Treatment perceptions and attitudes of older human immunodeficiency virus–infected adults

1. Introduction 

The over-50 demographic is witness to one of the fastest growing segments of acquired immunodeficiency syndrome (AIDS) cases.1 Estimates have suggested that by the end of 1999 more than 17% of individuals living with AIDS were older than 50 years, suggestive of a “graying” in the HIV epidemic. These data indicate that greater than 10% to 14% of all new AIDS cases and more than 5% of all newly diagnosed HIV infection in the United States occur among people 50 years and older.1, 2 Furthermore, 14.7% of male AIDS cases and 12.3% of female AIDS cases diagnosed in 1999 were among Americans older than 50 years1; however, an analysis of Medicare claims files from 1991 to 1993 indicates that there may be a severe underreporting of AIDS cases among people older than 65 years.3 The result is the emergence of a growing group of HIV-infected patients about which little is known. There is also growing evidence that suggests that median survival with HIV and progression time to AIDS are shortened in older patients.4 One recently reported meta-analysis of 38 individual studies comprising more than 10,000 patients suggests that the median time to progression of AIDS was 50% shorter among individuals 65 years and older compared with individuals who seroconverted at 25 to 34 years. Furthermore, the rate of progression to AIDS among the older individuals was double that of the 25- to 34-year-old comparison group during the study period. Lieberman also notes in a study of HIV in older Americans that prevention, counseling, testing, and research efforts have not been directed at this growing population of patients.5 Moreover, there are relatively few practitioners who are experts both in health problems associated with aging and HIV disease.

Although still limited, an increasing amount of research is now being conducted specifically among older adults who are infected with HIV. These studies have investigated several issues regarding HIV and its management, including general mental health,6 psychological distress,7, 8, 9 depression,7, 9 coping,8, 9 social support,8 and adherence to HIV medications.10, 11, 12

To advance this line of research, the current study was conducted to examine the health beliefs, sociodemographic characteristics, HIV transmission risk behaviors and perceived benefits, and barriers to treatment in a group of older adults diagnosed with HIV by using a modified version of the Health Belief Model (HBM) as its conceptual framework. The HBM is one of the most frequently cited and researched cognitive models of health behavior.13, 14, 15, 16, 17 One primary strength of the HBM is its incorporation of individual subjective assessments of health situations and their relationships with health behaviors and actions. The HBM consists of 6 dimensions: perceived susceptibility, perceived severity, perceived benefits, perceived barriers, cues to action, and modifying factors. The most notable dimensions of the HBM in theoretical propositions and empirical investigations are the 4 perception variables.16, 17 These perception variables are hypothesized to represent a decision process that individuals may go through when choosing among alternative courses of action related to health. The current study did not assess the HBM variable of “cues to action.”

Previous studies have used the HBM to predict risk behaviors in the primary prevention of HIV.18, 19, 20 The bulk of the evidence shows that HBM perceptions are predictive of HIV risk behaviors among both homosexual men and heterosexual populations.21, 22, 23 In one of the few studies conducted among older adults, Rose applied the HBM in a study of knowledge and beliefs about HIV among a group of older persons in Florida.24 She concluded that older adults in the sample had a relatively good knowledge base about AIDS; however, many participants in the study did not believe that they were susceptible to AIDS, even while engaging in high-risk behavior. In a follow-up study, Rose developed an AIDS education program on HIV/AIDS knowledge, perceived susceptibility to AIDS, and perceived severity of AIDS in older adults.25 The study concluded that the education program had resulted in a significant increase in total knowledge about AIDS, perceived susceptibility, and perceived severity of AIDS in the participants.

None of the reviewed studies evaluated secondary risk prevention in a population of older HIV-infected adults. The reported studies typically evaluated primary risk behavioral prevention among college students or young adults with specific behavioral risk. Because a significant portion of HIV-infected or AIDS patients are older Americans,1, 26 a basis for understanding their risk behaviors and treatment practices is needed. Patients' perceptions about the disease as well as their beliefs about and evaluations of treatment options affect their choices among alternative courses of actions; however, few known studies have used a theoretical model to examine how patients' perceptions about HIV disease and its treatment affect the risk and treatment behaviors of people who are already infected with HIV. The primary objective of this study was to apply the HBM in assessing the perceptions of older HIV patients regarding their disease. Furthermore, the study assesses the effect of disease and treatment perceptions on adherence to antiretrovirals, use of alternative therapies, and risk behaviors.

