PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Am J Health Promot. Author manuscript; available in PMC 2010 May 17.
Published in final edited form as:
PMCID: PMC2871324
NIHMSID: NIHMS124280

Engaging Community Businesses in HIV Prevention: A Feasibility Study

Abstract

Purpose

To explore the feasibility of engaging community businesses in HIV prevention.

Design

Randomly selected business owners/managers were asked to display discreetly wrapped condoms and brochures provided free-of-charge for 3 months. Assessments were conducted at baseline, mid-, and post-program. Customer feedback was obtained through an online survey.

Setting

San Diego, California neighborhood with a high rate of AIDS.

Subjects

Fifty-one business owners/managers representing 10 retail categories, and 52 customers.

Measures

Participation rates, descriptive characteristics, number of condoms and brochures distributed, customer feedback, business owners'/managers' program satisfaction and willingness to provide future support for HIV prevention.

Analysis

Kruskal-Wallis, Mann-Whitney U, Fisher's exact, and McNemar's tests were used to analyze data.

Results

The 20 business owners/managers (39%) who agreed to distribute condoms and brochures reported fewer years in business and more employees than those who agreed only to distribute brochures (20%) or refused to participate (41%), p <.05. Bars were the easiest of ten retail categories to recruit. Businesses with more employees and customers distributed more condoms and brochures, p < .05. More than 90% of customers supported distributing condoms and brochures in businesses and 96% of business owners/managers described their program experience as “positive.”

Conclusion

Businesses are willing to distribute condoms and brochures to prevent HIV. Policies to increase business participation in HIV prevention should be developed and tested.

Keywords: Condoms, commerce, social marketing, social environment, environment design, environmental policy

Purpose

Approximately 56,300 people are newly infected with HIV each year in the US.1 Annual costs for HIV-related medications and outpatient and hospital services range from about $10,000 to $50,000.2 The economic burden of HIV and the limited reach of existing HIV prevention programs has increased recognition of the need to build environments supporting safer sex.

The Behavioral Ecological Model (BEM)3 provides a framework for building environments to support safer sex. According to the BEM, increases in physical and social cues prompting safer sex increase its likelihood. Physical cues include displays of condoms or AIDS education materials, while social cues include peers modeling condom-carrying and discussing safer sex. Safer sex cues interact within, and between ecological levels of society; e.g., condom distribution in public settings may prompt safer sex discussions and increase the likelihood that people will be reinforced for engaging in safer sex, leading to safer sex norms.

Community businesses hold promise for increasing public exposure to repeated safer sex cues. The average US adult visits community businesses at least three times per week,4 compared to three times per year for physician offices.5 Furthermore, the Centers for Disease Control and Prevention6,7 recommend involving more levels of society, including businesses, in HIV prevention. However, only a few published U.S. studies report HIV prevention programs in businesses.e.g.,8-11 None of these studies evaluated the feasibility of distributing discreetly wrapped condoms and brochures in diverse business types. Feasibility studies are an important starting point for policy making, and for refining recruitment and intervention procedures prior to larger experimental trials.12,13 Feasibility studies assess an intervention's reach, acceptability, costs, and political and economic influences on intervention sustainability.12,13 The current study explored the feasibility of engaging community businesses in HIV prevention.

Methods

Design and Sample

Hillcrest, San Diego (92103 zip code: population = 31,744, median age = 40.2, 51% male, 70% Caucasian, 19% Hispanic, median household income = $61,250)14 was selected as the study's setting because it had the county's highest AIDS rate (County of San Diego Health and Human Services, personal communication, March 2004) and diverse business types. It is known as a gay friendly community, but with numerous non-gay residents and customers. A database was developed of all businesses within Hillcrest's government-defined boundaries (n = 329) meeting the following criteria: street-level, privately owned, not requiring an appointment, not providing health services, and not selling condoms. Stratified random sampling was used to recruit equal numbers of businesses from each of the following ten retail categories: bars, beauty salons, books/music/videos, clothing, clothing care/repair, general merchandise, general services, home furnishings, ice cream/treats, restaurants. Businesses were recruited sequentially (August-November, 2006) based on computer-generated random number assignments. The lowest number in each category was targeted for initial recruitment. If the business owner/manager could not be interviewed face-to-face after six in-person (walk-in) attempts, the next listed business was targeted.

