Sexually Transmitted Diseases  >  Program Guidelines  >  Partner Services

Program Operations Guidelines for STD Prevention
Partner Services

Previous Page

Table of Contents

Next Page

Sections on this page:

• COMMUNITY-BASED OUTREACH
• Overview of Interventions
• Social Network Analysis
• Targeted Screening and Field Testing
• Community Outreach
• STD Clinic Outreach

COMMUNITY-BASED OUTREACH

Public health and STD prevention programs, in particular, have a duty to warn individuals that they may have been exposed to a sexually transmitted disease. In response, most STD prevention programs provide for the notification and evaluation of exposed partners who have been identified by an infected index case or partner. Examples of such services include partner notification (PN), clinical evaluation and testing of partners, the concurrent provision of prophylaxis, and risk-reduction counseling. Some have stated that this duty to warn extends also to individuals who were exposed but who could not be located through PN (Peterman, 1997). However, not all STD prevention programs directly provide for the evaluation of persons who have not been located through PN or who have not been identified by an infected index case or partner. Examples of strategies that address this expanded charge include clinical evaluation and testing of patients who come to the clinic as volunteers, cluster interviewing with resultant disease screening and prophylaxis, review of epidemiologic data collected through ethnographic means, targeted outreach, screening, and public awareness campaigns. It is important for STD prevention programs to evaluate their local situations and to employ interventions which complement PN. Such interventions include social network analysis in conjunction with PN, targeted screening and field testing, and other forms of outreach..

Overview of Interventions

Social Network Analysis

Network analysis is defined as the study of how people connect in social structures and of its implications (Potterat, 1998). A detailed discussion of social network analysis is available in Appendix K. Several different social networking methods are commonly used. One method is to collect information about core environments in addition to partner names. Another is to make a programmatic commitment to investigate the networks where disease is located rather than investigating only individuals known to have a STD. Clustering, the technique by which infected and uninfected patients are interviewed about their associates (as well as their partners), may provide extra information about the identity and location of sought-after sex partners. Clustering also can be used to identify geographic areas or for narrowing criteria for targeted screening. Additional methods of social networking are spot-mapping home addresses and hangouts, studying partner mixing patterns, and performing old-fashioned shoe-leather epidemiology.

Research often centers around which behavioral and social features are necessary to continue disease transmission in epidemic numbers and around the exploration of which interventions could be implemented to halt transmission. Transmission dynamics are primarily dependent upon the effects of small populations with varying levels of sexual activity (i.e., frequency, number of different sex partners) (Oxman, 1996a). For example, in Oregon the number of syphilis infections was found to be affected by the number of clusters of women who have a large number of casual or anonymous partners. Oxman found that when the actual number of women in the group or the number of partners exchanged in any given time period, or both, was reduced, the rate of the group's infection decreased. When an epidemic begins, the number of infected people rises quickly to a peak that appears to be closely linked to the sexual behavior characteristics of the involved population.

There is support for the notion that syphilis outbreaks in heterosexuals, which are extremely difficult to control once underway, are a result of core transmission (a small number of interactive and networked individuals). STD control programs can incorporate social network methods, e.g., mapping, cluster interviewing, to identify the populations and conditions within its jurisdiction that facilitate disease transmission, especially by high-frequency (core) transmitters. In addition, programs may be able to prevent outbreaks by limiting disease occurrence in core transmitters. This should be done in partnership with communities that the program serves (Oxman, 1996a).

Woodhouse et al. researched how a group's social, sexual, and injection drug-sharing relationships might help or hinder the spread of various STD, including HIV (Woodhouse, 1994). What they found, surprisingly, was that the majority of the infected individuals were not part of the larger interconnected group engaging in high-risk activity, but in fact were connected to much smaller groups with no links to the larger group. In other words, it is not just the presence of infection that produces transmission, but other social factors such as group dynamics and behaviors, group size, and geographic area. Programs can take the information gathered from network analysis and review and can create and implement policy and operations that take into consideration the dynamics of day-to-day transmission within their jurisdictions (Woodhouse, 1994).

Rothenberg and Narramore outlined how social networking analysis was used by public health officials to address an increase in early syphilis cases in certain areas of Nashville, Tennessee (Rothenberg, 1996b). A map was created of all the addresses reported by individuals with early syphilis. As a result, staff were able to identify that 89.7% of all persons with early syphilis lived within nine well-defined geographic areas. This pattern had actually been occurring for several years. In addition, staff observed the use of crack cocaine in this network. In response, the Nashville STD control program implemented a network-informed approach to their syphilis prevention activities. Such efforts included the assignment of public health workers to specific geographic areas, staff having continuing contact with persons at risk and other community leaders, and mandatory reinterviewing of all infected persons to gather additional information on personal networks.

