The Delivery of Public Health Interventions Via the Internet: Actualizing Their Potential

The Internet increasingly serves as a platform for the delivery of public health interventions. The efficacy of Internet interventions has been demonstrated across a wide range of conditions. Much more work remains, however, to enhance the potential for broad population dissemination of Internet interventions. In this article, we examine the effectiveness of Internet interventions, with particular attention to their dissemination potential. We discuss several considerations (characterizing reach rates, minimizing attrition, promoting Web site utilization, use of tailored messaging and social networking) that may improve the implementation of Internet interventions and their associated outcomes. We review factors that may influence the adoption of Internet interventions in a range of potential dissemination settings. Finally, we present several recommendations for future research that highlight the potential importance of better understanding intervention reach, developing consensus regarding Web site usage metrics, and more broadly integrating Web 2.0 functionality. 273 A nn u. R ev . P ub lic . H ea lth . 2 00 9. 30 :2 73 -2 92 . D ow nl oa de d fr om w w w .a nn ua lr ev ie w s. or g by U ni ve rs ity o f O re go n on 0 3/ 27 /1 1. F or p er so na l u se o nl y. ANRV370-PU30-15 ARI 15 February 2009 13:8 RCT: randomized controlled trial RE-AIM: Reach, Efficacy/Effectiveness, Adoption, Implementation, Maintenance INTRODUCTION In the early morning of December 22, 1982, Jack Buchanan, MD, posted a simple message to a USENET newsgroup—a still widely used Internet-based social-networking system. Responding to an earlier inquiry, Buchanan copied a passage from a recent edition of the CDC’s Morbidity and Mortality Weekly Report (which, for the uninitiated, he described as a “respected newsletter type of thing”). The passage described the known etiology of a “frightening” new condition on which newsgroup members were actively seeking information. With his message, Buchanan helped to stimulate a major paradigm shift in public health intervention: This conversation was the first time that AIDS information was shared on the Internet. More than a quarter century later, the Internet has reached near ubiquity and is generally regarded as an indispensable communication tool throughout the developed world. With the dramatic increase in Internet access has been a parallel increase in the use of the Internet as a platform for the delivery of public health interventions across a wide range of conditions and population segments. We operationalize the term Internet interventions to refer to systematic treatment/prevention programs, usually addressing one or more determinants of health (frequent health behaviors), delivered largely via the Internet (although not necessarily exclusively Web-based), and interfacing with an end user. These interventions are typically highly structured, mostly self-guided, interactive, and visually rich, and they may provide tailored messaging based on end-user data (115). A veritable explosion in the number of randomized controlled trials (RCTs) testing Internet interventions has taken place, most emerging during the past half decade. At present, we find ourselves at an inflection point: The Internet as a platform has largely been deemed efficacious, and as the next generation of trials begins, greater attention will be needed to determine both the effectiveness and the dissemination potential of public health Internet interventions (42). Building on several recent reviews in this area (48, 69, 110, 130, 144, 152), our aims were threefold: (a) to review evidence on the effectiveness of public health Internet interventions, (b) to discuss considerations related to the dissemination potential of Internet interventions, and (c) to identify issues and trends that may prove fruitful for future research. Throughout, our comments are focused specifically on RCTs of Internet interventions for consumers/end-users and include both primary and secondary prevention interventions. To limit the scope, we have chosen not to include Internet systems for health professionals, clinical data management (e.g., personal health records, electronic medical records, training programs, guidelines/practice standards), assessment, survey administration, or telemetry (unless utilized within an intervention). Given our overarching interest in the dissemination of Internet interventions, we have employed the RE-AIM (41) framework to organize the discussion. RE-AIM is a planning and evaluation framework that focuses on factors critical for translating research into practice (Table 1). WILL THEY COME? THE POTENTIAL REACH OF INTERNET INTERVENTIONS Data from the Pew Internet and American Life Project show that almost three-quarters of U.S. households have access to a home computer and nearly 75% of adult Americans are regular Internet users (those who use the Internet and send/receive email “at least occasionally”). Internet use is strongly patterned by sociodemographic characteristics. Whereas more than 92% of adult Americans aged 18–29 years old use the Internet, only 37% of adults aged 65 or older are regular Internet users (106). Internet use is more prevalent among non-Hispanic whites (76%) and English-speaking Hispanics (79%) compared with non-Hispanic blacks (56%). Among the Hispanic population, English fluency is a major driver of Internet use; although 76% of bilingual Hispanic adults are regular Internet users, only 32% of primarily 274 Bennett · Glasgow A nn u. R ev . P ub lic . H ea lth . 