Factors Associated With Adherence to Antiretroviral Therapy

Factors Associated With Adherence to Antiretroviral Therapy

Features Factors Associated With Adherence to Antiretroviral Therapy Marcia K. McDonnell Holstad, DSN, RN, C, FNP James C. Pace, DSN, RN, MDiv, ANP-C...

172KB Sizes 18 Downloads 48 Views


Factors Associated With Adherence to Antiretroviral Therapy Marcia K. McDonnell Holstad, DSN, RN, C, FNP James C. Pace, DSN, RN, MDiv, ANP-CS Anindya K. De, PhD, MS, MStat Darla R. Ura, MA, RN, APRN, BC

The purpose of this study was to examine the relationship of sociodemographic characteristics, patient perceptions, and patient characteristics including spirituality, self-reported adherence, and highly active antiretroviral therapy. The convenience sample consisted of 120 English-speaking adults (60% male, 35% female, 5% transgendered) with HIV/AIDS from two HIV service agencies in a large metropolitan city in the southeastern United States. The mean self-reported adherence was 83.1% (SD ⫽ 15.7%). Adherence was significantly correlated with perceived support and absence of barriers, strong intentions to adhere, perceived effectiveness of the medications, higher levels of perceived general health, fewer years of HIV disease, and fewer years on antiretroviral medications. Existential well-being (e.g., viewing life as positive and having meaning) was a weak significant correlate. Backward regression analysis was conducted to identify a parsimonious model of predictors of adherence. The final model included presence of support and absence of barriers, fewer years of HIV infection, no current alcohol use, perceived severity of HIV, existential well-being, and male gender. This model explained 19.4% of the variance in adherence (observed F[6, 100] ⫽ 5.6, p ⬍ .0001). Key words: adherence, antiretroviral medications, HAART, spirituality

Highly active antiretroviral therapy (HAART) has changed the natural history of HIV/AIDS from a

disease characterized by a rapid downhill progression to a chronic illness. However, evidence indicates that poor adherence to antiretroviral drug therapy is a major problem that has the potential to diminish effective viral suppression, promote viral resistance, and place patients at risk for hospitalization and opportunistic infections (Paterson et al., 2000). Over the past several years, researchers have identified several factors associated with adherence to antiretroviral medications. Depression (Catz, Kelly, Bogart, Benotsch, & McAuliffe, 2000; Holzemer et al., 1999), severity of side effects, (Catz et al., 2000; Duran et al., 2001; Max, & Sherer, 2000), perceived stress and pessimism about HIV (Chesney, 1997), and inadequate coping mechanisms (Chesney, 1997; Singh et al., 1999) have been associated with lower levels of adherence. Having a sense of positive personal well-being and meaning to one’s life, or as Holzemer et al. (1999) termed it, “cherishing the environment,” had a positive association with adherence. Social support (Catz et al., 2000; Duran et al., Marcia K. McDonnell Holstad, DSN, RN, C, FNP, is assistant professor, Nell Hodgson Woodruff School of Nursing, Emory University. James C. Pace, DSN, RN, MDiv, ANP-CS, is professor of nursing, Vanderbilt University School of Nursing. Anindya K. De, PhD, MS, MStat, is senior service fellow, epidemiology program office, Centers for Disease Control and Prevention. Darla R. Ura, MA, RN, APRN, BC, is clinical associate professor, Nell Hodgson Woodruff School of Nursing, Emory University.

JOURNAL OF THE ASSOCIATION OF NURSES IN AIDS CARE, Vol. 17, No. 2, March/April 2006, 4-15 doi:10.1016/j.jana.2006.01.002 Copyright © 2006 Association of Nurses in AIDS Care

McDonnell Holstad et al. / Factors Associated With Adherence to ART

2001), self-efficacy (Catz et al., 2000; Gifford et al., 2000), positive relationships with one’s care provider (Bakken, Holzemer, Brown, Powell-Cope, Turner et al., 2000; Freeman, Rodriguez, & French, 1996), and adequate coping (Singh et al., 1999) have also been associated with higher adherence. Inconsistent relationships have been noted between adherence and sociodemographic characteristics. For example, Gifford et al. (2000); Gordillo, Del Amo, Soriano, & Gonzalex-Lahoz (1999); and Ironson et al. (1999) identified that higher levels of education were associated with better adherence, but Singh et al. (1996, 1999) and Catz et al. (2000) identified no significant associations. Younger age (Chesney, 1997; Duran et al., 2001) or middle age (Gordillo et al., 1999) was associated with adherence in some studies. Although gender has not been consistently associated with adherence, some researchers have identified that women miss more doses of antiretrovirals than men (Laine et al., 2000; Ohmit et al., 1998; Wenger et al., 1999). Abuse of alcohol and recreational drugs has been associated with both adequate and poor antiretroviral medication adherence (Duran et al., 2001; Holzemer et al., 1999; Ostrop, Hallett, & Gill, 2000). Spirituality has been studied in relation to immune function (Woods, Antoni, Ironson, & Kling, 1999), psychological well-being (Coleman & Holzemer, 1999) and quality of life (Sowell et al., 2000; Tuck, McCain, & Elswick, 2000), but few investigators have studied this variable with regard to antiretroviral adherence. A previous study (Singh et al., 1996) identified no association between religious support and adherence to antiretroviral therapy. In addition, self-care ability has not been well studied in patients with HIV/AIDS (Chou & Holzemer, 2004) and in particular with respect to adherence. It has been studied in relation to adherence in elders (Pavlides, 1993; Weseman, 1991), Tuberculosis patients (McDonnell, Turner, & Weaver, 2001), and diabetics (Haynes, 1987) with mixed results. Although data are available on some of the factors associated with adherence to HAART, it is important to continue to identify relevant factors so that patients at risk for nonadherence can be identified and adherence-promoting interventions developed. In addition, since spirituality has been shown to correlate with other aspects of HIV care, the influence of


