FACTORS INFLUENCING MEDICATION ADHERENCE BELIEFS AND SELF-EFFICACY

P1: IZO AIDS and Behavior (AIBE)

PP1211-aibe-486486

June 2, 2004

9:51

Style file version June 22, 2002

C 2004) AIDS and Behavior, Vol. 8, No. 2, June 2004 (°

Factors Influencing Medication Adherence Beliefs and Self-Efficacy in Persons Naive to Antiretroviral Therapy: A Multicenter, Cross-Sectional Study Nancy R. Reynolds,1,7 Marcia A. Testa,2 Linda G. Marc,2 Margaret A. Chesney,3 Judith L. Neidig,1 Scott R. Smith,4 Stefano Vella,5 and Gregory K. Robbins6 for the Protocol Teams of ACTG 384, ACTG 731, and A5031s Received Dec. 6, 2002; revised Aug. 18, 2003; accepted Oct. 17, 2003

It is widely recognized that adherence to antiretroviral therapy is critical to long-term treatment success, yet rates of adherence to antiretroviral medications are frequently subtherapeutic. Beliefs about antiretroviral therapy and psychosocial characteristics of HIV-positive persons naive to therapy may influence early experience with antiretroviral medication adherence and therefore could be important when designing programs to improve adherence to antiretroviral therapy. As part of a multicenter AIDS Clinical Trial Group (ACTG 384) study, 980 antiretroviral-naive subjects (82% male, 47% White, median age 36 years, and median CD4 cell count 278 cells/mm3 ) completed a self-administered questionnaire prior to random treatment assignment of initial antiretroviral medications. Measures of symptom distress, general health and well-being, and personal and situational factors including demographic characteristics, social support, self-efficacy, depression, stress, and current adherence to (nonantiretroviral) medications were recorded. Associations among variables were explored using correlation and regression analyses. Beliefs about the importance of antiretroviral adherence and ability to take antiretroviral medications as directed (adherence self-efficacy) were generally positive. Fifty-six percent of the participants were “extremely sure” of their ability to take all medications as directed and 48% were “extremely sure” that antiretroviral nonadherence would cause resistance, but only 37% were as sure that antiretroviral therapy would benefit their health. Less-positive beliefs about antiretroviral therapy adherence were associated with greater stress, depression, and symptom distress. More-positive beliefs about antiretroviral therapy adherence were associated with better scores on health perception, functional health, social–emotional–cognitive function, social support, role function, younger age, and higher education (r values = 0.09–0.24, all p < .001). Among the subset of 325 participants reporting current use of medications (nonantiretrovirals) during the prior month, depression was the strongest correlate of nonadherence (r = 0.33, p < .001). The most common reasons for nonadherence to the medications were “simply forgot” (33%), “away from home” (27%), and “busy” (26%). In conclusion, in a large, multicenter survey, personal and situational factors, such as depression, stress, and lower education, were associated with less certainty about the potential for antiretroviral therapy effectiveness and one’s perceived ability to adhere to therapy. Findings from these analyses suggest a role for baseline screening for adherence

1 Ohio

5 Istituto

2 Harvard

6 Massachusetts

Superiore di Sanita, Rome, Italy. General Hospital, Boston, Massachusetts. 7 Correspondence should be directed to Nancy R. Reynolds, Ohio State University, 1585 Neil Ave., Columbus, Ohio 43210 e-mail: [email protected].

State University, Columbus, Ohio. School of Public Health, Boston, Massachusetts. 3 University of California, San Francisco, San Francisco, California. 4 University of North Carolina, Chapel Hill, Chapel Hill, North Carolina.

141 C 2004 Plenum Publishing Corporation 1090-7165/04/0600-0141/0 °


PP1211-aibe-486486

June 2, 2004

9:51


142

Reynolds et al. predictors and focused interventions to address modifiable factors placing persons at high risk for poor adherence prior to antiretroviral treatment initiation. KEY WORDS: AIDS; HIV; adherence; beliefs; self-efficacy.