At baseline, we measured businesses' willingness to distribute condoms/brochures, reasons for refusing participation, and descriptive characteristics. During the program, we assessed the feasibility of obtaining brief customer feedback via a website, number of condoms/brochures distributed, and business and research staff suggestions for program improvement. At post-program, we measured business owners'/managers' program satisfaction, condom packaging preferences, perceptions of customer and employee reactions, and political and economic support for AIDS prevention. Because no similar feasibility assessments have been conducted in businesses, assessments were designed based on feasibility study guidelines,12,13 and the BEM,3 which recommends measuring contingencies (e.g., customer reactions, costs) likely to influence business participation.a We named the program “Business Environments Against the Transmission of AIDS” (BEAT AIDS). We used discreetly wrapped condoms and HIV-testing brochures based on pilot work indicating the importance of discreet packaging and offering businesses a range of options for contributing to HIV prevention.

Recruitment—business owners/managers

Business owners/managers agreeing to a baseline interview were asked: “If we give you condoms like these free of charge, would you be willing to display them?” We presented a clear plastic container 6”×9 ¾”×8 ¾” containing colorful miniature envelopes. Each envelope was covered with a cute slogan (e.g., “I'm protective, handy, and flexible”), and included a brand-name condom, 2 chocolate mints, and a card with condom-use instructions and a web-link for a survey/raffle. The condom container did not identify the envelopes' contents, but included a label stating “Free!; Please Take One,” along with the project name and logo. Next, we presented colorful brochures listing free local HIV-testing sites, and asked: “If we give you brochures like these free of charge, would you be willing to display them?” Each brochure had a cute slogan (e.g., “I'll show you mine if… [outside cover], you show me your test [inside cover].”)

Business owners/managers who refused to display the condoms and/or brochures, were asked: “Would you be willing to keep the condoms/brochures behind the counter and/or hand them out or put them in a shopping bag when adults make a purchase?” Participating businesses were immediately given the materials, informed that they could display them in any location in their businesses except restrooms, and visited at least once weekly for 3 months to count and restock condoms and brochures. All businesses interviewed received an explanation of study procedures and informed consent forms, approved by San Diego State University's Institutional Review Board (IRB).

Recruitment—customers

Within each condom pack left with businesses, a card invited customers to visit a website to “enter a raffle” and “complete a 3-minute survey to let us know what you think.” Customers choosing to visit the website (www.surveymonkey.com) were presented with study procedures and gave their informed consent to participate by clicking the “next” button to proceed to the survey, as approved by the IRB. The survey was advertised only in condom packs at participating businesses.

Measures

Descriptive and demographic characteristics of businesses

Age, gender, race, education, years in business, number of employees, and proportion of gay/lesbian customers were measured by business owner/manager report for all businesses interviewed at baseline, including those not participating in the BEAT AIDS program. To further characterize businesses that agreed to participate, at post-program these business owners/managers reported the most common gender, age range, and ethnicity of their customers, and number of walk-in customers per day.

Online survey: Customers' descriptive characteristics and feedback

Customers' age, gender, race, education, sexual orientation, zip code of home residence, and source of obtaining condom pack(s) were assessed. In addition, customers rated on a 5-point Likert scale if they thought it was a “good idea” for different types of Hillcrest businesses to hand out free condoms and HIV-testing brochures. Customers also provided comments regarding their thoughts about providing customers with free condoms and brochures. The online survey was accessible throughout the program. It was developed based on guidelines for feasibility studies that recommend assessing the acceptability of program materials, and consumer characteristics.12,13

Business owner/manager and research staff feedback

During recruitment, we kept a record of reasons given by business owners/managers for refusing to participate in the BEAT AIDS program. During the program, we conducted a mid-program open-ended interview to assess satisfaction of participating businesses and ways to improve the program. Research staff also recorded variables that affected implementation fidelity and quality based on observations at weekly business visits.

Number of condoms and brochures distributed

Condoms and brochures were counted by project staff during once weekly business visits for three months using standard procedures.

Business satisfaction

At post-program, owners/managers rated the item “Overall, I thought my experience in the BEAT AIDS program was…” on a 5-point Likert scale. Agreement-disagreement was also measured with: the BEAT AIDS program “did not take a lot of time or effort;” and “made me feel I was making a positive contribution to the community.”

