to males and females. 

This 3.4 percent is equivalent to a Percentage of males 15-44 years of age who have ever had oral or anal sex with a same-sex partner: United States, 2002 estimated 1.9 million women 18-44 years of age (table 15). 

+ About 15 percent of men 15-44 years of age had been tested for HIV, outside of blood donation, in the 12 months before the interview (table 18). 

Among men who have ever had same-sex sexual contact in their lives, 29 percent were tested in the last year, compared with 14 percent of men with no same-sex sexual contact (figure 9 and table 18).

Among sexually experienced men 15-44 years of age, 7 percent reported having been treated for a sexually transmitted infection (STI) other than HIV. About 17 percent of those who ever had same-sex sexual contact had been treated for a non-HIV STI, compared with 7 percent of those who had never had same-sex sexual contact, as shown in figure 10 and table 19.

In the 12 months before the survey, 29 percent of men who had ever had sex with another male had been tested for HIV outside of blood donation and 11 percent had been tested for STIs, but not tested for HIV. Men without same-sex sexual contact were much less likely to have been tested for HIV or other STIs in the last year (table 18). 

+ Among men 15-44 years of age who had at least one sexual partner in the last 12 months, 39 percent used a condom at their most recent sex. 

Among never married males, this figure was 65 percent, compared with 24 percent of married males. 

Among males who had ever had sexual contact with another male, 91 percent used a condom at their last sex, compared with 36 percent of men who had never had sex with another male (table 21).

Background This report presents national estimates of several types of sexual behavior among men and women 15-44 years of age in the United States in 2002. 

Number of males and females 15-44 years of age and percentage who have ever had any same-sex sexual contact in the 12 months prior to interview and in lifetime, by selected characteristics: United States, 2002 Men Women Number in Last Number in Last 12 Characteristic thousands Lifetime 12 months thousands Lifetime months Percent Percent All respondents 15-44 years of age. 

About one-half of all STI occur to persons 15-24 years of age, and the direct medical cost of these diseases to 15-24-year-olds alone was estimated at $6.5 billion in the year 2000 (4). 

These diseases may not only affect the infected individuals but also be transmitted to their spouses, their partners, and among pregnant women, to their babies. 

Data for HIV/AIDS cases (in 33 areas with confidential name-based reporting) in 2002 suggest that 49 percent of HIV cases diagnosed in 2002 were transmitted by same-sex sexual contact among males, another 34 percent by heterosexual vaginal intercourse, 15 percent by injection drug use, and the other 2 percent by other or unknown means (1). 

Therefore, approximately 83 percent of cases were acquired through sexual behavior. 

In addition to concerns about the potential health consequences of sexual behavior among adults, there has been considerable public concern about sexual activity among teenagers—in part because teenage birth and pregnancy rates in the United States are higher than in many other industrial countries (5,6) and in part because of concerns about STIs (2). 

Educational campaigns in recent years have encouraged teenagers to delay sexual activity and some concern has been raised that teenagers may be responding to this message by engaging in oral sex, in order to prevent pregnancy (7-9). 

There is evidence, however, that certain diseases can be transmitted through oral sex, including gonorrhea, chlamydia, chancroid, and syphilis (10,11). 

Some groups may also be at elevated risk of HIV transmission through oral sex, including men who the General Social Survey, which has included some questions on sexual behavior in its national samples of adults 18 years of age and over since 1988 (18-22) These surveys were based on in-person interviews and used national probability samples. 

The data published here from the 2002 NSFG, however, are based on larger samples and are more recent than Figure 6. 

Median number of opposite-sex sexual partners in lifetime, by age and sex: all but the General Social Survey data. 

United States, 2002 In addition, some surveys limited to teenagers have collected data on some Figure 7. 

Percent of males and females 15-44 years of age with 15 or more opposite-sex sexual partners in their lifetime, by Hispanic origin and race: United States, 2002 have sex with men and certain drug users (12). 

Therefore, this report contains data on selected aspects of sexual behavior that were collected in the 2002 National Survey of Family Growth (NSFG, also known as NSFG Cycle 6). 

Comparisons to other national surveys are made to the extent possible, to shed light on the reliability of the results and to suggest whether any marked trends have occurred. 

Selected previous studies Several nationally representative studies of sexual behavior have been conducted in the United States in the last two decades. 

These include: aspects of sexual behavior, including the CDC's Youth Risk Behavior Survey, a biennial survey of high school students which included data on condom use, sexual intercourse, and numbers of sexual partners (23). 

The Urban Institute’s National Surveys of Adolescent Males (NSAM) collected and published data on the sexual activity and contraceptive use of national samples of males 15-19 years of age in 1988 and 1995 (24,25). 

National telephone surveys of sexual behavior in the adult population were conducted by Catania et al. 

(26-28) in the 1990s. 

Finally, results from a 1970 survey conducted by the National Opinion Research Center (NORC) were published in 1989 and contained some data on same-sex sexual contact among males (29). 

In response to the need for reliable national data on sexual behavior related to public health concerns about HIV/AIDS and other STI (30-33), the staff of the NSFG worked with CDC’s Division of HIV/AIDS Prevention, other collaborating agencies, and experts who had conducted surveys on closely related topics (13-19), to develop questions that would collect the most reliable data possible on the sexual and drug-related behaviors that affect the risk of HIV and other STIs. 

These data were collected in the 2002 NSFG using Figure 8. 

Percent distribution of men and women 18-44 years of age by sexual orientation: United States, 2002 Figure 9. 

Percentage tested for HIV (ever and in the last 12 months), by sex and whether ever had a same-sex partner: United States, 2002 Audio Computer-Assisted Self- Interviewing, or ACASI, in which the respondent listens to the questions through headphones, reads them on the screen, or both, and enters the response directly into the computer. 

This method affords the respondent greater privacy and it has been found to yield more complete reporting of sensitive behaviors (34). 

ACASI may also make it possible for persons with lower literacy to complete the self-interview by listening to the questions instead of reading them. 

The data on types of sexual behavior that are shown in this report were collected using ACASI, unless otherwise noted. 

Results of the 2002 NSFG on the heterosexual activity, contraceptive use, and childbearing experience of American teenagers in 2002 (5), and the contraceptive use of women 15-44 years of age (35), have been reported previously. 

A forthcoming report (30) will show national estimates of HIV testing and counseling for men and women 15-44 years of age. 