spirituality on adherence is important to investigate. This correlational study attempted to validate the results of others on the relationship of sociodemographic characteristics, patient perceptions, and patient characteristics on adherence in a low-income, predominately minority population. It also explored the relationship of the dimension of spirituality to self-reported adherence to HAART in this group.

Conceptual Framework The conceptual framework for the study merged Orem’s self-care deficit theory of nursing (Orem, 1991) with the model of adherence developed by DiMatteo and DiNicola (1982) and Gritz, DiMatteo, and Hays (1989). Orem’s self-care theory is based on the premise that individuals or their agents will perform the activities necessary to care for themselves. The capacity and ability to care for oneself is called self-care agency. Self-care agency, according to Orem, is learned and requires decision-making and deliberate action. The development of self-care agency is dependent on many factors. An individual moves through three self-care operations as he or she moves toward self-care behaviors. The conceptual model of adherence as described by DiMatteo & DiNicola (1982) and Gritz et al. (1989) is based on a social-psychological perspective that views influence as the essential condition for behavior change. Elements from other theoretical and conceptual models are reflected in this model. At the core of the model is Fishbein’s model of reasoned action, which states that intentions are predictors of behavior and that intentions are influenced by one’s attitudes toward the behavior and the feelings of significant others about the behavior (subjective norms). Components from the health belief model include personal attitudes and beliefs that directly affect one’s intention to adhere. In the combined framework for this study, medication adherence is viewed as a self-care behavior that is associated with and affected by characteristics of the health problem, the patient, and the health care system. Variables that were measured in the study correspond to the basic conditioning factors proposed by Orem (1991) (factors that are internal or external to the individual that may affect medication adher-


JANAC Vol. 17, No. 2, March/April 2006

Figure 1. A Conceptual-theoretical-empirical structure for adherence to antiretroviral therapy as a self-care behavior

ence), the antecedents proposed by DiMatteo et al. (1983) (perceptions about HIV and its treatment), and the capacity to care for oneself. Thus, in this framework, adherence is viewed as a self-care behavior that is the result of incorporating recommended therapeutic activities, for example medication-taking, into one’s lifestyle. The decision to adhere is influenced by the relationship with the health care professional and significant others in addition to the person’s attitudes, beliefs, perceptions, and spirituality. The framework is depicted in Figure 1.

Methods This correlational study was implemented at two HIV/AIDS service organizations in a large metro-

politan city in the southeastern United States. One agency provided a full range of both clinical care and mental health care, as well as social services for clients with a diagnosis of AIDS. The other agency provided case management services and clinical care for clients with CD4 counts over 200. The study was approved by the institutional review board of a large medical center and received approval by separate research committees from each of the agencies. Agency staff prescreened potential participants for their interest and eligibility to participate in the study. Patients were eligible if they spoke English and had been on HAART for a minimum of 1 month. Women were purposely oversampled to reflect the increasing numbers of HIV/AIDS cases in this group. If interested, participants were scheduled for an in-

McDonnell Holstad et al. / Factors Associated With Adherence to ART Table 1.


Demographic Characteristics (%) (N ⴝ 120) (Number)/Percentage




Sexual preference



Religion Alcohol


Male Female Transgender African American White Hispanic Asian-Pacific Islander Others ⬍12th grade High school/GED Some college College degree Heterosexual Homosexual/gay Bisexual Less than $11,000 $11,000-$20,999 $21,000-$29,999 $30,000 or more Unemployed Employed Disabled Christian Other None Past use Weekly Occasional None Past use Daily Occasional

(72) 60 (42) 35 (6) 5 (79) 66.4 (32) 26.9 (3) 2.5 (1) .8 (4) 3.4 (29) 24.2 (23) 19.2 (39) 32.5 (29) 24.1 (49) 41.5 (55) 46.6 (14) 11.9 (69) 59.5 (29) 25.0 (14) 12.1 (4) 3.4 (44) 36.7 (54) 45.0 (22) 18.3 (112) 93.4 (8) 6.4 (51) 43.6 (8) 6.8 (6) 5.1 (52) 44.4 (87) 73.1 (15) 12.6 (4) 3.4 (13) 10.9