Businesses' perceptions of condom packaging

At post-program, owners/managers distributing condoms rated their agreement with the items, “I would be willing to continue to display condoms if they are wrapped in envelopes or other discreet packaging”, and “I would be willing to continue to display condoms even if they are not wrapped in envelopes or other discreet packaging.”

Businesses' perceptions of customer and employee reactions

At post-program, owners/managers rated their agreement with the following: “The BEAT AIDS program”… “helped attract new customers;” “helped me keep existing customers;” “ resulted in positive comments from customers;” and “resulted in positive comments from employees.”

Businesses' willingness to provide political support for HIV prevention

At post-program, owners/managers rated their agreement with the following: “I would like to have my participation in this program publicized in a positive way in print, TV, and radio media; I would be willing to put a decal in the window of my business showing I am participating in the BEAT AIDS program; I would be interested in meeting with other Hillcrest business owners several times a year to discuss ways to prevent AIDS in this community; I would be willing to sign a letter recommending that other Hillcrest businesses participate in the BEAT AIDS program; and I would be willing to sign a letter addressed to a member of Congress asking for tax breaks for businesses that hand out condoms or AIDS prevention brochures.”

Businesses' willingness to provide economic support for HIV prevention

At post-program, owners/managers rated the following items as “Yes” or “No.”

  1. A 1-year supply of condoms/AIDS prevention brochures for your business would cost approximately ____ dollars. Would you be willing to pay this amount to continue receiving condoms/AIDS prevention brochures for 1 year?
  2. If “no” to the question above, would you be willing to pay for part of the cost for condoms/AIDS prevention brochures?

Businesses were asked only about those materials (condoms/brochures) they had displayed. The estimated cost for condoms/brochures was individually tailored to each business's distributions during the program.

Analysis

The Shapiro-Wilk test for distribution normality, and Levene's test for homogeneity of variances indicated that most variables did not meet parametric assumptions. Therefore, non-parametric tests were used including the Kruskal-Wallis followed by Dunn's post-hoc and Jonckheere tests, and the Mann-Whitney U, Fisher's Exact, and McNemar's tests, with alpha set at .05. Analyses were conducted with SPSS-version 15.0.

Results

Business participation

We visited 102 businesses; 34 had closed in the previous year and were excluded, leaving 68 eligible for recruitment. Of the 51 businesses agreeing to an interview, 30 (59%) agreed to participate in the BEAT AIDS program; 20 agreed to distribute condoms and brochures (CB group); 10 agreed to distribute brochures only (B group), and 21 refused participation. The response rate was 75% (51/68) for obtaining an interview, and 44% (30/68) for agreeing to participate.

The thirty participating businesses included three businesses from each of the ten retail categories. Bars were the easiest retail category to recruit: of the three bars interviewed 100% agreed to participate. Beauty salons and clothing care/repair businesses (e.g., dry cleaners, tailors) were hardest to recruit: 3/7 (43%) beauty salons and 3/7 clothing care/repair businesses agreed to participate. Over the 3-month program, 4 businesses (13%) dropped out. The 26 businesses at post-program included 18 in the CB group and 8 in the B group.

Demographic characteristics of business owners/managers

The 51 business owners/managers interviewed at baseline reported a median age of 47 (range = 23-78), and were 71% male, 78% Caucasian, 14% Latino, 4% Asian/Pacific Islander, and 2% African American; 33% had graduated from high school, 28% had graduated from college/equivalent, 22% held a bachelor's degree, and 12% held a graduate degree. No significant demographic differences were found between business owners/managers in the CB group, B group, and those refusing to participate.

Descriptive characteristics of businesses and customers

Compared to businesses in the B group and those refusing to participate, businesses in the CB group had been open fewer median years (CB = 6, B = 19, Refusers = 13.5, Kruskal-Wallis and Dunn post hoc tests, p's < .05), and had a higher median number of employees (CB = 6, B = 2, Refusers = 3, Kruskal-Wallis and Dunn post hoc tests, p's < .05). CB businesses reported a higher proportion of gay and lesbian customers than B businesses (CB = “about half,” B = “less than half,” Kruskal-Wallis and Dunn post hoc test, p < .05), but did not differ from businesses refusing to participate. There were no significant differences in descriptive characteristics between businesses in the B group and those refusing to participate.