INTRODUCTION The efficacy of potent regimens combining antiretroviral medications offers the possibility of dramatic clinical improvement and prolonged life for persons infected with the human immunodeficiency virus (HIV) (Karon et al., 2001; Mannheimer et al., 2002; Murphy et al., 2001a; Palella et al., 1998). Although it is widely recognized that adherence to antiretroviral regimens is critical to treatment success, rates of adherence (proportion of prescribed number of pills taken) are frequently suboptimal (Bangsberg et al., 2000; Nieuwkerk et al., 2001; Paterson et al., 2000). As many as half of the individuals on combination therapies fail to take antiretroviral medication in accordance with dosage, time, and dietary instructions (Murphy et al., 2001b; Nieuwkerk et al., 2001). There is a substantial body of research exploring the problem of adherence to antiretroviral medications. Numerous studies have attempted to quantify measures of adherence, identify factors contributing to poor adherence, and determine the effectiveness of interventions for improving adherence (Chesney et al., 2000a; Haddad et al., 2002; Ickovics and Meade, 2002). Little is known of factors predictive of adherence in persons naive to antiretroviral therapy. It is, however, increasingly understood that successful long-term therapy requires durable, first-line antiretroviral treatment (ART) regimens (Volberding, 2002; Yeni et al., 2002). While adherence is vital to treatment response, it has been found repeatedly that health care providers are not able to accurately predict which individuals will successfully adhere to their antiretroviral regimens (Bangsberg et al., 2002) so that they might be targeted for early intervention. The problem of targeting persons at high risk for low adherence is even more difficult and complex for persons naive to therapy. For these individuals without an adherence track record, related behaviors, beliefs, and personal and situational characteristics might prove to be informative predictors of future antiretroviral therapy adherence. If such a set of predictors could be identified, they could be used to select persons at risk for suboptimal adherence who then could be targeted for intervention to reduce the likelihood of subtherapeutic adherence before therapy is initiated.

The health belief model (Becker, 1974, 1988; Janz and Becker, 1984; Montgomery et al., 1989) proposes that a strong internal commitment to a health-related behavior influences the ability of an individual to adopt and incorporate that behavior in his or her daily functioning. According to this model, two major factors influence the likelihood that individuals will adopt a recommended health protective behavior, namely (1) they must feel susceptible to a disease with serious or severe consequences and (2) they must believe that the benefits of taking the preventive action outweigh the perceived barriers. In this regard, a strong belief in one’s ability to adhere to antiretroviral medications (antiretroviral therapy adherence self-efficacy) could potentially improve the individual’s ability to comply with the therapeutic regimen. Furthermore, factors that the individual identifies as barriers to high adherence behavior could be linked to modifiable behavioral and situational characteristics and targeted by pretreatment counseling programs. We recently presented data showing that higher antiretroviral adherence self-efficacy is associated with higher antiretroviral therapy adherence, and that higher antiretroviral therapy adherence is associated with a more favorable therapeutic response (Reynolds et al., 2003). To evaluate those factors that might influence antiretroviral therapy adherence selfefficacy and barriers to adherence, we examined the baseline correlates of self-reported medication adherence history and attitudes and beliefs toward antiretroviral adherence in 980 persons naive to therapy at the baseline visit for a multicenter clinical trial.

METHODS Design A cross-sectional analysis of self-administered baseline survey data was conducted. Participants were enrolled in the AIDS Clinical Trials Group (ACTG) protocol 384. Data were collected prior to participant knowledge of random treatment assignment of initial antiretroviral therapy. ACTG 384 (design and selection criteria have been described fully previously; Smeaton et al., 2001) was a randomized, partially double-blinded, controlled trial with 980 participants


PP1211-aibe-486486

June 2, 2004

9:51


Factors Influencing Medication Adherence Beliefs from 58 U.S. (n = 896) and 23 Italian (n = 82) centers. The trial was designed to evaluate different strategies of initiating antiretroviral treatment in HIV-1infected individuals (HIV-1 RNA copies ≥500/mL) with less than 7 days of previous antiretroviral therapy. The research protocol was reviewed and approved by the institutional review board of each of the participating centers. Written informed consent was obtained from study participants.

143 a measure of different dimensions of self-perceived functioning and well-being, including perceived ability to work and perform daily role (Role Functioning), perceived health functioning (Functional Health), perceived vitality, memory and reasoning, and affect (Social-Emotional-Cognitive Functioning), and a self-rating of perceived general health (Health Perception). Statistical Analyses