The present report presents data on heterosexual vaginal intercourse as well as other types of sexual activity from the ACASI portion of the interview.

The NSFG has been conducted six times by NCHS: in 1973 and 1976 with samples of married and formerly married women; in 1982, 1988, and 1995, with samples of women of all marital status categories; and in 2002, with national samples of both women and men 15-44 years of age (36). 

Each time, the interviews have been conducted in person by trained female interviewers in the selected persons' homes. 

The sample is a nationally representative multistage area probability sample drawn from 121 areas across the United States. 

Large areas (counties and cities) were chosen first; within each large area or "Primary Sampling Unit," groups of adjacent blocks called segments were chosen at random. 

Within segments, addresses were listed and some addresses were selected at random. 

The selected addresses were visited in person, and a short screener interview was conducted to see if anyone 15-44 years of age lived there. 

If so, one person was chosen at random for the interview and was offered a chance to participate. 

To protect the respondent's privacy, only one person was interviewed in each selected household. 

In 2002, teenagers and black and Hispanic adults were sampled at higher rates than others. 

All respondents were given written and oral information about the survey and were informed that participation was voluntary. 

Adult respondents 18-44 years of age were asked to sign a consent form but were not required to do so. 

For minors 15-17 years of age, signed consent was required first from a parent or guardian, and then signed assent was required from the minor; if the parent declined to give written consent, or the minor did not give written assent, the minor did not participate in the survey. 

Respondents were assured that the confidentiality of their information would be protected. 

The response rate for the survey was 79 percent overall: 80 percent for women, 78 percent for men, and 81 percent for teenagers. 

Percentage of sexuallly experienced men who have ever had a sexually transmitted infection other than HIV, by whether ever had sex with another man, and number and sex of partners in the last 12 months: United states, 2002 Over 200 female interviewers were hired and trained by the survey contractor, the University of Michigan’s Institute for Social Research, under the supervision of NCHS. 

Interviewing occurred from March 2002 until early March 2003. 

Much of the data in the survey were collected by Computer- Assisted Personal Interviewing, or CAPI, in which the questionnaire was stored on a laptop computer, and administered by an interviewer, but most of the data in this report, including the most sensitive items, were collected using ACASI, (described previously). 

Respondents in the 2002 survey were offered $40 as a ‘‘token of appreciation’’ for their participation. 

The NSFG questionnaires and materials were reviewed and approved by both the CDC/NCHS Research Ethics Review Board and the University of Michigan Institutional Review Board. 

The female questionnaire lasted an average of about 85 minutes and the male questionnaire an average of 60 minutes. 

More detailed information about the methods and procedures of the study was published in a separate report (36). 

In this report, the term "intercourse" refers to heterosexual vaginal intercourse. 

The term "sex" refers to all types of sexual activity including vaginal intercourse, oral sex, and anal sex. 

The definition of Hispanic origin and race used in this report takes into account the reporting of more than one race, in accordance with OMB guidelines that are described further in the Technical Notes. For convenience in writing, the short term black or non-Hispanic black will be used instead of the full phrase, non-Hispanic black or African American, single race. Similarly, the term white or non- Hispanic white will be used instead of the full phrase, Non-Hispanic white, single race. The full terms are shown in the tables.

The NSFG has a rigorous probability sampling design, so the estimates can be generalized to the national household population.

The response rate for the NSFG was 79 percent, which is considered high, and suggests that the data for most statistics can be generalized to the population with confidence.

Questions asked on the NSFG have undergone testing and review in an effort to make them understandable to persons participating in the survey.

Sensitive questions associated with sexual behavior, reproductive health, or drug abuse were collected using ACASI methods. 

These methods have been found to yield more complete reporting of sensitive behaviors, and they also avoid the large amounts of missing data often found on paper- and-pencil self-administered questionnaires (20,21,29).

The questionnaire was administered in both English and Spanish; those who preferred to answer the interview in Spanish were interviewed by bilingual interviewers. 

The translation of the questionnaire into Spanish was done with particular attention to making it understandable and culturally appropriate to major Hispanic groups—including Mexicans, Puerto Ricans, recent immigrants, and those with limited education (36,37).

As a household-based sample survey, the NSFG excludes from the sampling frame the homeless, persons who are incarcerated or otherwise institutionalized, and those living on military bases in the United States. 

The results cannot be generalized to those populations; they may have different patterns of sexual behavior.

As in any survey, nonresponse errors could affect the results. 

The NSFG makes use of extensive quality control procedures to try to minimize the effects of such errors.

The results could be affected by underreporting of sensitive behaviors, although using ACASI as used in the NSFG has been found to yield more complete reporting of these items than other types of questionnaires (34).

The NSFG provides national estimates, but cannot provide State or local estimates of the behaviors described in this report.

The age range of the NSFG is 15-44 years of age. 

Therefore it is not possible to measure the behavior of those under age 15 or over 44 years of age. 

+ Given the sample size of the 2002 NSFG, the numbers of men and women in the sample who have had sexual contact with same-sex partners, while larger than in most other studies, are still relatively small, so the sampling errors of percentages for these groups are larger than they are for larger groups. 

It also means that the amount of subgroup analysis (for example, by age, race or ethnicity, and other characteristics) that can be done with acceptable sampling errors for these populations is limited. 

+ The scope of this report is limited to a few measures of sexual behavior, attraction, and orientation. 

There are no data in this report on topics such as civil unions or family life among same-sex couples. 

Measurement of sexual behavior This section will define some sexual behaviors measured in this report, in part by showing the wording of the questions asking about them. 

The NSFG is historically and primarily a study used to measure factors related to pregnancy and birth rates (3,38). 

For this purpose, much of the main part of the interview, administered by the interviewer, is concerned with behaviors that are related to birth and pregnancy rates—namely, contraceptive use, infertility, breastfeeding, heterosexual marriage and cohabitation, heterosexual vaginal intercourse, and demographic characteristics such as age, education, marital status, and race. 

The self- administered, or ACASI, part of the interview was focused primarily on topics related to the risk of acquiring or spreading HIV and sexually transmitted infections (STI). 

The ‘‘Technical Notes’’ show a comparison of data on whether respondents have ever had heterosexual vaginal intercourse, as collected in the interviewer-administered part of the interview, and in the self-administered ACASI.