Differences in Adherence (p Value)a .65 .50








NOTE: GED ⫽ general educational development [credential]. a. Student’s t-test for differences in level of antiretroviral general adherence scores.

terview with a member of the research team to obtain informed consent. After informed consent was obtained, paper-and-pencil questionnaires were administered in a private area at each agency, and reading assistance was provided if needed. Participants were reimbursed $20 for their time. Setting and Sample The demographic characteristics of the sample are found in Table 1. The sample consisted of 120 English-speaking adults with HIV/AIDS who received their health care at one of two HIV-specific service

agencies. All participants had been on HAART for at least 1 month. Participants ranged in age from 21 to 55 years (mean: 36.5 years, SD ⫽ 8.5), and they had been HIV positive for an average of 5.8 years (SD ⫽ 4.0). They reported a wide range of years on antiretroviral therapy with a mean of 3.1 years (SD ⫽ 2.5); the mean duration of their current regimen was 16.7 months (SD ⫽ 15.9). On average, they rated their general health status as 2 on a scale of 1 (healthy) to 6 (unhealthy). The majority of participants were men, but because of purposeful oversampling, 35% were women. For the year data were collected (2000), 25%


JANAC Vol. 17, No. 2, March/April 2006

Table 2.

Summary Statistics for Questionnaires (N ⴝ 120) Scale

Spiritual involvement and beliefs Existential well-being subscale Religious well-being subscale Spiritual well-being scale Self as carer Interpersonal aspects of care Perceived utility Perceived severity Perceived susceptibility Intentions to adhere Supports/barriers Subjective norms

Items and Range

Sample Range



Score (%)

20 20-134 10 10-60 10 10-60 20 0-120 40 40-240 8 8-40 8 8-40 4 4-20 4 4-20 4 4-20 4 4-20 6 ⫺18 - ⫹18













































⫺4 - ⫹18




NOTE: SD ⫽ standard deviation.

of AIDS cases were women (Centers for Disease Control and Prevention, 2001). Five percent of the sample (n ⫽ 6) described themselves as transgendered. Slightly over 66% were African American. About 41% were heterosexual, 46% homosexual/ gay, and about 12% bisexual. Over half were educated beyond high school and either unemployed or disabled. Over half had an annual income of less than $11,000. Almost half reported alcohol use, mostly on an occasional basis, and most did not currently use drugs. The majority were Christian, with the greatest number Baptists, a predominant denomination in the southern United States. The participants were regular churchgoers; the median time since they had last attended a religious service was 2 weeks. Instruments The instruments chosen for the study reflect the variables noted in the conceptual framework in Fig-

ure 1. Summary statistics for each instrument and subscales are presented in Table 2. The Antiretroviral General Adherence Scale (AGAS) was used to measure self-reported general tendency to adhere to the HAART regimen. The AGAS was adapted from the Tuberculosis General Adherence Scale (McDonnell et al., 2001) and is composed of five items that focus on the ease and ability of taking antiretroviral medications as the health care provider recommended. All responses were on a 6-point Likert scale from none of the time to all of the time. Four items focused on ease and ability of taking antiretroviral medications as the health care provider recommended, and a final item asked the participant to rate how often in general he or she was able to take the medications as recommended in the past 4 weeks. To facilitate comparison of adherence, raw scores can be linearly transformed to a percentage score by dividing the achieved score by the maximum possible score. The

McDonnell Holstad et al. / Factors Associated With Adherence to ART

alpha reliability coefficient for the AGAS was .74 for this sample. The Self as Carer Inventory (SCI), a 40-item selfreport questionnaire, was used to measure the participant’s perceived ability to care for himself or herself and the amount of self-care he or she provided (Geden & Taylor, 1991). Scores ranged from 40 to 240, with low scores indicating high perceived self-care ability. The alpha coefficient for this questionnaire was .95 for this sample. Spirituality was measured using two instruments: The Spiritual Well-Being Scale (SWB) by Paloutzian and Ellison (1982) and the Spiritual Involvement and Beliefs Scale (SIBS) developed by Hatch, Burg, Naberhaus, and Hellmich (1998). Each scale measures a slightly different dimension of spirituality. The SWB assesses one’s spiritual quality of life, whereas the SIBS assesses one’s spiritual beliefs and attitudes and includes questions about specific spiritual activities such as prayer and meditation. The SWB is a 20-item questionnaire, and each item is rated on a 6-point Likert scale. The possible range is 0 to 120, with higher scores indicating higher levels of spiritual well-being. The SWB contains two subscales: the Existential Well-Being Scale (EWB) and the Religious Well-Being Scale (RWB). The EWB subscale focuses on life purpose and satisfaction, and the RWB subscale focuses on one’s relationship with God. Previous test-retest reliability coefficients using a student sample were .93 (SWB), .96 (RWB), and .78 (EWB) (Paloutzian, 1983; Paloutzian, & Ellison, 1982). Coefficient alphas for the total SWB score and for the RWB and EWB subscales from the present sample were .92, .92, and .89, respectively. The SIBS developed by Hatch et al. (1998) assesses spiritual beliefs and levels of involvement in certain activities such as prayer and meditation. The 20-item instrument yields a score from 20 to 134; the higher the score, the higher the level of spiritual belief and involvement. In this study the alpha coefficient was .89, which is consistent with that of the original authors. The Antiretroviral Adherence Determination Questionnaire (AADQ), a 38-item scale modified from the Tuberculosis Adherence Determination Questionnaire (McDonnell et al., 2001), measured perceptions of seven theoretical determinants of adherence. The seven subscales are interpersonal as-