Participants Of the 980 participants enrolled in ACTG 384 who completed the baseline survey, 181 (18.5%) were female. The median age was 36 years, with 22% between ages 13 and 29 years, 43% between ages 30 and 39 years, 25% between ages 40 and 49 years, and 11% at 50 years of age and older. The study population was ethnically diverse, with 47% White, 35% Black, and 17% Hispanic, 88% completed the assessment questionnaire in English, 7% in Italian, and 3% in Spanish, and 2% failed to complete the assessment. The median plasma HIV-1 RNA was 86,868 copies/mL and the median CD4 cell count was 278 cells/mm3 . Measurement Survey data were collected with the ACTG Baseline Adherence Questionnaire (Chesney et al., 2000b), the ACTG Symptom Distress Module (Justice et al., 2001), and the HIV MOS Questionnaire (Wu et al., 1991). The ACTG Baseline Adherence Questionnaire (Chesney et al., 2000b) provided a selfreport measure of personal and situational variables including (1) beliefs about the effectiveness of antiretroviral medication and confidence in one’s ability to adhere to antiretroviral medications as directed (ART Adherence Self-Efficacy), (2) satisfaction with the amount of support and anticipated help with antiretroviral therapy adherence (Social Support), (3) reasons for failure to comply with current (nonantiretroviral) prescribed medication (Non-ART Medication Adherence), (4) symptoms of depression (Depression), (5) symptoms of anxiety, coping, and well-being (Perceived Stress), (6) alcohol and illicit drug use (Alcohol Use, Illicit Drug Use), and demographic variables. The ACTG Symptom Distress Module (Justice et al., 2001) provided a measure of the number and type of symptoms commonly experienced by HIV-positive persons (HIV Symptoms). The HIV MOS Questionnaire (Wu et al., 1991) provided

To standardize the scale ranges and simplify interpretation, all Likert item-based scale ranges were converted to 0–100 with 0 indicating worst functioning or well-being and 100 indicating best functioning or well being. In addition, the subset of subjects who had taken non-ART medication during the last month were asked to rate how often during the last month they missed taking medications (never = 3, rarely = 2, sometimes = 1, often = 0) for each of 14 specific reasons (e.g., were away from home, simply forgot, felt sick/ill). From these ratings the Non-ART Medication Adherence scale was calculated, with 0 indicating the worst score (subject checked often for all 14 reasons) and 100 the best score (subject checked never for all 14 reasons). Item mean values were generated for each scale item that was unanswered by a study participant using a computerized missing replacement routine. This algorithm required a fixed number of nonmissing items for each score depending on the number of items in the scale as well as the distribution of missing values, but in all cases required at least one-half of the items be present for a scale score to be nonmissing. Means, standard deviations, frequency distributions, and Cronbach’s alpha reliabilities were calculated for each scale. Pearson correlations were used to examine relationships between continuous variables of interest. Regression (linear and logistic) techniques were used to construct the explanatory models of ART Adherence Self-Efficacy. The Type 1 error (alpha) level was set at 0.05. SPSS statistical software was used for all analyses (SPSS Inc., 2002). RESULTS Personal and Situational Characteristics and Quality-of-Life Assessments Scale profiles showed that participants had generally positive beliefs about antiretroviral therapy effectiveness and confidence in his or her ability to


PP1211-aibe-486486

June 2, 2004

9:51


144

Reynolds et al. Table I. Adherence and Quality-of-Life Constructs, Measures, Reliability, and Baseline Meansa Construct assessed

Beliefs about the effectiveness of antiretroviral medication and confidence in one’s ability to adhere (ART Adherence Self-Efficacy) Reasons for failure to comply with current (non-ART) medication regimen (Non-ART Medication Adherence) Depressive symptoms (Depression) Perceived symptoms of anxiety/stress (Perceived Stress) Satisfaction with perceived social support (Social Support) Number and type of HIV-related symptoms (HIV Symptoms) Perceived ability to work and perform daily role (Role Function) Perceived health functioning (Functional Health) Perceived vitality, memory and reasoning, and affect (Social–Emotional–Cognitive Function) Self-rating of perceived general health (Health Perceptions)

Measureb ACTG Baseline Adherence Questionnaire ACTG Baseline Adherence Questionnairec ACTG Baseline Adherence Questionnaire ACTG Baseline Adherence Questionnaire ACTG Baseline Adherence Questionnaire ACTG Symptom Distress Module Multidimensional Health Status Multidimensional Health Status Multidimensional Health Status Multidimensional Health Status