Women saw questions on the computer screen in ACASI that asked about a number of topics, including: Oral sex with a male: ‘‘Has a male ever put his mouth on your vagina (also known as cunnilingus or oral sex)?’’ ‘‘Have you ever put your mouth on a male’s penis (also known as fellatio or oral sex)?’’ Anal sex with a male: ‘‘Has a male ever put his penis in your rectum or butt (also known as anal sex)?’’ Same-sex contact: Women read a question on the computer screen that asked, ‘‘The next question asks about sexual experience you may have had with another female. 

Have you ever had any sexual experience of any kind with another female?’’ Note that this question is worded in such a way that a variety of experiences could be reported. 

The wording of this question may elicit more ‘‘yes’’ answers than the more restrictive or behavior-focused wording used for males in the NSFG (see the following text), and the wording used in some other surveys, which asked females specifically if they had had oral sex with another female.

Males were asked the following questions in ACASI (on the computer screen or over headphones). 

Males who were under 20 years of age and had never been married were asked: ‘‘If you have ever done this at least one time with a female, answer yes. 

If you have never done this, answer no. 

Has a female ever touched your penis until you ejaculated, or came Other questions in the self-administered portion of the interview asked of all males included: Oral sex with a female:

Has a female ever put her mouth on your penis (also known as oral sex or fellatio)?

Have you ever put your mouth on a female's vagina (also known as oral sex or cunnilingus)?

Anal sex with a female:

Have you ever put your penis in a female's rectum or butt (also known as anal sex)?

Same-sex contact: Men were asked in ACASI (on the computer screen or over headphones), The next questions ask about sexual experience you may have had with another male. 

Have you ever done any of the following with another male? 

Put his penis in your mouth (oral sex)? 

Put your penis in his mouth (oral sex)? 

Put his penis in your rectum or butt (anal sex)? 

Put your penis in his rectum or butt (anal sex)?

A yes answer to any of these four questions was classified as same-sex sexual contact. Note that these questions are more specific than the single question on female same-sex behavior. 

Questions on sexual attraction and orientation After answering questions in ACASI on all of these sexual behaviors, as well as other behaviors related to risk for STI and HIV, respondents were asked questions on sexual attraction and orientation. 

Like the other questions in ACASI, they were read on the computer screen or heard over headphones; they were not asked by an interviewer. 

For females, the questions were: People are different in their sexual attraction to other people. 

Which best describes your feelings?  Are you. 

Only attracted to males Mostly attracted to males Equally attracted to males and females Mostly attracted to females Only attracted to females Not sure.

Do you think of yourself as Heterosexual Homosexual Bisexual Or something else?’’

For males, these questions were: People are different in their sexual attraction to other people. 

Which best describes your feelings?  Are you. 

Only attracted to females Mostly attracted to females Equally attracted to females and males Mostly attracted to males Only attracted to males Not sure.

Do you think of yourself as Heterosexual Homosexual Bisexual Or something else?’’ The wording of these items was based on wording used in previous studies (13-18), along with consultations with the directors of many of those studies and other experts (36). 

In the rest of this report, when the text says that respondents were asked something in ACASI, it means that they saw the question on the computer screen, or heard the question through headphones, or both. 

Statistical analysis All estimates in this report are based on sampling weights that are designed to produce unbiased estimates for the U.S. population. 

The statistical package SAS Version 9 was used to produce all estimates of percentages and numbers in this report (www.sas.com). 

SUDAAN software was used to estimate the sampling errors of the statistics; this software takes into account the use of weighted data and the complex design of the sample in calculating estimates of standard errors and significance tests (39). 

The standard errors of the percentages shown in this report are shown in some of the tables of this report and in Appendix tables. 

In the description of the results below, when the percentage being cited is below 10 percent, the text will cite the exact percentage to one decimal point. 

To make reading easier and to remind the reader that the results are based on samples and subject to sampling error, percentages above 10 percent will be shown rounded to the nearest whole percent. 

Readers should pay close attention to the sampling errors and confidence intervals of small groups, such as subgroups of teenagers or men who have sex with men. 

In this report, percentages are not shown if the denominator is less than 50 cases, or the numerator is less than 3 cases. 

When a percentage or other statistic is not shown for this reason, the table contains a symbol(*) that means ‘‘Statistic does not meet standards of reliability or precision.’’ For most statistics, the numerators and denominators are much larger.

Sexual activity of any type in the last 12 months Table 1 shows men 15-44 years of age by the number of female sexual partners they had in the 12 months before the interview. 

This answer includes vaginal intercourse and oral or anal sex with females. 

For all males 15-44, 10 percent had never had sex, 6.4 percent had had sex in their lifetimes but not in the last 12 months, and 63 percent had had one sexual partner in the last year. 

Another 8.0 percent had two female partners in the last 12 months, and 10 percent had three or more female partners in the last 12 months. 

Comparing all men and all women in tables 1 and 2, about 18 percent of men and 14 percent of women reported having two or more partners in the last 12 months (tables 1 and 2 and figure 1). 

The highest proportions of men with three or more partners in the last 12 months were found among men 20-24 years of age (19 percent) and black or African-American men (22 percent, table 1). 

Among women, those 18-19 years of age had the highest proportion reporting three or more male partners in the last 12 months (17 percent). 

These variations in tables 1 and 2 appear to be strongly related to marital status differences among the groups: for example, the proportions of black males with more than one partner in the last 12 months is related to the fact that they are less likely to be married than white males. 

Tables 1 and 2 also contain data by marital status on the date of interview. 

Since a mutually monogamous relationship is one way to reduce the risk of sexually transmitted infections, including HIV, previous studies have shown data on the proportion of married men and women who have had more than one partner in the last 12 months (17, p. 

102; 18,27,28). 

About 4.5 percent of all married men (table 1) and 3.8 percent of married women (table 2) reported more than one partner in the last 12 months. 

These results are generally consistent with the results of other surveys for men and women, but it is important to note that these statistics are not certain evidence of infidelity. 

Some marriages were started within the last 12 months, and it is possible that temporary separations could have occurred within the last 12 months in a small percentage of marriages. 

Tables 3 and 4 show data on the proportion of males and females 15-24 years of age who have ever had any sexual contact with the opposite sex; and the proportion who have had vaginal intercourse, oral sex, anal sex, any same-sex contact, and no sexual contact. 

These data relate to concerns that teenagers may be delaying vaginal sexual intercourse or fulfilling virginity pledges by engaging in other types of sexual contact (7-9, 40, 41). 