Among the 26 businesses that completed the BEAT AIDS program, CB businesses reported a higher median number of walk-in customers per day than B businesses (CB = 65, B = 15, Mann-Whitney U, p < .05). In addition, 58% of CB and B owners/managers reported that their customers were about half male and half female, 89% reported that most customers were Caucasian, and 92% reported that most customers were 25-44 years (no significant differences between CB and B groups).

Online survey: Customers' descriptive characteristics and feedback

Consistent with owners'/managers' description of their customers, the 52 customers completing the online survey were 60% male, 71% Caucasian, 19% Latino, 2% Asian, 2% Native Hawaiian/Pacific Islander, 2% African American, and 4% multi-racial, with a mean age of 34.02 (range = 18-63). About 52% reported a bachelor's degree, 48% self-identified as homosexual/lesbian, 42% as heterosexual, and 10% as bisexual. Of the 20 CB businesses displaying condoms, 65% were listed at least once as a source of the condom packs. Customers reported living in 16 zip codes representing diverse San Diego County neighborhoods; 31% reported living in Hillcrest.

Ninety-four and 96 percent of the customers agreed that it was a “good idea” for businesses to display condoms and brochures, respectively. Table 1 describes customers' open-ended comments regarding the BEAT AIDS program.

Table 1
Feedback on Program Acceptability and Feasibility

Business owner/manager and research staff feedback

Reasons given by business owners/managers for refusing to participate in the BEAT AIDS program included:

  • “It's not our place to do this.” Some businesses suggested we visit local health centers.
  • “Space is money,” and “I don't know where I'd put this.” Businesses reported paying high rent and wanting every square inch of space to return a profit.
  • “We have middle-aged married women/older women/children coming in.”
  • “I don't want to get involved in something political.”
  • “I just don't see condoms in here.”

Strategies for improving the BEAT AIDS program based on businesses' reasons for refusing, mid-program interviews with owners/managers, and research staff observations are outlined in Table 1.

Number of condoms and brochures distributed

Of 13,224 condoms and 409 brochures distributed to participating businesses, 12,757 condoms and 318 brochures were confirmed as taken. Table 2A shows the median number of condoms taken per month for the 20 businesses distributing condoms (range = 3.7-1,527) and the median number of brochures taken per month for the 30 businesses distributing brochures (range = 0-19.7), with estimated program costs. The median number of brochures distributed per month in the CB group (1.24) and the B group (0.98) did not differ (Mann-Whitney U, p = .58); therefore, these groups were combined to explore factors influencing rates of brochure distribution (Table 2A).

Table 2
Number of Condoms and Brochures Taken per Month by Business Characteristics, Estimated Costs, and Willingness of Businesses to Pay for Condoms and Brochures

Bars had the highest rates of condom and brochure distribution, but there was significant variation in distribution rates among the three bars in this study (range = 100-1,527 condoms and 5.2-19.7 brochures per month), and among the other retail categories. Therefore, we explored business characteristics that could influence distribution rates. Table 2A shows that higher numbers of customers and employees (by tertiles) predicted higher median condom (Kruskal-Wallis and Jonckheere test, p's < .01) and brochure distribution (Kruskal-Wallis and Jonckheere test, p's < .05). No other characteristics of businesses or owners/managers predicted condom and brochure distribution rates.

Business satisfaction

Nearly all owners/managers (96.2%) described their experience in the BEAT AIDS program as “somewhat positive” or “very positive;” 96.2% agreed that the program did not take a lot of time or effort, and 92.3% agreed that they were making a positive contribution to the community.

Influence of condom packaging

Most owners/managers who displayed condoms agreed that they would continue displaying freely provided condoms with the discreet packaging used in this study (94.4%), but this dropped to 33.3% if plain condom packs were substituted (McNemar's test, p = .001).

Businesses' perceptions of customer and employee reactions

Few owners/managers agreed that the program attracted new customers (3.8%), or kept existing customers (19.2%), although 72.2% of those in the CB group agreed that the program resulted in positive customer comments, compared to 33.3% in the B group (p = .04, Fisher's exact test). Nearly 60% in each group reported that the program resulted in positive employee comments.