N Items Chronbach’s α Mean score (±SD) 954

3

0.59

73.1 ± 22.4

325

14

0.89

89.1 ± 14.6

950

7

0.83

75.2 ± 19.6

950

10

0.83

63.7 ± 16.9

613

2

0.41

67.1 ± 29.8

938

20

0.89

84.0 ± 15.6

966

5

0.82

79.1 ± 21.5

942

4

0.83

79.3 ± 25.6

955

9

0.84

72.7 ± 17.4

950

4

0.82

64.3 ± 22.3

with missing data were excluded from the analysis (total N = 980). raw scores were rescaled to a 100-point scale (0 = worst health/lowest rating to 100 = best health/highest rating). c Participants who were naive to antiretroviral agents, but taking other medications (N = 325).

a Patients b The

adhere (ART Adherence Self-Efficacy) at baseline (Table I). Among the 325 participants currently taking medication, the Non-ART Medication Adherence mean score was 89, indicating that relatively few barriers interfered with the subjects’ ability to take medications during the last month. The 7 depressive symptoms as measured by the Depression scale were taken from the original 20 items for the CES-D (Radloff, 1977). On the original CES-D scale a score of less than 63 (recalibrated to range 0–100 for comparison purposes) corresponded to moderate to severe depression, while a score between 63 and 75 corresponded to mild depression. While the thresholds for the abbreviated version used in this study are not exactly comparable, the high internal consistency among these items allows for a fair comparison. In our sample, 25% of subjects scored less than 62 on the depression scale, while slightly less than 50% scored lower than 75, indicating a fairly high prevalence of depressive symptoms. The other psychosocial well-being scores (Perceived Stress, Social Support) reflected a similar pattern of impairment. The perceived general health (Health Perceptions) scale was considerably lower than for healthy

individuals of comparable age and sex. On the 0– 100 linear analogue health rating item (worst possible health/death = 0 to best possible health = 100), the mean was 74.9 (SD = 19.2), and 25% of subjects rated their health lower than 65. These scores were similar to those obtained by Wu et al. (2002) using the ED-5 visual analogue scale (scored 0–100) in ACTG 204. For comparison purposes, the visual analogue health rating scale from the ED-5 yielded a mean of 87 in a national survey in the United Kingdom for individuals aged 30–39 years in the middle (grade III) social class (Kind et al., 1998), with only individuals aged 80 years and older reporting a mean score lower than 75. Other health function subscales (HIV Symptoms, Role Function, Functional Health, Social– Emotional–Cognitive Functioning) also reflected relatively lower health functioning as compared to the general population of comparable age. Beliefs About Antiretroviral Therapy In contrast, overall beliefs about antiretroviral therapy effectiveness and adherence (ART Adherence Self-Efficacy) were relatively high. However,


PP1211-aibe-486486

June 2, 2004

9:51


Factors Influencing Medication Adherence Beliefs

145

Table II. Beliefs About the Effectiveness of Antiretroviral Medication and Confidence in One’s Ability to Adhere (ART Adherence Self-Efficacy) Percentage responding in each category Question

contenta

(number responding)

Not sure

Somewhat sure

Very sure

Extremely sure

2.3

8.2

33.7

55.7

7.8

19.1

36.6

36.5

11.2

12.5

28.2

48.1

Certainty about ability to take all or most of the antiretroviral medications as directed (N = 953) Certainty that antiretroviral medication will have a positive effect on health (N = 950) Certainty that if antiretroviral medication is not taken exactly as instructed, the body will become resistant to HIV medications (N = 952) a These

three items formed the ART Adherence Self-Efficacy scale.

ication benefits ( p = .001). Gender was not a significant correlate ( p = .233), and the degree of association between education and ART Adherence Self-Efficacy was fairly comparable between genders (gender by education interaction, p = .414) as illustrated in Fig. 1. A median split [(low < 78, mean = 52.7 ± 16.5) versus (high ≥ 78, mean = 89.3 ± 9.4)] on the ART Adherence Self-Efficacy variable showed that individuals with higher scores also reported better quality of life and health functioning (Fig. 2). When Gender, Age, Race, Level of Education, Depression, Perceived Stress, HIV Symptoms, Role Functioning, Functional Health, Social–Emotional– Cognitive Functioning, and Health Perceptions were entered into a stepwise regression with ART Adherence Self-Efficacy as the dependent variable (n = 880 with data on all variables), the final model selected the variables Age, Level of Education, Perceived Stress, Health Perceptions, and Role Functioning (R2 = 0.10) (Table IV). After controlling for Age ( p < .0001) and Education ( p = .007), lower Perceived Stress ( p < .0001) and higher Health Perceptions ( p < .0001) remained significantly associated