Although these data cannot answer this question fully, they can show the proportion of young males and females who have had any sexual contact with another person and what types are most common in various subpopulations. 

It is important to recognize, however, that the data in tables 3 and 4 all refer to whether a person has ever done these things, even once, in their lives; the proportion who have done them recently, or are doing them currently, is lower. 

Concerns with oral sexual activity among middle-school children (11-14 years of age) cannot be addressed directly with these data because the youngest respondents in the NSFG were 15 years of age when interviewed. 

(In a recent survey conducted by Princeton Survey Research Associates International for NBC News, 4 percent of teens 13-14 years of age reported (in telephone interviews) that they have had oral sex: http://msnbc.msn.com/id/ 6839072 (accessed on February 7, 2005). 

Table 3 shows that 36 percent of 15-17-year-old males had had vaginal intercourse with a female; 28 percent had given oral sex to a female, and 40 percent had received it; 8.1 percent had had anal sex with a female, and 46 percent of 15-17-year-old males had had no sexual contact with another person. 

Among males 18-19 years of age, 66 percent had had vaginal intercourse with a female, while 52 percent had given oral sex to a female, 66 percent had received it, and 15 percent had had anal sex with a female. 

About 22 percent of 18-19-year-old males had not had sexual contact with another person. 

Table 4 contains statistics on the proportions of females 15-24 years of age with these same types of sexual contact. 

Among 15-17-year-old females, 39 percent had had vaginal intercourse at least once, 30 percent had given oral sex to a male, and 38 percent had received it from a male. 

Nearly one-half (49 percent) of 15-17-year-old females had not had any sexual contact with another person. 

At ages 18-21, the proportions of females who had had vaginal intercourse were about the same as the proportions who had had oral sex. 

At age 18-19, 17 percent of females had not had any sexual contact with another person, compared with about 22 percent of 18-19-year-old males. 

The data in tables 3 and 4 suggest that the prevalence of oral sex among teenagers differs by Hispanic origin and race. 

In table 3, black teen males were more likely than white teen males to have had vaginal intercourse (66 compared with 45 percent), but less likely than white teen males to have given oral sex to a female (45 percent of white male teens compared with 20 percent of black male teens). 

A similar pattern is found for female teens: black female teens were more likely to have had vaginal intercourse (62 compared with 52 percent), but less likely than white female teens to have given oral sex (25 compared with 51 percent). 

These results appear consistent with findings for adults on oral sex in the NSFG and the NHSLS (16; 17, p. 141). 

They are also consistent with a study by Smith and Udry (41) in one large Southern city, which found that white teens were more likely than black teens to engage in behaviors other than vaginal intercourse for a period of time during the teen years. 

Smith and Udry speculated that these behaviors may be related to the race difference in age at first intercourse and in teen birth rates, but larger samples are needed to determine whether the patterns found in tables 3 and 4 could account for a significant part of the differences in birth and pregnancy rates between these groups. 

It should also be noted that there is less difference between black and white teens when both types of oral sex are combined. 

For example, 58 percent of white females 15-19 years of age and 53 percent of black females 15-19 years of age reported ever giving or receiving oral sex at all; the comparable percentages were 57 percent among white males 15-19 years of age and 59 percent among black males 15-19 years of age. 

Table 5 and figure 2 show the percent distribution of males 15-24 years of age by type of sexual contact with the opposite sex. 

The proportion of males who had had oral sex but not vaginal intercourse was about 10- 14 percent of males at each age from 15-19 years of age (figure 3). 

At age 15-17, 36 percent of males had had vaginal intercourse with a female. 

Another 13 percent had had oral sex but not vaginal intercourse, and 4.1 percent had had anal sex or some other sexual contact. 

The proportion of females 15-17 years of age who had had vaginal intercourse was 39 percent; another 11 percent had had oral sex but not vaginal intercourse; the proportion that had had other sexual contact but not vaginal or oral sex was close to zero (table 6). 

About 66 percent of 15-year old girls had not had any sexual contact with an opposite-sex partner. 

At 18-19 years of age, 74 percent of females had had vaginal intercourse, another 9.3 percent of females had had oral sex but not vaginal intercourse, and 17 percent had not had any sexual contact with a male (table 6). 

The data in table 5 by Hispanic origin and race show that 15 percent of non-Hispanic white males 15-19 years of age, compared with 7.3 percent of Hispanic and 7.6 percent of black teen males, have had oral sex but not vaginal intercourse. 

For female teens, 12 percent of white, 9.9 percent of Hispanic, and 5.8 percent of black females 15-19 years of age had had oral sex but not vaginal intercourse (table 6). 

The difference between Hispanic males and white males is significant, but the other differences by race and origin for male and female teenagers are not statistically significant at the 5-percent level. 

In the bottom panel of tables 5 and 6, those who have not had vaginal intercourse are tabulated by the main reason they have not done so. 

Among males and females who said it was against their religion or morals to have had sexual intercourse at their age, 82 percent of males and 78 percent of females had not had any sexual contact with another person. 

Among those who said they did not want to get an STD, 70 percent of males and 81 percent of females had not any sexual contact with an opposite-sex partner. 

Among females (table 6), 74 percent of those who did not want to get pregnant and 74 percent of those who ‘‘haven’t found the right person yet’’ also had not had any sexual contact with a male. 

The proportion of young males and females who have had anal sex with a member of the opposite sex is about 1 in 5 at age 15-24, but this sharply increases with age (tables 3 and 4). 

For males, the proportion who have had anal sex with a female increases from 4.6 percent at age 15 to 34 percent at ages 22-24 (table 3); for females, the proportion who have had anal sex with a male increases from 2.4 percent at age 15 to 32 percent at age 22-24 (table 4). 

While the percentages who have had anal sex with the opposite sex are lower than the percentages who have had oral sex, the risk of HIV or STI transmission is higher for anal than for oral sex. 

The risk of disease transmission from oral sex is also believed to be lower than for vaginal intercourse. 

The public health concern about oral sexual activity, particularly among teenagers, includes the concern that young people may not be aware that oral sex carries some risk of disease transmission (7-12). 

Types of sexual behavior at ages 15-44 years Table 7 shows data for males and females 15-44 years of age on the proportion who have ever had specified types of sexual contact. 

Among males 15-44 years of age, 91 percent have had sexual contact with females. 