2. Methodology 

Study participants included a convenience sample of 100 older HIV-infected patients, 50 years and older, who were being treated at 2 large HIV clinics in Washington, DC, and who agreed to participate. Patients were enrolled if they were older than 50 years at the time of recruitment, had a positive serological confirmation of HIV infection, and could understand English. Patients were recruited by the use of flyers displayed in patient waiting areas of the participating clinics and by clinicians practicing there. Approximately 10 eligible patients who were recruited declined to participate. Patients participated in structured interviews and were queried regarding: demographic characteristics, HIV transmission risk behaviors and perceived benefits, and barriers to treatment. Patients were interviewed between April 1999 and March 2000. Reviews of patients' medical records were used to collect clinical status data, including the most recent viral load, and CD4 count prior to interview but within 3 months of interview date. Patients' self-reports of antiretroviral medication regimen were also validated from data in the medical record. Each participant was offered a financial incentive of $25 in exchange for voluntary participation. Each participant also completed an informed consent form approved by the Howard University Institutional Review Board (IRB) and by each clinic. These patients participated in a 30-minute structured interview designed to address the objectives of this study. The structured patient interview, conducted by trained study investigators, covered the broad areas of: (1) risk behaviors for transmitting HIV, (2) adherence to antiretroviral medications and barriers/benefits of such use, (3) use of alternative therapies and barriers/benefits of such use, and (4) demographic characteristics. All interviews were held at the clinics in private, quiet patient rooms during weekday mornings and afternoons. Sample size calculations were not conducted because of the exploratory nature of the study. Additional details about the methods used in this study have been described elsewhere.27

3. Results 

A total of 100 older HIV-infected patients were recruited to participate in this cross-sectional study from 2 metropolitan Washington, DC, HIV clinics. Patients ranged from 50 to 76 years at the time of interview. Eighty-five patients were between 50 and 59 years, and 2 patients were older than 70 years. The demographic characteristics of the sample are detailed in Table 1. The majority of study participants were black (75%) and male (78%). Most had a high school education or better (74%). The mean age of participants was 54.5 years, and they had been diagnosed with HIV for an average of 7.2 years. Sexual activity was the most common route of HIV transmission, with male-to-male transmission reported by 35% of the study group and heterosexual transmission reported by 21%. Intravenous transmission of HIV was reported by 24% of the study group. The vast majority (90%) of the participants indicated that they did not have sex in exchange for money or drugs nor shared needles to inject drugs in the 6 months prior to study enrollment. Around 15% of the respondents indicated that they had been diagnosed with a sexually transmitted disease in the previous 6 months. Regarding medication use, the mean number of antiretroviral medications taken by the participants during the study period was 2.8, and the majority of patients (86%) were on multiple drug combinations, including 55% who were receiving protease inhibitors as part of their combination therapy. Twenty-one patients (21%) reported using some form of alternative therapy to treat or ameliorate their HIV infection. Thirteen percent of patients had either refused antiretrovirals or were not otherwise receiving them. Of the 87 patients who were taking antiretrovirals, only 14 participants reported adherence rates less than 90%. The mean self-reported adherence was 94.4% (median=100%). Adherence ranged from 0% to 100%.

Table 1. Demographics characteristics (N=100)

		Frequency	Percent	Mean ± SD
	Sex
	Male	78	78
	Female	22	22
	Race
	Black (non-Hispanic)	75	75
	White (non-Hispanic)	20	20
	Hispanic	3	3
	Native American/Alaska Native	2	2
	Education
	Less than high school graduate/degree	26	26
	High school graduate/degree	27	27
	Some college	19	19
	College graduate/degree	14	14
	Graduate/professional school graduate/degree	14	14
	Marital status
	Unmarried	44	44
	Married or intimate relationship	24	24
	Divorced/separated	22	22
	Widowed	10	10
	Annual income
	Less than $10,000	53	53
	$10,000-$19,000	19	19
	$20,000-$29,000	9	9
	$30,000-$39,000	3	3
	$40,000-$49,000	3	3
	Greater than $50,000	4	4
	Do not know	1	1
	No answer	8	8
	Mode of transmission
	Male-to-male (homosexual)	35	35
	Intravenous drug use	24	24
	Heterosexual (sex with HIV-infected person)	21	21
	Transfusion	4	4
	Unknown/do not know	14	14
	Other	2	2
	Age (in y)			54.5 ± 5.1
	Number of years since diagnosis of HIV			7.2 ± 5.2
	Number of antiretroviral medications			2.8 ± 0.9
	Number of non-HIV medications			1.6 ± 1.8