pects of care; perceived severity of HIV/AIDS (PSV); perceived susceptibility to AIDS (PS); perceived utility of the treatment plan (PU) such as costs, benefits, efficacy; subjective norms (SN) or the influence of family, friends, and others on the decision to adhere; intentions to adhere to the treatment plan (INT); and presence of supports and absence of barriers to adherence (SB). Responses range from 1 (strongly disagree) to 6 (strongly agree). In this study, alpha reliability coefficients for the subscales ranged from .54 to .82, with PSV, PS, and SB having coefficients below .7, which is comparable to previous reports for other populations (DiMatteo et al., 1993; McDonnell et al., 2001). Data about the participants’ sociodemographics, current living conditions, religion, alcohol and drug use, HIV infection, and antiretroviral treatment were also determined.

Results Adherence Rates Self-reported antiretroviral adherence scores ranged from a low of 10 (33.3%) to a perfect 30 (100%) with a mean score of 24.9 (SD ⫽ 4.7) or 83%. Forty-three percent of the sample (52) responded all of the time to the last item, which asked, in general, how often they were able to take their medications as prescribed over the last 4 weeks. These scores indicated moderate levels of adherence and ease of taking the antiretroviral medications. Analysis of variance and Student’s t-tests were performed to examine differences in adherence scores based on gender, income, education, ethnicity, and HIV service agency. For these analyses, alcohol and drug use were dichotomized into categories of current use and no current use; income was classified as less than $11,000 per year and greater than or equal to $11,000 per year; education was classified as less than high school or above high school; and gender was categorized as male or female (including the 6 transgendered participants). For the analysis by race/ethnicity only, African American and White/ Caucasian were used because of the small number of participants of other races (n ⫽ 8). The p values for these analyses are displayed in Table 1. There were


JANAC Vol. 17, No. 2, March/April 2006

no significant differences in adherence based on any of the sociodemographic characteristics. Although not statistically significant, a closer examination of the adherence scores by gender, sexual preference, and substance use yields some interesting differences. When compared by gender, women had a slightly lower mean percentage score of 82.5% than men (mean score 83.8%), and the 6 transgendered persons had the lowest mean adherence score of 77.8%. When compared by sexual preference, bisexuals had the highest mean score (88.0%), followed by heterosexuals (mean score 84.6%), and homosexuals (mean score 80.9%). Current alcohol users had lower mean AGAS scores (mean score 81.3%) than nondrinkers (mean score 85%), and current drug users also had slightly lower mean scores (82.7%) than non-drug users (83.2%). The data in Table 2 summarize statistics for the seven subscales of the AADQ, the SCI, SIBS, SWB and its subscales RWB and EWB. Based on the mean scores, participants were satisfied with the interpersonal aspects of their care (i.e., communication, concern, trust). They expressed a very good perception of the effectiveness of the medication (PU). They thought their HIV disease was moderately severe, and they had a moderate perceived susceptibility to AIDS. Participants reported strong intentions to take their medications and felt they had the support needed to do so. Despite this, however, family, friends, and relatives had little influence on their ability to follow their HAART regimen (SN). As noted in Table 2, participants reported high levels of self-care capacity on the SCI scale. The last item of this questionnaire is a rating of the amount of care provided by oneself. Based on the mean of 1.6 on a scale of 1 (all) to 6 (none), these participants provided almost all of their own care. The high mean scores on both spirituality instruments indicated a high self-reported level of spiritual involvement and belief and spiritual well-being. Correlations between the variables are presented in Table 3. As anticipated, the Spiritual Involvement and Beliefs Scale was highly correlated with the Spiritual Well-Being Scale (r ⫽ 0.68, p ⬍ .01) and its components religious well-being (r ⫽ .71, p ⬍ .01) and existential well-being (r ⫽ .473, p ⬍ .01). There was a weak positive correlation between ad-