Among adults 25-44 years of age, 97 percent of males had had vaginal intercourse with a female, 90 percent had had oral sex, and 40 percent anal sex with a female (figure 4). 

Overall, and in most of the adult age groups, about 5-8 percent of males had had oral or anal sexual contact with another male at some time in their lives. 

Table 7 also shows similar statistics for females. 

Among females 25-44 years of age, 98 percent of women had had sexual contact with a male: 98 percent had had vaginal intercourse; 88 percent had had oral sex; and 35 percent had had anal sex. 

Among females 25-44 years of age, 11 percent reported that they had had a sexual experience with another female at some time in their lives (table 7). 

Among women 20-29 years of age, about 14 percent have had sexual contact with another female. 

The question on same-sex experience for females was significantly different from the set of questions used for men. 

See the section of the text on ‘‘Measurement of sexual behavior’’ for details and specific question wording. 

Significantly more white males (87 percent) have had oral sex with a female compared with black (79 percent) and Hispanic (74 percent) males. 

Across Hispanic and racial categories, the proportions of males who had ever had anal sex with a female were similar: 31 percent of black, 33 percent of Hispanic, and 35 percent of white males. 

Similarly among females, the proportion who had ever had oral sex with a male was higher for white women (88 percent) than for black (75 percent) or Hispanic (68 percent) women. 

(These findings are similar to those in reference 17, p. 

141.)  Unlike males, there is a significant difference by race in the percentage of females who have ever had anal sex—34 percent of white women compared with 23 percent of Hispanic and 22 percent of black women report having had anal sex with a male. 

Table 8 and figure 5 show more detailed data on the proportion of males and females who had ever had sexual contact with a member of the same sex. 

Among men 15-44 years of age, 6.0 percent had ever had sexual contact with a male: 3.7 percent had had anal sex and 5.7 percent had had oral sex with another male (figure 5). 

Among adult men 25-44 years of age, 6.5 percent had had sexual contact with another man (table 8). 

After accounting for the sampling errors of these statistics, there is little or no trend by age among males in the proportions who had ever had sex with another man. 

Among women who have had more male partners, larger percentages had also had sexual contact with other females: 2.2 percent of women with 1 male partner in their lives had had sexual contact with another female compared with 32 percent of women with 15 or more male sexual partners. 

Finally, the proportion of men 25-44 years of age who have had oral sex with another man was 8.2 percent for those who have ever been in a jail, prison, or juvenile detention center, and 5.5 percent for those who have not. 

These data are shown for men 25-44 years of age. 

Women were not asked about prison or jail experience in the NSFG, because the NSFG’s sample size is too small to make reliable estimates for women. 

Table 9 shows that education and poverty level income are not generally associated with large differences in the proportions of males or females who have ever had same-sex contact. 

Data are also shown for men by whether they live in one of the 12 largest metropolitan areas, another metropolitan area, or a nonmetropolitan area. 

A somewhat larger proportion of males living in the central cities of the 12 largest metropolitan areas (6.2 percent) had had anal sex with another male compared with 2.3 percent of men in nonmetropolitan areas, a significant difference. 

One previous study found that the proportion with same-sex contact was substantially higher in the central cities of the 12 largest metropolitan areas than in the rest of the United States in 1988-93 (16, p. 

302-309; 42); these authors speculate that migration to these large cities may be one factor in this pattern. 

The data in table 9 provide partial confirmation of this pattern. 

By Hispanic origin and race, it appears that Hispanic women are less likely than white or black women to have had same-sex contact: 6.5 percent of Hispanic women have had sexual contact with another female compared with 11 percent of black women and 13 percent of white women. 

No comparable difference is apparent for Hispanic males (table 9). 

Table 10 shows the proportion of men 15-44 years of age by the number of female sexual partners they have had to this point in their lives. 

Table 11 shows similar data for females. 

The choice of categories shown in these tables was determined by the distribution of the data. 

While they are not meant to indicate any particular thresholds of risk for HIV or STI, many studies have documented that higher numbers of sexual partners are associated with increased risk of STIs. 

These tables show, as have other surveys in the United States and other countries, that men report more opposite-sex partners than women (15-19). 

For example, males 30-44 years of age reported an average (median) of 6-8 female sexual partners in their lifetimes. 

Females 30-44 reported an average (median) of about 4 male sexual partners in their lifetimes, as shown in figure 6. 

Some researchers have suggested that some of this difference is due to a small percentage of men who report very large numbers of partners (43). 

Laumann et al. (16, p. offered several possible explanations to consider: the ages of men and women at marriage, at cohabitation, or in sexual relationships are often similar, but not identical: some women, for example, may be having sex with men outside the age range covered here (particularly over age 44); some men who are not in the household population of the United States and therefore not covered in the NSFG sample (for example, in the military or in prison) may be having sex with women who are in the household population; or there may be a few women with very large numbers of male partners who do not respond to surveys. 

In addition, Laumann et al. speculate that ‘‘men and women may differ in what they consider a sex partner’’—that is, some men may include a brief sexual encounter when they report the number of partners they have had, while some women may not count a similar event (16, p. 185). 

Another explanation suggested by Laumann and colleagues is that the numbers of men and women in this age range may not be equal, particularly within subgroups of the population where numbers of sexual partners may be higher. 

For example, table 7 shows (in the ‘‘Number in thousands’’ column) that there were 61.147 million males and 61.561 million females 15-44 years of age in the household population of the United States in 2002, a difference of only 0.7 percent. 

But this imbalance differs in both direction and size for white, black, and Hispanic groups. 

Among non-Hispanic whites, there were 39.498 million females and 38.738 million males, a difference of 2.0 percent more females than males. 

Among Hispanics, however, there were 10.188 million males and 9.107 females 15-44 years of age, a difference of 1.081 million or a surplus of males of 10.6 percent. 

Among blacks, there were 8.250 million females and 6.940 million males, a difference of 1.310 million, or a surplus of females of 18.6 percent. 

A shortage of males (or an absence of males from the household population, due to higher percentages in the military or in prison), then, may be one factor that partly explains the larger number of partners reported by males than females, especially among the black population. 

While these factors may contribute to an explanation of the male-female difference in the reported number of sexual partners, a complete or satisfactory explanation of this difference is not available. 

In table 10, the median number of partners reported by men 15-44 years of age is 5.6. 

The median number of partners does not vary significantly by metropolitan residence, and there is no simple pattern by education. 

