Six hundred and sixty two (N = 662) usable surveys were returned from 42 states over a 5-month period. Some additional surveys were discarded either because the respondents were not HIV-positive or because they left portions of the survey blank. The base on which we calculated the following percentages fluctuates somewhat due to item nonresponse. The results that follow should be understood only in light of the shortcomings of our survey distribution procedures. First, although it is typical in survey research to calculate and report a response rate, our distribution procedures prevent us from doing so. We know that we distributed 10,500 instruments to various organizations, but it is impossible to know how many of those surveys were actually offered to possible respondents. Based on calls to various organizations, we know that many survey packages did not reach a staff member who could recruit participants for the study, despite the fact that we mailed over 95% of them to a specific, named individual. Several organizations also later said they could not participate in the study due to time constraints and the deadline for returning completed surveys. Still other organizations requested additional surveys in quantities that later turned out to be overly optimistic given the time that their staff members had to speak with clients about the study. Our distribution technique also prevented us from forming a randomized sample. We specifically targeted organizations that served diverse clientele, and we do not know how the surveys were actually distributed at the various sites. We included instructions on how to distribute the survey, but health and social service professionals face hectic schedules and sometimes overwhelming demands. We expect that they did the best they could under these circumstances.
Finally, because our survey was distributed at health and service organizations, it only made it into the hands of PLWHA who utilize such organizations. Unfortunately, we can say nothing about the information preferences or practices of those individuals who either have no access to or choose not to use these institutional resources. Still, we believe we achieved our goal of surveying many of the diverse groups affected by the epidemic. As shown by the demographic breakdown in , we reached most of the targeted groups. However, we did not succeed in getting sufficient data from younger adults. Indeed, selection bias and underutilization of service organizations may explain why we received so few responses from this group, one in dire need of further study. Some demographic points of interest include: the range of annual incomes, from none to over $60,000; different work situations, with 162 retired or not employed, 171 employed, and 302 on disability; and T-cell counts from 0 to 2,000.
Table 1 Demographic characteristics of sample
Information source preferences
The survey explored respondents' assessments of HIV information sources along several lines. One question asked, “How do you best like to get HIV information?” and provided a series of options and space to write in “other” sources not listed. Respondents were instructed to put the numbers “1,” “2,” and “3” next to their top three choices. As shown in , 43% of respondents selected doctors as their 1st choice, and 70% ranked doctors in their top 3 sources. HIV-positive counselors and magazines had the next highest frequencies but were more than 30% lower than doctors, and brochures and newsletters followed but had very low “first choice” frequencies.
Sources of HIV informationPercentages were calculated using the number of actual responses to the question. Some respondents selected fewer than three sources.
It is important to note that while the lowest rated information sources in are not widely preferred, they serve some groups in the HIV/AIDS community more than others. For example, people who did not complete high school were three times more likely than college graduates to choose videos as a top-three choice. In general, white men who identified as homosexual preferred newsletters, the Internet, educational forums, and peers, while other groups preferred brochures, pamphlets, videos, classes, nurses, and HIV-positive counselors. There was also variation based on how long respondents had known they are HIV-positive. Doctors were still the top source by far, but for PLWHA who had lived with HIV for less than ten years, HIV-positive counselors were ranked second. For those who had lived with HIV for more than ten years, newsletters and forums were more commonly preferred sources.
The Internet was not rated highly overall, but whites were about twice as likely to choose it as African Americans or Hispanics. Moreover, 25% of college graduates listed it in their top 3 choices, while it was only selected by 5% of those who did not finish high school, 8% of high school graduates, and 15% of those with some college. Twenty-five percent of respondents living in major metropolitan areas listed the Internet among their top three sources. Those living in other areas listed it less than half as frequently: large cities, 12%; suburbs and large towns, about 10%; and small towns or rural areas, 6%. Despite the subjectivity inherent in how respondents described their location, these data revealed a notable divide in Internet preference: PLWHA in major metropolitan areas favored the Internet more than those who did not live in such areas.