Responses to the study measures are described in Table 2. The mean score for perceived severity of HIV was 14.3 (±2.6) and above the scale's median value. The mean score for perceived susceptibility to early progression of AIDS was 8.5 (±3.1), just below the scale's median value. The mean scores for perceived benefits of antiretrovirals and perceived costs (barriers) of antiretrovirals were 16.4 (±2.8) and 18.5 (±6.3), respectively. The mean scores for perceived benefits of alternative therapies at 12.2 (±6.8) and perceived costs (barriers) of alternative therapies at 16.4 (±6.4) were close to the midpoint of the scales' midpoint values. Of the 6 multiple-item HBM perception scales, 3 exhibited acceptable reliabilities of 0.68 (Cronbach's α) or higher, 2 exhibited marginal reliabilities of 0.52 and 0.54, and 1 scale (perceived susceptibility) exhibited a reliability score less than 0.5. Although reliability scores of 0.7 (or even 0.6) or higher are in general considered acceptable,29 Anastasi and Urbina suggest that for studies with smaller sample sizes (100 or fewer participants) a Cronbach's α score of 0.5 can be considered appropriate.30 The items retained in the scales are the ones that provided best overall internal reliability for each scale.

Table 2. Study variables (N=100)

		Frequency	Percent	Mean ± SD	Cronbach's α
	Clinical status
	CD4 count (cells/mm3) (within 3 mo of study interview)			432.0 ± 283.3
	Viral Load (copies/ml HIV RNA) (within 3 mo of study interview)			52,555.7 ± 159 606.0
	Health perceptionsa
	Perceived severity of HIV (4 items)			14.3 ± 2.6	0.52
	Perceived susceptibility to early progression to AIDS (3 items)			8.5 ± 3.1	0.39
	Perceived benefits of antiretrovirals (4 items)			16.4 ± 2.8	0.54
	Perceived costs (or barriers) of antiretrovirals (8 items)			18.5 ± 6.3	0.82
	Perceived benefits of alternative therapies (5 items)			12.2 ± 6.8	0.86
	Perceived costs (or barriers) of alternative therapies (6 items)			16.4 ± 6.4	0.68
	Health behaviors
	HIV transmission risk behaviors (in past 6 mo)b			1.0 ± 1.0	0.48
	Shared needles	1	1
	Diagnosed with a sexually transmitted disease	15	15
	Had sex in exchange for money	6	6
	Had sex in exchange for drugs	4	4
	Had multiple sexual partners	31	31
	Did not always use condoms during sexc	21	35
	Compliance/adherence to antiretroviral therapy (percentage)			94.4 ± 13.8
	Use of alternative therapies
	Yes	21	21
	No	79	79

a These variables were measured on a 5-point Likert-type scale anchored by 5, strongly agree, and 1, strongly disagree. Possible range of scores is derived by multiplying the number of items (possible low score) to 5 (highest score); eg, a 4-item measure has a possible range of 4 to 20. b Risk behaviors were coded as yes (1 point) or no (0 points) with the exception of condom use and number of sexual partners. Condom use was measured as always (0 points) or less than always (1 point). Number of sexual partners was recorded as 0 (0 points), 1 (1 point), or greater than 1 (2 points). c Based on the respondents who participated in sexual activity in the previous 6 months (n=60).

Based on the results of Student t tests (Table 3), users of alternative therapies had higher perceptions of the severity of HIV disease (t=2.33, df=98, P=.03), higher perceived barriers to the use of antiretroviral medications (t=3.89, df=98, P < .001), and higher perceived benefits of the use of alternative therapies (t=4.66, df=98, P < .001) compared with nonusers of alternative therapies. Logistic regression analyses (Table 4) of use (nonuse) of antiretrovirals and alternative therapies for the 6 HBM perception variables revealed that after adjusting for age, race, and sex, patients who perceived a greater burden of using antiretrovirals were more likely to use alternative therapies (OR=1.1; 95% CI, 1.0 to 1.3; P=.04) and that patients who perceived greater benefit of using alternative therapies were more likely to use these therapies (OR=1.3; 95% CI, 1.1 to 1.5; P=.001). None of the perception variables were significant in predicting the use of antiretrovirals.