For men 25-44 years of age, the median number of female partners is 6.7; for women 25-44 years of age, the median is 3.8 male partners (table 11). 

Men who have served in the military report a larger number of partners than those who have not served: As with prison experience, this information is shown for men 25-44 years of age to allow for reporting of military service. 

Among men 25-44 years of age who have ever served in the military, the median number of female partners was 10.4, compared with 5.3 among those 25-44 years of age who have never served in the military. 

The median number of female partners among black men was 8.3 compared with 5.3 for white and 4.5 for Hispanic men. 

Significantly more black men had had 15 or more partners (34 percent) compared with white and Hispanic men (22 and 18 percent, respectively) (table 10 and figure 7). 

In table 11, the median number of male partners reported by women 25-44 years of age was 3.8. 

These medians do not vary markedly by characteristics of the woman, except by current marital status; similar to the data presented for males, formerly married (separated, divorced, and widowed) women report more partners in their lifetimes than others, in part because they are, on average, older than the other marital status groups.

sexual attraction as stated at the date of interview. 

These researchers have suggested that it is difficult to classify the U.S. population into two separate and distinct groups (heterosexual and homosexual), because each measure gives a somewhat different estimate of the number or percent of men and women in the United States who have sex with members of their own gender (13-18, 42). 

While data on sexual orientation and attraction provide important risk markers, the 2002 NSFG data on same-sex sexual behaviors can contribute to public health and prevention research concerning the risk factors for HIV and STI transmission. 

Tables 12 and 13 show the proportions of males and females in the 2002 NSFG who recorded, in the self-administered part of the interview, an answer to the following question: ‘‘Do you think of yourself as heterosexual, homosexual, bisexual, or something else?’’ The results are tabulated for those 18-44 years of age, in part to facilitate comparisons with other surveys that have only collected this information for those 18 and over. 

Among men, 90 percent identify themselves as heterosexual (figure 8). 

Another 2.3 percent identify themselves as homosexual, 1.8 percent as bisexual, 3.9 percent as something else, and 1.8 percent did not report their orientation (table 12). 

Among women as well, 90 percent identified themselves as heterosexual (table 13). 

Among men who have ever had sexual contact with another male, one-half (49 percent) consider themselves heterosexual (table 12). 

Among women who have had sexual contact with another woman, nearly two-thirds (65 percent) consider themselves heterosexual (table 13). 

It is noteworthy that 7.3 percent of Hispanic or Latino men, and 7.5 percent of black men, reported that their sexual orientation was ‘‘something else,’’ and another 3-4 percent of each group did not report an answer to the question.

Further study of those who answered something else and those who did not report an answer to the question is needed, in part to characterize their sexual and risk behaviors. 

The following tables provide some additional information on the ‘‘something else’’ and ‘‘did not report’’ categories. 

Tables 14 and 15 show another measure, also collected in the self- administered part of the interview, which was based on a question (quoted previously) about whether the respondent is ‘‘sexually attracted to’’ males, females, or both. 

Table 14 shows the results for males and table 15, the results for females. 

For males in table 14, 92 percent answered that they were attracted only to females, 3.9 percent mostly to females, 1.0 percent to both males and females, and another 2.2 percent only or mostly to males. 

The figures were generally similar, accounting for sampling error, for Hispanic, black, and white males. 

The results for females are somewhat different. 

About 86 percent of females in table 15 said they were attracted only to males, 10 percent of females answered that they were attracted mostly to males, 1.9 percent to both males and females, and 1.5 percent mostly or only to females. 

Looking at the results in table 14, 99 percent of males answering that they were heterosexual said they were only or mostly attracted to females; this reflects considerable correspondence between the measures of sexual orientation and attraction for this large group. 

Among the much smaller group of those reporting themselves as homosexual, 74 percent said they were attracted only or mostly attracted to males, but 23 percent said they were attracted only to females. 

This suggests that these measures could be measuring overlapping but somewhat different things, as Laumann and colleagues (16,17,42) suggested. 

Comparisons With Previous Surveys Tables 16 and 17 compare results from the 2002 NSFG to data from other previous national surveys that collected data on these topics. 

The intent of the comparisons is to establish whether the NSFG findings are consistent with previous national surveys. 

Comparisons with the other surveys are difficult, because the surveys were conducted in different years; their age ranges differ; their response rates differ; each is a sample and their estimates have sampling errors; and the wording and context of their questions on these topics were often similar but not identical. 

Therefore, differences among surveys do not always reflect trends or differences in behavior; they may reflect, at least in part, sampling error and differences in survey questions or procedures.

Unprotected anal sex is a common means of transmission of some diseases, including HIV. In the National Survey of Men (NSM), conducted in 1991 with 3,321 males 20-39 years of age, 20 percent of men reported that they had had anal sex with a female in the last 10 years (13). 

In the 1992 National Health and Social Life Survey (NHSLS), 26 percent of men 18-59 years of age reported ever having anal sex with a female (16) and in the 2002 NSFG, that figure was 38 percent. 

The higher proportion reporting heterosexual anal sex in the NSFG than in the NHSLS may be related to the fact that the NHSLS included cohorts (45 years and over in 1992) that were less likely to have had anal sex (16,17; p. 140). 

The difference in the percentages reporting anal sex between the 2002 NSFG and the 1991 NSM may be related to the 11-year difference between the surveys, or to the fact that the NSM estimate is based on a question covering ‘‘the last 10 years,’’ rather than the entire lifetime to date.

These surveys also obtained data on oral sex. 

Note that for men (table 16), 75 percent of men 20-39 years of age in the NSM reported that they had given oral sex to a female compared with 77 percent of men 18-59 years of age in the NHSLS, and 80 percent of men 18-44 years of age in the NSFG (table 16). 

The proportion of men who had ever received oral sex from a female was 79 percent in the NSM, 79 percent of men in the NHSLS, and 85 percent in the NSFG, so it appears that the three surveys gathered comparable data on oral sex, and no marked trends occurred among adult men in that time period. 

Data in table 16 also show that 73 percent of women in the NHSLS said they had received oral sex from a male compared with 83 percent in the NSFG, but the overall total in the NHSLS may be lower in part because those 50-59 years of age in 1992 were much less likely to have had oral sex than those 18-49 years of age in 1992 (17, p. 140). 