Businesses' willingness to provide political and economic support

At post-program, most owners/managers agreed that they would be willing to sign a letter to Congress asking for tax breaks (80.8%), put a BEAT AIDS decal in their business window (76.9%), be given positive publicity for participating in BEAT AIDS (69.2%), and sign a letter recommending that other businesses participate (65.4%), while somewhat fewer (42.3%) agreed that they would be interested in meeting with other Hillcrest business owners to discuss ways to prevent AIDS.

Table 2B shows that owners/managers in the CB group reported greater willingness to pay for AIDS prevention initiatives than those in the B group.

Discussion

The present study suggests that community businesses could contribute to expanding the range of settings for HIV prevention. Approximately 60% of the businesses interviewed agreed to participate, most were satisfied with the program, and more than half were willing to provide future political or economic support for AIDS prevention. Although some businesses expressed concern about customer reactions to HIV prevention initiatives, most customers responding to our online survey had positive comments. There is a need to explore customer reactions to AIDS prevention efforts with more diverse samples.

Over 12,000 condoms and about 300 brochures were taken. Compared to brochures, condoms may have greater immediate/future usefulness, or be easier to pocket. Furthermore, while the brochures were designed by a graphic artist to attract interest, their frequent display near other community brochures may have reduced their salience. In contrast, the condoms were the only ones in each business. Future studies should package brochure information together with condoms or, in the case of a brochure-only group, together with another popular product to increase brochure uptake. One business owner increased brochure-taking from 1 to 9 brochures per month by pairing brochures with a community guidebook. In addition, given that most business owners indicated that they would have been unwilling to display condoms without discreet packaging, and customers' preference for our packaging (Table 1), future studies might benefit from using discreet condom packaging.

Businesses that distributed more condoms and brochures had more customers and employees. Not surprisingly, distribution rates increased with more people available to take materials, and increases in patrons taking materials may have cued others to take materials.3 To maximize recruitment success and condom/brochure distribution, businesses with more customers and employees should be prioritized for future invitation into similar programs. However, based on the BEM,3 we also recommend targeting businesses with fewer customers and employees as increasing a community's density/amount of safer sex cues increases opportunities to reach more diverse customers, increases the range of settings for condom access, and may reduce social discomfort associated with discussing and practicing safer sex.

Limitations of this study include small sample size, inclusion of only one community with high AIDS awareness, and having the same staff conduct the intervention and assessments. In addition, while the assessments showed evidence of discriminate validity in differentiating between study groups, additional validation is warranted using controlled trials and measures of change in customer practices. Self-selection bias is also a possibility, as participating businesses and customers may have had greater interest in AIDS prevention. Finally, it is unknown if condoms were taken repeatedly by specific customers, or by different customers. Study findings must be interpreted with caution, as results may not generalize to other communities.

To expand this study, and to promote policies supporting AIDS prevention in businesses, we recommend evaluating if study findings generalize to a broader range of communities, followed by larger quasi- and fully controlled experimental studies to explore if economic or policy incentives increase businesses' willingness to conduct AIDS prevention activities.

Policy change will also require increasing the economic sustainability of AIDS prevention interventions in businesses. Because asking customers to pay part of the cost of condoms may reduce by about half the number of condoms taken,15 other funding sources should be explored. These might include corporate funding of condoms in exchange for advertising on condom packs, and giving businesses graduated tax breaks for incremental contributions to AIDS prevention. We also recommend reducing program costs and identifying optimal settings for future intervention by harnessing computer technology to monitor when materials need to be resupplied and when, where, and how many condoms are opened. Finally, community action has been an integral mechanism of policy change for sustainable public health initiatives, including tobacco and infectious disease control.16 Thus, we recommend building partnerships with community health organizations, state and local health departments, and public interest groups that can help advocate for policies supporting AIDS prevention in businesses.

While businesses have said “it's not my place” to prevent HIV, the BEM suggests that expanding the range of settings that provide ongoing cues and reinforcement for safer sex is critical for building social norms supporting HIV risk reduction.3 Achieving reliable condom and brochure distribution in a critical mass of businesses could facilitate changes in community-wide safer sex behavior and culture.

SO WHAT? Implications for Health Promotion Practitioners and Researchers

This study suggests that community businesses can be engaged in AIDS prevention, and that policy and economic incentives are needed to expand business participation. Combined with other research, there is preliminary support for the assertion that community businesses can increase public access to condoms and exposure to safer sex cues. Repeated exposure to safer sex cues may increase opportunities for conversations about safer sex and change safer sex cultural norms and behaviors. If this assertion holds true, public health practitioners and researchers should explore how to collaborate with policymakers, corporations, and community organizations to establish stable funding for AIDS prevention efforts, and how to harness technology to monitor condom distribution and use.