herence scores and existential well-being (r ⫽ .18; p ⫽ .055). Self-reported adherence scores were significantly positively correlated with SB (r ⫽ .34, p ⬍ .01), INT (r ⫽ .19, p ⬍ .05), and PU (r ⫽ .25, p ⬍ .01). Adherence was significantly negatively correlated with years HIV-infected (r ⫽ ⫺0.27, p ⬍ .01), years on antiretroviral medications (r ⫽ ⫺0.21, p ⬍ .05), and perceived health status (r ⫽ ⫺.20, p ⬍ .05). As a result, in this sample the longer the persons had HIV disease or were taking antiretroviral medications, the less adherence they reported. Because perceptions of health status were scored on a 1 (healthy) to 6 (unhealthy) scale with a lower score indicating better health, the negative correlation with health status actually indicates that a high level of perceived general health was associated with a high level of adherence. Variables that had been significantly associated with self-reported adherence (i.e., years HIV-infected, years on antiretroviral medications, perceived health status, supports/barriers, intentions, perceived utility, and existential well-being), theoretical variables from the AADQ (i.e., interpersonal aspects of care, perceived severity, perceived susceptibility, subjective norms) and those that had been reported in others’ work to be associated with adherence (i.e., self-care capacity, current drug use, current alcohol use, and gender) were entered into a backward-regression equation. For this analysis, gender was dichotomized to male/nonmale. The final model had the highest adjusted R2, was parsimonious, and was consistent with results from other studies. This model, displayed in Table 4, accounted for 19.4% of the variance in adherence (observed F[6, 100] ⫽ 5.6, p ⬍ .0001).

Discussion The resulting six-factor model accounting for 19.4% of the variance in adherence was consistent with the conceptual framework, with the exception of self care ability and the only system factor, interpersonal aspects of care. Important sociodemographic factors included male gender, current alcohol use, and years of HIV infection; the latter two of which had a negative influence on adher-

Table 3.

Correlation Matrix of Variables (N range: 106-120)



.10 .18 ⫺.10 .01 ⫺.15 .09 .25b .06 ⫺.05 .19a .34b .14 ⫺.20a

.47b .71b .68b ⫺.38b .22a .30b ⫺.05 ⫺.29b .23a .26b .08 ⫺.09

.51b .86b ⫺37b .24b .38b ⫺.29b ⫺.39b .30b .27b .02 ⫺.38b

.87b ⫺.29b .15 .28b ⫺.17 ⫺.27b .18 .12 .10 ⫺.11

⫺.36b .23a .37b ⫺.26b .38b .23a .21a .07 ⫺.32b







⫺.27b ⫺.21a

⫺.14 ⫺.05

⫺.02 ⫺.04

⫺.16 ⫺.10

⫺.08 ⫺.11

.11 .06







⫺.40b ⫺.32b .59b .18 ⫺.20a ⫺.15 .29b ⫺.25b ⫺.27b .46b ⫺.17 .42b .75b ⫺.03 ⫺.05 b b ⫺.38 .26 .43b ⫺.12 ⫺.21a .37b ⫺.28b .21a .32b ⫺.30 ⫺.11 .21a .33b .33b ⫺.20a ⫺.27b .24b .35b ⫺.12 ⫺.20a .31b ⫺.16


.20a ⫺.18a .09 ⫺.07 ⫺.12 ⫺.09



.07 ⫺.04

⫺.23a ⫺.15

.20a .06 .09 ⫺.05

⫺.15 ⫺.15

⫺.10 ⫺.01


.11 0.76

.12 .01


NOTE: AGAS ⫽ Antiretroviral General Adherence Scale, ARV ⫽ antiretroviral, EWB ⫽ Existential Well-Being Scale, INT ⫽ intentions to adhere, IPAC ⫽ interpersonal aspects of care, PU ⫽ perceived utility, PS ⫽ perceived susceptibility, PSV ⫽ perceived severity, RELWB ⫽ religious well-being, SB ⫽ supports/barriers, SCI ⫽ self as carer, SIBS ⫽ Spiritual Involvement and Beliefs Scale, SN ⫽ subjective norms, SWB ⫽ spiritual well-being. a. p ⬍ .05 b. p ⬍ .01

McDonnell Holstad et al. / Factors Associated With Adherence to ART

SIBS EWB RELWB SWB SCI IPAC PU PSV PS INT SB SN Health Status Amount of self-care Time HIVinfected (yr) Time on ARV medications (yr)


Time Time on Amount HIVARV Heath of Self- Infected Medications Status Care (yr) (yr)



JANAC Vol. 17, No. 2, March/April 2006

Table 4.