The data in table 16, from NSAM and NSFG, suggest that there was little or no change (accounting for sampling error) in the proportion of males 15-19 who had ever had heterosexual oral or anal sex between 1995 and 2002. 

Data on trends in these behaviors are not available for female teenagers. 

Same-sex contact Finally, these surveys also asked men whether they had ever had oral or anal sexual contact with another man. 

In the 1991 NSM, 2.3 percent reported having had sex with another man in the last 10 years. 

In the 1992 NHSLS, 4.9 percent of men reported same-gender sexual activity since age 18. 

In the 2002 NSFG, 6.2 percent of men 18-44 years of age reported ever having had oral or anal sex with another male up to this point in their lives. 

For women, 4.1 percent reported oral sex with another woman in the NHSLS (18-59 years of age in 1992). 

In the 2002 NSFG, 11.5 percent of women 18-44 years of age reported a "sexual experience" with another woman; although this is higher than in the other surveys, at least some of the difference may be because of differences in the wording of the questions on female same-sex sexual contact between the NSFG and the NHSLS. 

Table 17 shows the estimated proportions of men and women in several national surveys who reported sexual behavior with the same sex, or same-sex sexual orientation or attraction. 

Comparisons are not always direct for the reasons explained previously: the surveys were done in different years, they had different age ranges, their response rates were not identical, each survey has sampling error, and the wording and context of the questions was not always identical.

First, the proportions of males in these surveys who reported sexual activity with other men can be compared. 

In the NSFG, 6.0 percent of males 15-44 years of age in 2002 reported that they had had sex with another man at some time in their lives. 

In the NHSLS, this figure was 4.9 percent (similar considering the sampling errors of the two surveys). 

In the NSM, a somewhat different measure was used—same-sex contact in the last 10 years—which yields a lower figure (2.3 percent). 

In the 2002 NSFG, 2.9 percent of men reported having sex with another man in the last 12 months (confidence interval 2.3-3.5 percent), a figure similar to the 1.8 percent found by Rogers and Turner with the combined samples of four national surveys in 1970-90 (21). 

In the NHSLS in 1992, 97 percent of men reported that they were heterosexual, 2.0 percent called themselves homosexual, and about 1 percent bisexual or something else. 

The General Social Survey did not collect data on sexual orientation. 

In the 2002 NSFG, which was conducted 10 years later on a younger population (15-44 years of age rather than 18-59 years of age), 90 percent of males 18-44 years of age reported that they were heterosexual, 2.3 percent homosexual, 1.8 percent bisexual, and 3.9 percent something else; 1.8 percent did not answer the question (table 14). 

It appears that while the NSFG’s findings are generally similar to the other surveys, the NSFG has somewhat higher proportions reporting that their sexual orientation is ‘‘something else’’ (table 17).

NSFG data from 2002 indicate that 11 percent of women 15-44 years of age have ever had a sexual experience with another woman (6.9 million); 4.4 percent of women (2.7 million) have had a same-sex partner in the last 12 months, and just 1.3 percent (800,000) had only had female partners in the last 12 months (table 17 and table B). 

Looking at sexual orientation among women, based on the 2002 NSFG (table 13), about 90 percent answered heterosexual. 

The proportion of women who answered that they are homosexual was 1.3 percent, while 2.8 percent answered bisexual. 

The total of these two figures was 4.1 percent or about 2.3 million women 15-44 years of age. 

As with men, nearly 4 percent answered something else. 

The proportion of women who reported that they were attracted only or mostly to women was 1.5 percent; 1.9 percent of women said they were attracted to both sexes; and 10.2 percent of women said they were attracted ‘‘mostly’’ (but not only) to males (table 17). 

In each of the surveys described in table 17, the proportion who answered heterosexual is at least 90 percent; the proportion who answered homosexual is about 1-2 percent of men and 1 percent of women; and the proportion bisexual ranges from about 1-3 percent. 

Given the sampling errors of the surveys, these figures are fairly similar. 

The principal difference between the findings of the NSFG and the findings from other in-person surveys based on national samples is (a) the proportion giving an answer of ‘‘something else’’ is low in the surveys done in the early 1990’s, but about 4 percent in the NSFG; and (b) the proportion of women, particularly women under 30 years of age, who reported a sexual experience with another woman was higher in the NSFG than in the other surveys. 

Only 4.4 percent of women had had same-sex sexual activity in the last 12 months, compared with the 11 percent who had had such contact at some time in their lives. 

The proportion of men 18-44 years of age who have sex with men could be defined behaviorally as same-sex contact at least once in his life (3.7 million), or same-sex contact in the last 12 months (1.65 million), or exclusively same-sex contact in the last year or in a longer time period (1.0 million). 

Similarly, the prevalence of women 18-44 years of age who have had a sexual experience with another woman could be defined as having done so at least once in her life (6.9 million women), or having had sex with another woman in the last 12 months (2.7 million), or having had sex exclusively with other women in the last year (0.8 million) or in a longer time period. 

It is important to note again that these numbers do not include those 45 years of age and over, or those in prison or jails or other institutions, so they are not complete estimates of the total number of people in the United States who have these characteristics. 

Health behaviors and characteristics The next section of this report shows some behaviors and characteristics associated with sexual orientation and behavior. 

Table 18 shows whether men and women have been tested for Human Immunodeficiency Virus, or HIV, the virus that causes AIDS, in their lifetimes and in the last 12 months (excluding blood donations). 

Among men 15-44 years of age, 47 percent (with a 95 percent confidence interval of 44.2-49.0) reported that they had been tested for HIV at some time in their lives compared with 55 percent of women (95 percent confidence interval of 53.1-56.7 percent). 

This difference may be related to the fact that many women are tested for HIV in connection with pregnancy and prenatal care (31). 

About 15 percent of men and 17 percent of women 15-44 years of age were tested for HIV, outside of blood donation, in the last 12 months. 

Among men who have ever had sex with another man, 60 percent had ever been tested for HIV (in their lives) and 29 percent had been tested in the last year. 

Some have recommended annual testing for men who have sex with men (MSM) (44,45), and these data indicate that this ambitious goal has not yet been attained. 

Among men who had never had sex with another man, 46 percent had ever been tested, and 14 percent had been tested in the last year. 

The pattern for women is similar (figure 9). 

About 14 percent of men who consider themselves heterosexuals, 38 percent of men who consider themselves homosexuals, and 23 percent of bisexual men, were tested for HIV in the last 12 months. 