2 00 9. 30 :2 73 -2 92 . D ow nl oa de d fr om w w w .a nn ua lr ev ie w s. or g by U ni ve rs ity o f O re go n on 0 3/ 27 /1 1. F or p er so na l u se o nl y. ANRV370-PU30-15 ARI 15 February 2009 13:8 Table 1 RE-AIM model elements, definitions, and Internet intervention example RE-AIM element Definition Internet intervention example Reach The number and percent of those invited and eligible who participate and their representativeness Sixteen percent of diabetes patients invited to an Internet self-management intervention participated. Those declining were more likely to be Latino and male. Effectiveness The amount of change in temporally appropriate outcomes and impact on quality of life or any adverse (iatrogenic) effects Thirty percent of those randomized to an Internet smoking-cessation program quit compared with 12% in the control condition. The study showed no differences between conditions of weight change or quality of life. Adoption The number, percent, and representativeness of settings and staff invited who participate Forty-six percent of work sites approached to participate in an Internet health-promotion program evaluation took part. Work sites participating were larger, had more white-collar employees, and offered more wellness activities. Implementation The extent to which a program or policy is delivered consistently, and the time and costs of the program The average number of log-ins in an Internet physical activity intervention was 5.2. Usage decreased over the eight-week intervention, and the number of overall log-ins and use of the social support forum were associated with greater improvement. Maintenance (individual level) The long-term effects on key outcomes and quality-of-life impact At a 6-month follow-up, an Internet weight-loss program experienced 60% attrition. Those responding lost an average of 9 pounds. A mail follow-up of initial nonrespondents revealed an average weight loss of 8 pounds among this group. Maintenance (setting level) The extent to which a program or policy is sustained, modified, or discontinued following initial trial or study period Of 24 schools participating in an online drug-abuse prevention program, 6 continued the program unchanged, 10 requested substantial changes or added their own components, and 8 discontinued the program. Spanish-speaking Hispanics are (106). There remains a strong socioeconomic gradient in Internet use, with high levels found among those with a college education (93%), relative to those with high school education (67%) or less (38%). More than 50% of Internet users have a broadband connection (103). Broadband access is more prevalent among younger adults, in higher socioeconomic status households, and in urban/suburban areas (55). The racial/ethnic gap in broadband access has narrowed considerably and largely disappears after controlling for other sociodemographic characteristics and Internet use (55). The world penetration of Internet technologies is estimated at 21.9%, with the highest penetration found in the United States (73.6%), Europe (43.4%), and Asia (15.3%). Internet usage, however, is higher in Europe and Asia than in the United States. In Europe, the highest numbers of Internet users are found in Germany, the United Kingdom, and France, whereas in Asia, Internet usage is highest in China, Japan, and India. Growth in Internet usage (from 2000 to 2008) has been high across the world, but particularly in Africa and the Middle East (97, 159). For many, the Internet serves as a ubiquitous source of access to health information. Health information on the Internet is largely trusted; nearly three-quarters of online health information seekers reported that they do not consistently check the source and date of the information that they discover (40). Given the variable quality of Internet-based health information (37), this finding is concerning. However, it perhaps lends additional support to the notion that the Internet landscape is well positioned for public health intervention. The combination of high potential reach and efficacy suggests several possible advantages of Internet interventions compared with other modalities (41). Internet-based www.annualreviews.org • Public Health Internet Interventions 275 A nn u. R ev . P ub lic . H ea lth . 2 00 9. 30 :2 73 -2 92 . D ow nl oa de d fr om w w w .a nn ua lr ev ie w s. or g by U ni ve rs ity o f O re g