Final Adherence Regression Model (n ⴝ 115)

Independent Variables

Standardized Beta


Supports/barriers Years HIV-infected Current alcohol use Perceived severity of HIV Existential wellbeing Male

.292 ⫺.231 ⫺.177

.001 .009 .065



.146 .121

.104 .210

Adjusted R2


ence. Influential patient characteristics and perceptions included the perceived presence of supports and the absence of barriers to adherence, the perceived severity of HIV/AIDS, and the degree of existential well-being. The moderate mean score of self-reported general adherence reported in this study— 83.1% for the preceding 4 weeks—is slightly higher than the 74% rate over 6 months reported in a previous study (Hecht, Colfax, Swanson, & Chesney, 1998). This rate would be considered acceptable for most chronic diseases (Mehta, Moor, & Graham, 1997); however, based on a previous report (Paterson et al., 2000) rates of 95% or greater are needed to maintain viral load suppression and prevent the occurrence of opportunistic infections and hospitalization. Thus, patients must be perfect or near-perfect pill-takers to remain “healthy” with HIV disease. The 42 women in this study had slightly lower mean adherence rates than men; however, these differences are small and not statistically significant. Transgendered persons (n ⫽ 6), although too small a subsample on which to generalize, had the lowest adherence rates. These results suggest that special adherence needs might exist for each gender including transgendered persons (Schilder, Laframboise, Hogg, Goldstone, Trussler et al., 1998). The presence of supports and the absence of barriers to medication-taking and the number of years of HIV infection had the strongest influence in the model. Social support has been consistently associated with adherence to HAART and a variety of medications and in a variety of medical conditions. The negative influence of the number of years of HIV disease on adherence levels suggests that those who

have lived longer may have learned what to do to live, how to adapt, and how to manipulate their medication regimen to maintain their health. Although it did not enter into the final regression, the number of years on antiretroviral medications was positively correlated with the number of years of HIV infection (r ⫽ .46, p ⬍ .0001) and negatively correlated with adherence; this result lends credence to this idea that the longer one has the disease and takes HAART, the more adherence tends to wane. In addition, it has also been noted in the general adherence literature that the longer one takes any medication, the more difficult it is to adhere (Meichenbaum & Turk, 1987). It is also possible that a subset of our patients might not have needed to take the prescribed dose, in total, to maintain viral suppression, or they may have possessed other characteristics, such as a low density of CCR5 coreceptors on their CD4 cells, that kept them healthy (Reynes et al., 2001). In general, the sample mean PSV score showed only a moderate perception of the severity of HIV/ AIDS; however, the regression suggested that perception of severity of illness might be a positive motivator for medication adherence. Adherence was correlated with the perception of health, so the overall effect of HAART in improving health may be a reinforcer to continue to adhere. The capacity to care for oneself has not been previously studied in relation to adherence to HAART, although patients with HIV/AIDS have complaints of fatigue and weakness, and many are on disability. Our population considered themselves relatively healthy and provided most of their own care. The results did not support a link between the ability to care for oneself and adherence to HAART. This study also explored the relationship between spirituality and religious beliefs and adherence to antiretroviral medications. The influence of existential well-being is of particular interest. Persons who loved life, found meaning in their activities, viewed life as positive, felt some hope in the future, were satisfied with life and themselves, enjoyed life, and felt a definite purpose in living (all components of the EWB scale, without “religiosity”) were more adherent. Similarly, a previous study (Holzemer et al., 1999) identified that patients who reported having a meaningful life, felt comfortable and well cared for, used their time wisely, and took time for important

McDonnell Holstad et al. / Factors Associated With Adherence to ART

things were more adherent to their medications and more likely to follow the advice of their providers. Coleman and Holzemer (1999) noted that existential well-being, a spiritual indicator of meaning and purpose, was related to the participants’ psychological well-being more than religious well-being. The participants in our study were spiritually healthy, which was associated with better antiretroviral adherence. The importance of a positive relationship with the health care provider has been positively associated with antiretroviral adherence. In this study, satisfaction with the interpersonal aspects of care had no direct relationship to adherence scores. Future research should include the development and testing of models of factors related to adherence to HAART using random sampling and multiple measures of adherence (subjective, objective, and physiological). The relationship of spirituality and, in particular, existential well-being to adherence needs further exploration. Existential well-being might be the converse of depression, and studies that examine both variables may uncover additional information about this variable. The result of lower adherence in transgendered participants warrants further study of this group’s adherence rates and factors related to adherence. Alhough limited by convenience sampling and the social desirability of self-report, the results of this study have several implications for nursing practice. Based on these results, the authors recommend that the nursing adherence assessment should include the length of time the client has been HIV-infected (and on antiretroviral medications), a sense of how severe the client feels his or her HIV disease is, whether the client feels he or she has supports for or barriers to medication-taking, current alcohol use, the client’s perception of the meaning of his or her life, and gender-related issues. Nursing adherence interventions should include the provision of additional supports and elimination of barriers to taking the medication, activities to promote and enhance a positive outlook on life such as support groups and mental health referrals, and referral to substance abuse programs if deemed appropriate. Gender-appropriate support groups and social services (e.g., child care and transgender services) also might be helpful. Patient education should include a discussion regarding how serious the client


thinks his or her HIV disease is and the client’s perceived health status. Information that emphasizes the effectiveness, importance, and benefits (i.e., utility) of taking the medications for one’s health are important, as well as are methods to secure a strong intent to take the medications such as contracting. Further assessment of the meaning of spirituality and religion in the lives of those who are HIV positive offers many avenues for individualized interventions. The positive effects of existential well-being including hope, courage, stamina, increased coping abilities, increased self-esteem, and a sense of integration with life and living might contribute to an overall sense of well-being and healthfulness. Spiritual interventions might help those who struggle with disapproving religions and yet still have very spiritual ways of interacting with self, others, God, and the world.