The proportion ever tested for HIV was 60 percent of bisexual and 67 percent of homosexual men, compared with 46 percent of heterosexual men. 

Since several of the groups shown in table 18 are small proportions of the population, statistics for them have relatively large sampling errors; statistics for these small groups should be interpreted with caution. 

Table 19 and figure 10 show the proportion of men and women 15-44 years of age who reported any sexually transmitted infection, or STI, other than HIV. Table 19 is limited to those who have ever had sexual contact with another person. 

Men were asked if they had ever been diagnosed with genital herpes, genital warts, or syphilis, and whether he had been treated for gonorrhea or chlamydia in the last 12 months. 

Women were asked about these diseases and also about pelvic inflammatory disease (PID), which is often caused by sexually transmitted organisms. 

It is known that some persons have an STI and do not know it, and therefore these data should be viewed as low estimates of the prevalence of these infections (31,46). 

But they are measures of the proportion of persons who are aware that they have had STIs and therefore may modify their behavior as a result of that awareness. 

About 17 percent of women reported having had an STI or PID (confidence interval 15.4-18.2), and 7 percent of men reported having had an STI (7.4 percent, confidence interval 6.2-8.6 percent). 

For men and women, the proportion reporting an STI is higher for men who report having sex with men than for men who have had sex only with women, as expected, given the higher risk of STI transmission for male same-sex activity. 

For example, 17 percent of men who have ever had sex with men (confidence interval = 13.1-21.5 percent) report having had an STI, significantly higher than for those who have not had same-sex sexual contact (6.6 percent, confidence interval = 5.4-7.8 percent). 

The comparisons are similar by sexual orientation and by same-sex contact in the last 12 months, but the confidence intervals of those statistics are larger (table 19 and figure 10). 

Further study of these data is warranted and additional data may be presented in forthcoming reports based on the 2002 NSFG. 

Table 20 shows the percent distribution of men and women by whether they have been tested for an STI, for HIV, or both. 

For men, about 80 percent had not been tested for either HIV or STI in the 12 months before the survey; 5.9 percent had been tested for HIV only, 7.3 percent for STI only, and 7.1 percent for both. 

Nearly one-half (48 percent) of men who identified themselves as homosexual in the self-administered part of the interview had been tested in the past year, along with 30 percent of those reporting that they are bisexual. 

Nearly 40 percent of men who had ever had sexual contact with another male, reported being tested for HIV, an STI, or both during the last 12 months. 

Among women (table 20), testing was more common than for men: 70 percent had not been tested for either HIV or STI, while 7.5 percent had been tested for HIV only, 13 percent for STI only, and 10 percent for both STI and HIV. Research suggests that some of the male-female difference in testing is associated with medical visits for contraceptive use and prenatal care (30,35). 

Table 21 shows data for men and women on the proportion who used a condom at their most recent sex (excluding same-sex sexual contact among females). 

Overall, 40 percent of men reported using a condom at their most recent sex. 

The proportion using a condom was higher for groups that may be at higher risk of STIs: 66 percent of teenagers, 65 percent of never married noncohabiting men, 88 percent of homosexual men, 76 percent of bisexual men, and 91 percent of men who have ever had a same-sex partner. 

Fewer white men (35 percent) reported using a condom at their most recent sexual intercourse than Hispanic or black men (46 and 56 percent, respectively). 

About half as many women compared with men, reported that a condom was used at their most recent sex—22 percent of women reported their male partner used a condom and, as noted previously, 40 percent of men reported they had used a condom. 

For both sexes, condom use is highest among teens and lowest among those 30-44 years of age. 

Females 15-19 years of age were more likely than women of other ages to have a partner who used a condom at their most recent sexual intercourse (44 percent). 

Black women (32 percent) were more likely than Hispanic (23 percent) or white (20 percent) women to have a partner who used a condom. 

Table 22 shows data on men and women by the type of health insurance coverage they have, if any. 

This information is presented for subgroups reporting different sexual orientation, sexual behavior, and STI history. 

Overall in 2002, 20 percent of men 15-44 years of age did not have any health insurance coverage at all compared with 16 percent of women 15-44 years of age. 

Much of this difference is accounted for by the higher proportion of women (14 percent) than men (7.4 percent) who were covered by Medicaid, Medi-Gap, CHIP, or other government health coverage. 

While 19 percent of heterosexual men had no health insurance coverage, 27 percent of homosexual men and 35 percent of those who reported their sexual orientation as something else had no coverage (table 22). 

In addition, heterosexual men were less likely than other men (6.6 percent compared with 11- 17 percent) to rely on Medicaid or other government-sponsored, means-tested programs.

This report was intended to provide national estimates of some basic statistics related to certain types of sexual behavior, orientation, and attraction for men and women 15-44 years of age in the household population of the United States, in order to (a) measure the populations at risk of STIs, and (b) to provide insights about factors related to birth and pregnancy rates, especially among those 15-24 years of age, based on the most recent and most reliable national statistics currently available. 

This report was not intended to be a complete or exhaustive treatment of these very complex and sensitive subjects. 

Specifically, the data described here are national, not State, estimates that are limited to the household population of the United States, ages 15-44. 

The data presented here do not cover persons who are homeless, currently incarcerated, currently living on military bases or in U.S. territories, or persons under age 15 or above age 44.

This report is intended to provide reliable national estimates of some basic statistics on certain types of sexual behavior, sexual orientation, and sexual attraction based on data collected in the United States in 2002. 

These statistics help to estimate the size and characteristics of populations at elevated risk of sexually transmitted infections, including Human Immunodeficiency Virus or HIV, the virus that causes AIDS. 

This report is not an exhaustive analysis of either sexual behavior or of the risk of sexually transmitted infections and is limited to men and women 15-44 years of age in the household population of the United States in 2002. 

The data used for this report are from the 2002 National Survey of Family Growth (NSFG). 

The survey was based on in-person interviews conducted between March 2002 and March 2003 in the homes of 12,571 people (4,928 men and 7,643 women) 15-44 years of age in the household population of the United States. 

The response rate for the survey was 79 percent. 

Most of the demographic characteristics used in this report were collected by an interviewer, but the measures of sexual behavior reported here were collected in the self- administered portion of the 2002 NSFG survey in which respondents read the questions (or heard them through headphones) and entered their own answers into a laptop computer, thus ensuring their privacy.