A recent report described the importance of social network and ethnographic tools in the investigation of a cluster of syphilis cases in Georgia (Rothenberg, 1998). Several complementary methods were used, including the interviewing of as many people as possible who were believed to be involved in transmission (both infected and uninfected people); the detailed ethnographic exploration and documentation of sexual and social patterns; the collection of interview information on standard CDC interview forms; the conversion of interview data into databases to which network analysis software could be applied (including programs that allowed for graphic representations of patterns); and follow-up interviews several months to a year later to examine the social and sexual patterns that followed the outbreak. Ethnographic interviews revealed the existence of a complex sexual picture that predated the diagnosis of the first case by one year and that people without infection were often as central in the network and as important in transmission as infected people. In addition, uninfected people were as likely to identify partners with infection as people without infection. This approach underscores that if programs interview only people known to have infection, they will miss important people, including infected partners and individuals who do not have a STD, but who, by their connectedness within a network, sustain transmission. With appropriate training in ethnographic and social network methods and the use of databases such as STD*MIS, a network informed approach can be incorporated into STD prevention program activities.

An example of such incorporation has been attempted in an inner city area of Atlanta, Ga. with high syphilis rates. A DIS team, spending approximately 80% of its time in the field (compared to interviewing infected persons in the clinic and then seeking the partners), used network and ethnographic methods to identify an interconnected group of over 300 persons with a six month syphilis incidence of 12.6% (Rothenberg, in press). By identifying such groups at risk, the field team is in a position not only to interrupt disease transmission but to predict and respond to changing disease trends. These approaches provide direct observations of behavior change in a community (e.g., adoption of condom use, limiting numbers of anonymous partners, decreasing the frequency of sex and drug partner change) (Rothenberg, 1995), and provide a built-in mechanism for appropriate targeting.

Traditionally, public health has focused on specific behaviors or on some overall assessment of risk, which has often resulted in a broad characterization of various social groups, i.e., gay men and teenagers. Research has shown that specific behaviors determine the risk for infection and that social networks determine the extent of the disease within that given population. Over the years, partner notification has shown that social networks do play an important role in the public health approach to disease control. Initial work suggests that social structures can influence STD transmission. Social structures also can increase the effect of risky behaviors within each social setting. Epidemics do not result just from many risky acts, but are the result of complex interactions embedded in a social and geographic context (Rothenberg, 1996a). It is crucial that programs take into account both the risky behaviors of individuals and the risky behaviors ingrained within the culture of the social network when critiquing, revising, or developing disease control interventions.

History has shown that the act of segmenting social networks, such as closing bathhouses and shooting galleries, or housing disruption in economically impoverished areas, may result in higher rates and widespread disease for a period of time (Rothenberg, 1996a). Disease that was once self-contained in a small segregated community can expand beyond its previous boundaries and, as a result, create new possibilities for disease transmission. Instead of dividing social networks, programs can use social networking methods to identify those individuals who hold influence and who can potentially act in partnerships with health professionals in disease prevention.

A more formal approach to social network analysis has been shown to be very effective in reducing the incidence of disease transmission by targeting specific areas (Rothenberg, 1996b). However, it can be a very labor-intensive process and is recommended only if program staff are familiar with the techniques of data collection and evaluation and have the resources to process the information gathered. First, programs can expand the scope of partners to include close friends, acquaintances, persons within the same social group, roommates, former or occasional sex partners, and anyone else deemed at risk. Local protocol should dictate the exact criteria. Second, people identified would be (cluster) interviewed to determine the appropriateness of prophylaxis, to pursue further partners and associates, to identify what other social groups may be involved, to determine the behaviors associated with the groups, and to gauge the strength of associations within the social network. Care must be taken to assure that there is no violation of confidentiality nor the perception of violation. Subsequently, programs can document what they have learned about individual communities, with a focus on the mixing patterns, frequency of partner change, and social hierarchy. Once these elements are understood and discussed by program staff, it will be easier to tailor and implement disease control methods toward the specific dynamics of disease transmission within the social network.

Social network analysis, in essence, means reducing the emphasis on individuals and looking at the commonalities among individuals with a STD and their associates. Experts believe that increased focus on STD transmission analysis or intervention should be placed on the social network rather than solely on the individual. It is widely thought that disease control methods targeted to the general population may be less valuable than approaches that focus resources on important group structures (Rothenberg, 1996a). Researchers add that since some social network analysis in the infectious diseases context may fall short in the area of sampling strategies and data collection, results should be used to stimulate further research in this area (Potterat, 1998). As a result, social network analysis can be seen as complementary to other models of infectious disease prevention.

Targeted Screening and Field Testing

Targeted screening can be defined as an activity to identify people with infection in a select group who are engaged in a behavior that puts them at greater risk for infection. Field testing is when public health workers offer testing at non-clinic locations associated with known cases and their partners.

As an example, the prevalence of Chlamydia trachomatis infection in inner-city youth was measured by collecting 486 urine specimens during a 20 month period (Rietmeijer, 1997). Specimens were collected both in the field and in clinic settings. The study found that positivity rates were higher in the field than in the clinic facilities (11.9% vs. 4.4%). Ninety-seven percent of all infected patients were treated within eight days of testing. Thus, screening can be done in nontraditional settings and still yield similar, if not better, results than screening done in standard clinic settings. Considering the substantial numbers of asymptomatic chlamydia infections in field-recruited male youths, the large number of recent sex partners, and a reluctance to seek clinic-based STD screening, it is doubtful whether, even with optimal access to STD treatment services, traditional clinic-based approaches will ever bring the chlamydia epidemic under control (Rietmeijer, 1997). In this context, the use of non-invasive screening methods embedded in targeted, community-level prevention programs has the potential to make significant contributions to STD control.

