Key points are not available for this paper at this time.
HISTORICALLY, THE PREDOMINANT focus of sexually transmitted disease (STD) epidemiology has been on the attributes and behaviors of individuals, and on the risk of acquiring, rather than transmitting, infection. This approach is consistent with the approaches of clinical medicine, chronic disease epidemiology, and psychology. However, when considered as the "sole" or "main" focus, it appears to be inconsistent with STD transmission dynamics,1 and it has been increasingly challenged over the past decade.2-5 Recent theoretical and empirical work emphasizes the implications of the infectious nature of STD on their epidemiology6-9 and suggests a number of new principles and parameters to guide the social (behavioral) epidemiology of STD in the near future. This set of principles and parameters may define a new paradigm for STD epidemiology, one that may become predominant in the new millennium. In this issue of Sexually Transmitted Diseases, Laumann and Youm present the results of their analyses of sexual behavior and sexual network data from the National Health and Social Life Survey (NHSLS), based on a nationally representative probability sample of men and women in the United States. The scientific approach, the characteristics of the data, and the findings of this study are remarkably in line with those of the new paradigm, and Laumann and Youm's findings are sufficiently important that the publication of this article may constitute a bench mark highlighting the paradigm shift mentioned above. Most interesting conclusions reached by the authors include the critical role social network patterns play in accounting for the known differentials in rates of infection across racial/ethnic groups. They found that the higher rates of sexual contact between the "core" group and the "periphery" among African Americans facilitates the spread of infection overflow into the African American population; whereas the "sexual segregation" of African Americans from other racial/ethnic groups results in sexually transmitted infections remaining inside this population.10 Emerging Principles The new paradigm appears to include at least four principles. First, in STD epidemiology, as in the epidemiology of other infectious diseases, one person's health outcome is highly dependent on other persons' health outcomes. Unless one's sex partner(s) is infected with a particular sexually transmitted pathogen, one cannot acquire that infection.1 The epidemiology of chronic diseases is different; in the case of chronic diseases, one's health outcome is independent of the health outcomes of others, even though one's risk exposure behaviors may be influenced by risk exposure behaviors of others, through social diffusion processes. Second, as a focus of research and interventions, transmission of infection and its prevention is as important and perhaps more important than acquisition of infection and its prevention. This approach focuses attention on infected individuals, and the role they play in the spread of infection, and differentiates between "source" and "spread" cases.8,9 Closely related to the emphasis on transmission of infection is the third principle, the consideration that characteristics of sex partners and partner selection processes are an important component of risk determination. Thus, attention shifts from focusing solely on risk behaviors of individuals to include behaviors of sex partners. Finally, a shift from data based solely on cases and their characteristics to one based on persons, including both those who are infected and those who are uninfected, is called for in the new paradigm. The analysis by Laumann and Youm10 constitutes an example of this approach. Emerging Parameters The predominant paradigm focused on characteristics and behaviors of individuals, including demographic and ecologic characteristics such as age, gender, race/ethnicity, zip code and, where available, social class. Recent work suggests new, population level parameters may be important determinants of STD rates. These may include: 1) Closed/open population: the extent to which a population or subpopulation may be open or closed, i.e., members of the population may have sexual connections with persons who are members of other populations (assortative/disassortative mixing). 2) Population composition and relevant mixing patterns: some populations have a young age composition with more people in younger age groups compared with older age groups, combined with age-mixing patterns that condone older individuals engaging in sexual interaction with younger individuals, whereas other populations have an old age composition with a larger proportion of the population in older age groups combined with age-mixing patterns that reinforce sexual interaction between same-age partners. The spread of sexually transmitted infections between different age groups is facilitated by the former combination of population composition and sexual mixing pattern, whereas the latter combination would limit such spread.11 Similar combinations of population composition and sexual mixing pattern with respect to race/ethnicity, social class, culture, etc., would have similar effects on the spread of sexually transmitted infections. 3) Concurrency: the extent to which individual members of a population have concurrent sexual partnerships12; the greater the proportion of individuals in a population who have overlapping durations of sexual interaction with their partners, the greater is the potential for the spread of sexually transmitted infections. 4) Absolute and relative size of "core" group: although data from a wide range of populations are not available, it seems clear that both the absolute and the relative (relative to the size of the general population) size of the core group, where core group is defined to include those individuals that have large numbers of sexual contacts so as to have the potential to infect large numbers of individuals if they themselves get infected, are important determinants of the spread of infection in the population. 5) Average level of risk engaged in by the "core" group: although the core group is defined so as to have riskier behaviors-i.e., greater numbers of partners compared with the general population-members of a core group may, on the average, have lower or higher numbers of partners. The extent to which members of a core group have, on the average, larger (rather than smaller) numbers of partners is an indicator of the rate of spread of sexually transmitted infections, at least in that core group. 6) Sexual interaction between the core group and the general population: the extent to which members of the core group have, as their sex partners, individuals who themselves are not members of the core group. This parameter is an indicator of the potential rate of spread of sexually transmitted infections beyond the core group into the general population. 7) Sexual interaction between subpopulations, particularly between core groups of subpopulations: the extent to which there are individuals who function as bridges between subpopulations is an indicator of the potential spread of sexually transmitted infections from one subpopulation to another.13 When the bridges are to the core group in the receiving subpopulation, the probability of spread in that subpopulation is greater.10 These parameters are more or less variable over time and across societies; mixing patterns, population composition by age, race/ethnicity, social class, culture, and sexual behaviors change over time. These changes are reflected in changing STD rates. Rates of STD marked by short infectious periods are influenced by current mixing patterns and respond to change rapidly, whereas rates of STD marked by long infectious periods are influenced by cumulative lifetime mixing patterns and respond to change only in the long run. Moreover, in considering the epidemiology of curable bacterial STDs, it is important to consider differences in health seeking behaviors and access to care across population groups. Implications for Interventions and Intervention Research The conclusions reached by Laumann and Youm and the new social structural paradigm point to the importance of structural interventions for the prevention of STD. They also provide additional insights into the mechanisms of action of structural factors as they influence STD rates. For example, analyses of the NHSLS data suggest that the "sexual segregation" of African Americans may contribute to the higher STD rates observed in this population. These analyses also show that there is more sexual contact between members of the core group and members of the periphery in the African American population compared with the white population. The remarkably low gender ratio in this population, which results in the scarcity of available young men,3 may play an important role in explaining the higher STD rates among African Americans. Lack of young men decreases women's ability to be discriminating in partner choice, to negotiate safer sex practices, and results in exposure to men who are more likely to be in the "core" or to have had sexual interaction with women in the "core." Increasing understanding of structural interventions is encouraging at a time when accumulating evidence points to reversals in the trend toward safer sexual practices among some high risk subgroups in the United States.14,15 Interventions aimed at changing individual behaviors need to be repeatedly implemented to prevent relapse after success in accomplishing behavior change and to ensure that successive younger cohorts are covered by the risk reduction interventions as they enter the sexually active age groups. Structural interventions may be easier to sustain. Moreover, a societal approach may facilitate a better understanding of the mechanisms of action that relate factors like poverty, inequality, political upheaval, and minority status to increased STD rates.
Sevgi O. Aral (Sat,) studied this question.