In two similar questions, respondents were asked (1) what people and (2) what information sources “encourage and support you to take positive actions to deal with your HIV?” For the first question, the doctor category was again rated highest, with 79% of respondents selecting it as one of their top 3 choices. Friends and family were not far behind, at 72%. Case managers were the 3rd highest, at 41%, but were more than 30% lower than friends and family. In the 2nd question, magazines and pamphlets were the highest rated information sources, at 64% and 63%, respectively, and newsletters were 3rd with 54% of respondents listing them in their top 3.
Table 2 Characterizing HIV information sources
In a separate question on barriers to using HIV information, availability was rated the lowest. As illustrated in , only 8% of respondents indicated that HIV information was “hard to find” as a top 3 barrier, while 35% listed that “too much” information was available. The 2 options most selected as top 3 barriers were “hard to understand,” at 40%, and “not sure whether to trust,” at 38%. The most frequently selected 1st choice related to applicability, with 15% listing “not enough information applies to me” as the biggest barrier. Variability across demographic groups was limited on this question.
Barriers to using HIV informationPercentages were calculated using the number of actual responses to the question. Some respondents selected fewer than three barriers.
Questions about information seeking and sharing were distributed across the survey, and each elicited similar responses, as shown in . A majority of respondents agreed or strongly agreed that they actively searched for information, were confident in their abilities to find information, and regularly read to learn more about HIV. They also agreed that new information helped keep them healthy and helped them feel good about themselves.
Table 3 Information seeking and sharing
Demographic breakdowns show 56% of Hispanics and 58% of African Americans strongly agreed that they actively searched for new HIV information, compared to 39% of whites. Confidence in finding information was also higher for African Americans (57%) and Hispanics (53%) than for whites (40%). Women appeared to be more active readers than men: 62% strongly agreed that they regularly read to learn about HIV, compared to 47% of men. Respondents with self-reported high treatment success also tended to be active information seekers, 75%, compared to 55% for those reporting low treatment success. Responses to the 6th entry shown in indicated that information sharing was a common practice. Eighty percent of respondents agreed or strongly agreed that they gave advice or told others where to get information. Among those who reported sharing information, a segment of respondents (105 in total) demonstrated faulty knowledge about the implications of T-cell count measures (they did not know at what number of T-cells a person risks getting opportunistic infections). These respondents were about as likely to share information as those who had correct understanding of the implications of T-cell counts for their health, raising important questions about the quality of information being transferred informally among PLWHA.
While most respondents actively sought information, their interactions with it were not always positive. The responses shown in the 7th entry in show that 71% agreed or strongly agreed that it was easy to feel overwhelmed by HIV information. The frequencies for this question were similar across all groups of respondents, including those who had been living with HIV/AIDS the longest. Interestingly, while a large majority of respondents were overwhelmed with information, a substantial segment (43%) strongly disagreed that at times it was better not to seek information. On the other hand, 31% either agreed or strongly agreed that not seeking information could be beneficial.
It also appeared that PLWHA's ability to make treatment choices and follow their medication schedules could benefit from better information support. Thirty-eight percent of respondents agreed or strongly agreed that they did not know enough to make good treatment choices. Also, education level was an important factor in understanding treatment choices. Few college graduates (6%) felt that they did not know enough to choose wisely, but 34% of those without a high school degree felt this way. Adherence to treatment is a critical issue in self-care and disease management. The World Health Organization recently defined adherence as “the extent to which a person's behavior—taking medication, following a diet, and/or executing lifestyle changes—corresponds with agreed recommendations from a healthcare provider” [26
]. Among PLWHA, not adhering to combination antiretroviral (ARV) regimens might result in drug resistance and treatment failure. Unfortunately, 35% of respondents strongly or somewhat agreed that they were not getting much useful information on how to stay consistent with taking their medications. The rates were higher for unemployed respondents (45%) and non– high school graduates (44%). Moreover, those with less adherence-related information might be managing their medications in a more haphazard manner. Among the 35% not encountering adherence-related information, 40% did not systematically keep track of when they missed taking their medications.