Table 3. Selected t test results

	Dependent variable	Independent variable	t	df	P value
	Perceived severity of HIV disease	Use of alternative therapy	2.33	98	.03
	Perceived barrier to use antiretroviral medications	Use of alternative therapy	3.89	98	.001
	Perceived benefits of using alternative therapies	Use of alternative therapy	4.66	98	.001
	Perceived severity of HIV disease	Race (white or minority)	2.24	98	.027

Table 4. Adjusted logistic regression model on the use of alternative therapies

	Independent variable	Coefficient	SE	OR (95% CI)	P value
	Perceived burden of using antiretrovirals	0.12	0.06	1.1 (1.0, 1.3)	.04
	Perceived benefit of using alternative therapies	0.26	0.08	1.3 (1.1, 1.5)	.001

Adjusted for sex, age, and race.

Other factors that were significantly associated with respondents' perceptions about HIV and the benefits and costs (or barriers) to the treatment of HIV were race and income. White patients had a higher perceived severity of HIV compared with patients of color (t=2.240, P=.027, df=98). Also, higher income (one aspect of socioeconomic status) was significantly associated with higher perceived benefits of antiretrovirals (r=0.23, P=.01). Perceived barriers to the use of antiretrovirals had an inverse relationship to clinical status. Perceived costs (or barriers) of antiretrovirals were inversely associated with CD4 count (r=−0.25, P=.01) and positively associated with viral load (r=0.33, P < .01).

In other words, higher perceived barriers were related to lower CD4 counts and higher viral loads. In general, the lower a person's CD4 count the higher is his/her susceptibility to infection. Moreover, a lower CD4 count potentially indicates a higher viral load in the patient. Perceived costs (or barriers) of antiretroviral use were also found to be inversely associated with age (r=−0.21, P=.02) and positively associated with having no prescription insurance (r=0.33, P < .01). That is, barriers to antiretroviral use decreased with increased age. Also, barriers to antiretroviral use increased in patients without prescription drug coverage.

In assessment of risk behavior, nearly a third of participants reported having multiple sexual partners in the previous 6 months, and 35% reported not always using a condom during sex. However, a vast majority of the participants indicated that they did not have sexual intercourse in exchange for money or drugs and had not been diagnosed with a sexually transmitted disease during the prior 6-month period. Bivariate analysis was conducted to determine if there was a correlation between the 6 HBM perception scales and the behavioral risk factors. The 6 risk behavior factors included: sharing needles to inject drugs, diagnosed with sexually transmitted disease, number of sexual partners, whether the participant had sex in exchange for money and/or drugs, and condom use. Bivariate correlation analysis did not indicate a significant relationship between the proposed risk factors and the HBM perception variables; however, there was a correlation between having sex in exchange for money and higher levels of viral load (r=−0.28, P=0.01).

Regarding the use of alternative therapies, Wutoh and colleagues previously reported that of the 21 patients who acknowledged use of alternative therapies, 16 used alternative therapies in conjunction with prescribed antiretrovirals, while the remaining 5 used alternative therapies instead of prescribed antiretrovirals. There was no significant difference in the use of prescribed antiretrovirals if the patients used alternative therapies (P=.097).27

4. Discussion 

The results of this study provide insight into the treatment attitudes and risk behavioral patterns of older HIV+ patients. The demographic characteristics and treatment patterns of this population were similar to the overall HIV caseload in the Washington, DC, area.31 The vast majority of patients were on multiple drug combinations, including 55% receiving protease inhibitors. Alarmingly, 13 patients either refused antiretrovirals or were otherwise not receiving them. The lack of therapy among these patients was related to their attitudes regarding antiretrovirals as well as to their use of alternative therapies. It is likely that the high self-reported adherence noted by patients was exaggerated; however, the authors in another publication note significant statistical associations between the self-reported adherence and HIV viral load in this study group.27 We did not find any significant difference between antiretroviral users and nonusers regarding the HBM perception variables, although this may be a product of the relatively small number of patients in the nonuse group and relatively low reliability scores on some of the HBM perception scales.