Acknowledgement This study was funded by a research grant from the Office of Research Affairs, Nell Hodgson Woodruff School of Nursing, Atlanta, Georgia.

References Bakken, S., Holzemer, W. L., Brown, M. A., Powell-Cope, G. M., Turner, J. G., Inouye, J., et al. (2000). Relationships between perception of engagement with health care provider and demographic characteristics, health status, and adherence to therapeutic regimen in persons with HIV/AIDS. AIDS Patient Care & STDs, 4, 189-197. Catz, S. L., Kelly, J. A., Bogart, L. M., Benotsch, E. G., & McAuliffe, T. L. (2000). Patterns, correlates, and barriers to medication adherence among persons prescribed new treatments for HIV disease. Health Psychology, 19, 124-133. Centers for Disease Control and Prevention. (2001). HIV/AIDS Surveillance Report 12. Atlanta: Author. Chesney, M. (1997). Adherence to drug regimens: A learned skill. Improving the Management of HIV Disease, 5, 12. Chou, F., & Holzemer, W. (2004). Linking HIV/AIDS clients’ self-care with outcomes. Journal of the Association of Nurses in AIDS Care, 15, 58-67. Coleman, C. L., & Holzemer, W. L. (1999). Spirituality, psychological well-being, and HIV symptoms for African Americans living with HIV disease. Journal of the Association of Nurses in AIDS Care, 10, 42-50.


JANAC Vol. 17, No. 2, March/April 2006

DiMatteo, M., & DiNicola, D. (1982). Achieving patient compliance: The psychology of the medical practitioner’s role. New York: Pergamon Press. DiMatteo, M., Sherbourne, C., Hays, R., Ordway, L., Kravitz, R., McGlynn, E., et al. (1993). Physicians’ characteristics influence patients’ adherence to medical treatment: Results from the medical outcomes study. Health Psychology, 12, 93-102. Duran, S., Spire, B., Raffi, F., Walter, V., Bouhour, D., Journot, V., et al., & The APROCO Study Group. (2001). Self-reported symptoms after initiation of a protease inhibitor in HIV-infected patients and their impact on adherence to HAART. HIV Clinical Trials, 2, 38-45. Freeman, R. C., Rodriguez, G. M., & French, J. F. (1996). Compliance with AZT treatment regimen of HIV-seropositive injection drug users: A neglected issue. AIDS Education and Prevention, 8, 58-71. Geden, E., & Taylor, S. (1991). Construct and empirical validity of the self-as-carer inventory. Nursing Research, 40, 47-50. Gifford, A. L., Bormann, J. E., Shively, M. J., Wright, B. C., Richman, D. D., & Bozzette, S. (2000). Predictors of selfreported adherence and plasma HIV concentrations in patients on multi-drug antiretroviral regimens. Journal of Acquired Immune Deficiency Syndromes, 23, 386-395. Gordillo, V., Del Amo, J., Soriano, V., & Gonzalex-Lahoz, J. (1999). Sociodemographic and psychological variables influencing adherence to antiretroviral therapy. AIDS, 13, 17631769. Gritz, E., DiMatteo, M., & Hays, R. (1989). Methodological issues in adherence to cancer control regimens. Preventive Medicine, 18, 711-720. Hatch, R. L., Burg, M. A., Naberhaus, D. S., & Hellmich L. K. (1998). The Spiritual Involvement and Beliefs Scale: Development and testing of a new instrument. Journal of Family Practice, 46, 476-86. Haynes, L. (1987). The relationship between perceived exercise of self-care agency of diabetics and reported compliance to the diabetic regimen. Master’s Abstracts International, 26, 106. (UMI No. AAT 1330830) Hecht, F. M., Colfax, G., Swanson, M., & Chesney, M. A. (1998, February). Adherence and effectiveness of protease inhibitors in clinical practice [Abstract No. 151]. Paper presented at the Fifth Annual Conference on Retroviruses and Opportunistic Infections, Chicago. Holzemer, W. L., Corless, I. B., Nokes, K. M., Turner, J. G., Brown, M. A., Powell-Cope, G. M., et al. (1999). Predictors of self-reported adherence in persons living with HIV disease. AIDS Patient Care & STDs, 13, 185-97. Ironson, G., Solomon, G. F., Balbin, E. G., O’Cleirigh, C., George, M.A., Kumar, M., et al. (1999). Adherence to combination antiretroviral therapies in HIV patients of low health literacy. Journal of General Internal Medicine, 14, 267-273. Laine, C., Newschaffer, C. J., Zhang, D., Cosler, L., Hauck, W., & Turner, B. (2000) Adherence to antiretroviral therapy by pregnant women infected with human immunodeficiency virus: A pharmacy claims-based analysis. Obstetrics & Gynecology, 95, 167-173.