Acknowledgments

This research was supported, in part, by grants #ID04-SDSU-060 from the Universitywide AIDS Research Program of the University of California, and #1K99 HL88017-01 from the NIH/NHLBI, awarded to Dr. Liza Rovniak, and by intramural support from CBEACH, GSPH, SDSURF, SDSU. We appreciate Stephen Scott's generosity in sharing his expertise running condom distribution programs. We thank Nicole Hannah for her graphic design support. We thank Juanita Andrews, Ding Ding, and Jason Daniel for their assistance in conducting this study.

Footnotes

aCopies of all assessments used in this study may be requested from the corresponding author.

References

1. Hall HI, Song R, Rhodes P, et al. Estimation of HIV incidence in the United States. JAMA. 2008;300(5):520–529. [PMC free article] [PubMed]
2. Hill A, Gebo K. Predicting HIV care costs using CD4 counts from clinical trials. Am J Manag Care. 2007;13(9):524–528. [PubMed]
3. Hovell MF, Wahlgren DR, Gehrman CA. The behavioral ecological model: integrating public health and behavioral science. In: DiClemente RJ, Crosby R, Kegler M, editors. New and emerging models and theories in health promotion and health education. San Francisco: Jossey-Bass; 2002. pp. 347–85.
4. Nielsen Consumer Panel Services. U.S. Consumers Making Fewer Shopping Trips. [December 5, 2008]. Available at: http://www.nielson.com/media/2008/pr_080317.htm.
5. Cherry DK, Hing E, Woodwell DA, Rechtsteiner EA. National ambulatory medical care survey: 2006 summary. Natl Health Stat Report. 2008;(3):1–39. [PubMed]
6. Centers for Disease Control and Prevention. Business Responds to AIDS Program—December 1992-February 1993. MMWR Morb Mortal Wkly Rep. 1993;42(5):95–101. [PubMed]
7. Centers for Disease Control and Prevention. A Heightened National Response to the HIV/AIDS Crisis Among African Americans. Atlanta: US Department of Health and Human Services, Centers for Disease Control and Prevention; Mar, 2007.
8. The CDC AIDS Community Demonstration Projects Research Group. Community-level HIV intervention in 5 cities: final outcome data from the CDC AIDS community demonstration projects. Am J Public Health. 1999;89(3):336–45. [PubMed]
9. Cohen DA, Farley TA, Bedimo-Etame JR, et al. Implementation of condom social marketing in Louisiana, 1993 to 1996. Am J Public Health. 1999;89(2):204–08. [PubMed]
10. Honnen TJ, Kleinke CL. Prompting bar patrons with signs to take free condoms. J Appl Behav Anal. 1990;23(2):215–17. [PMC free article] [PubMed]
11. Kelly JA, Murphy DA, Sikkema KJ, et al. Outcomes of a randomized controlled community-level HIV prevention intervention: effects on behavior among at-risk gay men in small U.S. cities. Lancet. 1997;350(9090):1500–05. [PubMed]
12. Suligoi B, Wagner TM, Ciccozzi M, Rezza G. The epidemiological contribution to the preparation of field trials for HIV and STI vaccines: objectives and methods of feasibility studies. Vaccine. 2005;23(12):1437–45. [PubMed]
13. Bender AP, Williams AN, Sprafka M, Mandel JS, Straub CP. Usefulness of comprehensive feasibility studies in environmental epidemiology investigations: a case study in Minnesota. Am J Public Health. 1988;78(3):287–90. [PubMed]
14. San Diego Association of Governments. Population and Housing Estimates: Zip Code 92103, SANDAG, Current Estimates (2008) [October 15, 2008]. Available at: http://www.sandag.org.
15. Cohen D, Scribner R, Bedimo R, Farley TA. Cost as a barrier to condom use: the evidence for condom subsidies in the United States. Am J Public Health. 1999;89(4):567–68. [PubMed]
16. Centers for Disease Control and Prevention. Achievements in public health, 1990-1999: changes in the public health system. MMWR Morb Mortal Wkly Rep. 1999;48(50):1141–47.