Study results suggested that although participants were aware of the severity of HIV, they did not personally view themselves at risk to progressing to AIDS. This may be because of the high perceived benefits of taking antiretrovirals and to high self-reported adherence levels. Overall, participants were aware of the benefits of antiretrovirals and were not particularly burdened by the cost (or perceived barriers) of antiretrovirals. Neither the perceived costs (or barriers) associated with using alternative therapies nor its benefits were rated to be very high by these patients; however, patients tended to use alternative therapies when they perceived higher benefits in using such modalities and/or when they perceived using antiretrovirals as a burden (cost). Such a finding was consistent with other researchers who have increasingly found higher use of alternative therapies among HIV-infected patients.32, 33 Because greater perceived barriers of using antiretrovirals was inversely related to CD4 count and positively related to viral load, this may suggest that patients who use alternative therapies in lieu of prescription antiretrovirals may be at increased risk of disease progression and AIDS. While other studies have reported on the use of alternative therapies among HIV-infected patients,32, 33, 34 the additional impact of higher age in this group may be of concern.

Regarding assessment of risk behaviors, there was a relatively high level of sexual activity among these older HIV-infected participants. Nearly one third reported having multiple sexual partners in the previous month. This highlights a concern that HIV-infected patients may still be continuing high-risk behavior, despite their HIV status. It is of greater concern among this group of older patients because societal preconceptions generally do not view older individuals as being very sexually active or provide sufficient risk reduction programs for older individuals. It is of particular concern that 35% of the participants who acknowledged sexual activity in the previous 6 months reported not always using a condom during sex. While the participants generally noted that they did not have sex in exchange for money or drugs, 6 patients were diagnosed with a sexually transmitted disease in the prior 6-month period. The low overall reliability of the risk behavior scale highlighted the deficiencies in this tool. One potential problem in the design of this scale may be the combination of sexual and drug-use risk to create a single measure of overall risk. Researchers have noted that a reduction in sexual risk may not necessarily correspond with a reduction in drug-use risk.35 However, when each element of the risk behavior scale was assessed individually, it provided interesting insight regarding the behaviors of this patient group.

5. Limitations 

The participants in this study were enrolled as a convenience sample and may have potentially self-selected based on a number of sociodemographic or other characteristics. Indeed, the majority of patients were male and African American. The demographic characteristics of the study population may not be consistent with other groups of older HIV-infected adults. Furthermore, as an exploratory study, the patient sample was not determined through sample size calculations. Another limitation was the low reliability of the perceived susceptibility to early progression to AIDS scale (0.39). Additionally, 2 other perception measures exhibited reliabilities below 0.6 (perceived severity of HIV and perceived benefits of antiretrovirals), thus impacting any conclusions that can be made from their use. Also, the data analyzed during this study were collected 4 years prior to preparing this manuscript, and it is possible that patients' attitudes and perceptions may have changed over this period.

Future research regarding HIV in older Americans should include the development of several behavioral methods and specifically a more reliable measure of risk behavior patterns among older patients. There is also a need to investigate the potentially high rate of sexual activity among older HIV patients and the corresponding alarming rate of inconsistent condom use among sexually active older patients. Future research should also focus on the use of alternative therapies among this population of patients to investigate the use of specific alternative therapies and to assess the effect of the use of alternative therapies on adherence to conventional antiretrovirals. While the high self-reported rate of antiretroviral adherence in this population was encouraging, it would be useful to compare self-reported adherence with some other objective measure of medication compliance.

A final limitation involves the cross-sectional nature of the study. It represents one point in time and does not reflect changes in behavior or regimens over time. Despite these potential limitations, the results of this study provide important insight regarding the treatment behaviors of older HIV-infected adults. Future research considerations will include evaluations of methods to improve antiretroviral adherence among older HIV-infected adults, assessment of the various types of alternative therapies used among older patients, development of tools to more adequately assess health perceptions and behavioral intentions, the development of specific prevention strategies, and assessment of specific treatment regimens that will improve antiretroviral adherence.

6. Conclusions 

The results of this study may help enable health care practitioners to better meet the needs of their older patients who are infected with HIV and perhaps to anticipate the needs of younger HIV-infected patients who are likely to survive into their middle and later years. Health care providers would be better equipped to plan strategies for the reduction of risk behaviors and for the management of HIV that have increased potential for effectiveness for older HIV-infected populations. The results of this study also can be applied in the development of specific interventions that are intended to decrease HIV transmission among older adults and to improve medication-taking behavior among those who are already infected with HIV. However, it is necessary to evaluate the applicability of other health behavioral models and concepts among this population in an effort to model the behavior of older HIV-infected adults and to examine the effect of health behaviors on health outcomes.