Max, B., & Sherer, R. (2000). Management of the adverse effects of antiretroviral therapy in medication adherence. Clinical Infectious Diseases, 30(Supplement 2), S96-S116. McDonnell, M., Turner, J., & Weaver, M. (2001). Antecedents of adherence to antituberculosis therapy. Public Health Nursing, 18, 392-400. Mehta, S., Moor, R., & Graham, N. (1997). Potential factors affecting adherence with HIV therapy. AIDS, 11, 1665-1670. Meichenbaum, D., & Turk, D. (1987). Facilitating treatment adherence. New York: Plenum Press. Ohmit, S., Schuman, P., Schoenbaum, E., Rompalo, A., Cohen, M., Richardson, J., et al. (1998, June). Adherence to antiretroviral (ART) among women in the HIV Epidemiology Research Study (HERS) and Women’s Inter-Agency HIV Study (WIHS) [Abstract 32347]. Paper presented at the International Conference on AIDS, Geneva, Switzerland. Orem, D. (1991). Nursing: Concepts of practice. St. Louis, MO: Mosby. Ostrop, N. J., Hallett, K. A., & Gill, M. J. (2000). Long-term patient adherence to antiretroviral therapy. Annals of Pharmacotherapy, 34, 703-709. Paloutzian, R. (1983). Invitation to a psychology of religion. Glenview: CA, Scott, Foresman. Paloutzian, R., & Ellison, C. (1982). Loneliness, spiritual wellbeing and quality of life. In: L. Peplau, D. Perlman (Eds.). Loneliness: A Sourcebook of current theory, research, and therapy. New York: John Wiley. Paterson, D. L., Swindells, S., Mohr, J., Brester, M., Vergis, E., Squier, C., et al. (2000). Adherence to protease inhibitor treatment and outcomes in patients with HIV infection. Annals of Internal Medicine, 133, 21-30. Pavlides, C. (1993). The relationship among complexity of medication, functional ability, and adherence to prescribed medication regimen in the homebound older adult. Dissertation Abstracts International, 54, 1336. (UMI No. ATT 93201172) Reynes, J., Portales, P., Segondy, M., Baillat, V., Andre, P., Odile, A., et al. (2001). CD4 T cell surface CCR5 density as a host factor in HIV-1 disease progression. AIDS, 15, 16271634. Schilder A. J., Laframboise, S., Hogg, R. S., Goldstone, I., Trussler, T., Schecter, M. T., et al. (1998, June). The health care experiences of HIV- positive transgendered persons can influence adherence to HIV antiretroviral therapy [Abstract 32395]. Paper presented at the International Conference on AIDS, Geneva, Switzerland. Singh N., Berman S. M., Swindells S., Justis J. C., Mohr J. A., Squier C., et al. (1999). Adherence of human immunodeficiency virus-infected patients to antiretroviral therapy. Clinical Infectious Diseases, 4, 824-830. Singh, N., Squier, C., Sivek, C., Wagener, M., Nguyen, M. H., & Yu, V. L. (1996). Determinants of compliance with antiretroviral therapy in patients with human immunodeficiency virus: Prospective assessment with implications for enhancing compliance. AIDS Care, 8, 261-269. Sowell, R., Moneyham, L., Hennessy, M., Guillory, J., Demi, A., & Seales, B. (2000). Spiritual activities as a resistance re-

McDonnell Holstad et al. / Factors Associated With Adherence to ART source for women with human immunodeficiency virus. Nursing Research, 49, 73-82. Tuck, I., McCain, N. L., Elswick, R. K. (2000). Spirituality and psychosocial factors in persons living with HIV. Journal of Advanced Nursing, 33, 776-783. Wenger, N., Gifford, A., & Liu, H., Chesney, M., Golin, C., Crystal, S., et al. (1999, January). Patient characteristics and attitudes associated with antiretroviral adherence [Abstract 98]. Paper presented at the Sixth Annual Conference on Retroviruses and Opportunistic Infections, Chicago. Weseman, R. J. (1991). Older adults: The relationship between medication adherence and the variables of self-care agency, functional status, and medication regimen complexity. Master’s Abstracts International, 30, 101. (UMI No. ATT 1344870) Woods, T., Antoni, M., Ironson, G., & Kling, D. (1999). Religiosity is associated with affective and immune status in symptomatic HIV-infected gay men. Journal of Psychosomatic Research, 46, 165-176.