Disease control efforts also have used targeted screening to find otherwise unseen or undiagnosed disease. It continues to be a very effective way to locate a high percentage of new cases (Gerber, 1989). When traditional means of disease control fall short, clustering others within the same social network of the infected patient and offering them testing can be extremely effective. In this setting, screening close social associates of infected patients is almost as effective as screening actual partners.

During the first half of 1990, traditional approaches to the control of syphilis were found to be ineffective in slowing a syphilis epidemic (Mellinger, 1991). Persons who were involved in the exchange of drugs or money for sex often could not or would not provide sufficient information about their sex partners. That prevented public health personnel from locating exposed partners. As a result, alternative case-finding methods were needed. Disease transmission was reduced by using cluster interviewing to identify friends and associates at risk for syphilis and by setting up targeted serologic screening for those identified and for others engaging in high-risk sexual activity. This process documented a 27% reactivity rate, with 3% of those newly infected diagnosed with either primary or secondary syphilis.

Similarly, staff involved in a different syphilis epidemic began to focus on identifying places associated with cases and partners, instead of just on partner names (Hutcheson, 1993). They discovered 21 places where affected people were most likely to meet sex partners. Subsequently, staff members familiar with the community visited the sites and took over 200 blood samples that were tested for syphilis. Thirty-one percent tested positive, and 17% were preventively treated. Of those testing positive, 78% received examination and treatment, and a majority were found to have additional STD. It is important to note that in this case a combination of innovative, conventional, and cluster interviewing and investigation methods were used to effectively identify previously undiagnosed syphilis cases.

Increased use of crack cocaine and the exchange of sex for money or drugs have been major contributors to the increased occurrence of syphilis in many areas throughout the country, affecting disproportionate numbers of people of color (Greenberg, 1992). Traditional syphilis control programs usually offer a combination of interventions, including serologic screening of asymptomatic individuals, diagnostic testing of individuals self- otivated by symptoms or by perceived risk, and DIS case management (Oxman, 1996b). To increase the effects on disease transmission, many programs have instituted targeted syphilis screening in areas connected with cases and their associates. In recent years, these targeted screenings have been provided for sex workers and their customers, and for the drug (crack cocaine) dealers and crack users. Since those groups tend not to use traditional health care, screening should be offered in non-clinic settings such as crack houses, bars, shelters, parks, jails, detention centers, back alleys, and other locations frequented by at-risk populations.

Traditional control of gonorrhea and chlamydia has often been clinic based and relied on the treatment of self-referred, mostly symptomatic patients in combination with the notification and treatment of their partners. However, delays between diagnosis and treatment are not uncommon and often result in recurrences of transmission and reinfection. To prevent the unintentional transmission of disease, suggestions have been made to expand targeted testing in non- clinic settings. Communities that have traditionally avoided clinical care will be more likely to seek care in non-traditional settings if given the opportunity. Field screening will become more practical with the increasing availability of convenient techniques for the detection of STD, such as urine testing.

While no one in the field of disease control is debating the usefulness of screening to identify undiagnosed disease, the common thought is that screening needs to target the highest prevalence areas to interrupt core transmission and, in turn, reduce disease rates. In general, STD prevention programs need to balance screening and DIS activity so that testing and field activity are complementary.

There is a strong movement to combine both field and clinic screening efforts. Each acts as a bridge to the other. Field screening results in patients accessing clinical care, and clinical care plays a very important part in disease control. Without traditional clinic screening, communities risk missing cases of other STDs, and reducing opportunities for such related prevention activities as pregnancy testing, Pap smear screening, risk-reduction counseling, HIV testing, hepatitis B vaccination, and the initiation of contraception.

Community Outreach

An effective strategy in reducing disease transmission is for DIS or other health professionals to develop relationships with the social and sexual leaders (core transmitters) within any given population. This requires that DIS build partnerships with people affected by STD. However, first it is necessary to establish an effective line of communication between those who analyze data and the field staff, so that programs (and particularly DIS) can identify the core transmitters within their areas, i.e., develop a picture of the sociosexual networks and transmission dynamics (Potterat, 1992). Once trust is established between the community and DIS, it may be much easier to locate partners and associates, set up effective targeted screening, provide risk-reduction counseling, and perform cluster interviews.

STD Clinic Outreach

Some practical approaches that STD prevention programs can use to help control STD, especially in populations who trade sex for money or drugs include: locating clinics close to high-incidence areas, adding evening hours, reducing waiting time, encouraging community participation in targeting behaviors to be changed, and immediately following up with infected patients (Dunn, 1991) Presumptive treatment of close associates and cluster suspects can be more effective than partner notification in controlling transmission of syphilis, especially in crack users, and the cost of this treatment may be negated by the cost savings of the cases prevented. Others have suggested that clinics consider gang boundaries and their effects when planning and implementing services.

Previous Page

Table of Contents

Next Page