while 56% were “extremely sure” of their ability to take all medications and 48% “extremely sure” that antiretroviral therapy nonadherence would cause resistance, only 37% were as sure that antiretroviral therapy would benefit their health (Table II). Associations Among Personal/Situational Characteristics, Quality of Life, and ART Adherence Self-Efficacy Among the bivariate correlations examined, Perceived Stress showed the highest association with participants’ beliefs about antiretroviral therapy effectiveness and his or her ability to adhere to antiretroviral medications as directed (ART Adherence Self-Efficacy, r = 0.236), with higher levels of stress corresponding to lower levels of selfefficacy (Table III). In a multivariable linear model which included age, sex, education, and their interactions, ART Adherence Self-Efficacy was found to be associated with level of education; the higher the level of education, the greater was one’s belief in his or her ability to adhere and confidence in med-

Table III. Correlations: Personal and Situational Factors, Quality of Life, and Symptom Distress with ART Adherence Self-Efficacy and (Nonantiretroviral) Medication Adherencea Antiretroviral Adherence Non-ART Medication Self-Efficacy (ASE) Adherence (ADHERE) Scale

r

n

p

r

N

p

Depression Health Functioning Role Functioning Health Perceptions Social–Emotional–Cognitive Functioning Perceived Stress Social Support HIV Symptoms

0.19 0.11 0.09 0.21 0.20 0.24 0.19 0.16

945 928 952 938 942 945 611 928

0.0001 0.001 0.005 0.0001 0.0001 0.0001 0.0001 0.0001

0.33 0.15 0.16 0.19 0.25 0.32 0.03 0.25

325 318 324 320 321 325 258 320

0.0001 0.008 0.005 0.001 0.0001 0.0001 0.601 0.0001

a Value

of the pearson correlation coefficient r for ASE with ADHERE = 0.25 ( p < .0001).


PP1211-aibe-486486

June 2, 2004

9:51


146

Reynolds et al.

Fig. 1. Comparison of mean Antiretroviral Adherence Self-Efficacy (men vs. women) stratified by levels of education (N = 954). There was no significant difference between men and women across levels of education; however, self-efficacy increased significantly as education increased ( p = .007) (see stepwise regression results).

with higher ART Adherence Self-Efficacy. The added impact of Role Functioning, once all of these variables were entered into the equation, was negative ( p = .016), however, most likely reflecting reduced self-efficacy caused possibly by the inability to meet the increased demands of higher role performance.

Barriers to Non-ART Medication Adherence Of the subset of participants (N = 325) reporting current nonantiretroviral medication use, the most common reasons reported for nonadherence were “simply forgot” (33%), “away from home”

Fig. 2. Bars represent the comparisons of quality-of-life means for individuals grouped by Antiretroviral Adherence Self-Efficacy (ASE) (low ASE <78 vs. high ASE ≥ 78). Nominal significance levels are reported for univariate between-group mean comparisons using the independent t test. ∗∗∗p < .001, ∗∗ p < .01, ∗ p < .05, and # P = .075.


PP1211-aibe-486486

June 2, 2004

9:51


Factors Influencing Medication Adherence Beliefs Table IV. Stepwise Regressiona Variable in mode1 Constant Age Education Stress Health perceptions Role

Regression coefficient SE coefficient 64.39 −0.33 1.70 0.27 0.14 −0.11

4.19 0.08 0.63 0.50 0.05 0.04

p .000 .000 .007 .000 .002 .016

for model = 0.309, R2 = 0.095, p value based on t statistic for regression coefficient. The order of entry for the variables was stress, age, education, health perceptions, and role.

aR

(27%), and “busy” (26%) (Fig. 3). Depression and Perceived Stress were most strongly correlated with nonantiretroviral medication adherence (Non-ART Medication Adherence) (r = .33 and 0.31, respectively; p < .001) (Table III). In a separate validation analysis of ACTG 384 on-treatment data, the correlation coefficient between actual on-study antiretroviral therapy adherence and barriers to baseline nonantiretroviral therapy adherence (Non-ART Medication Adherence) and antiretroviral therapy adherence self-efficacy (ART Adherence Self-Efficacy) was re-

147 ported to be 0.19 and 0.16, respectively (Spearman’s rho), p < .001 (Reynolds et al., 2003). DISCUSSION Baseline correlates of beliefs about antiretroviral therapy effectiveness, ability to adhere, and selfreported adherence to nonantiretroviral medication were examined in a diverse population of 980 persons naive to therapy. A greater belief in one’s ability to adhere to antiretroviral therapy and confidence in the benefits of antiretroviral therapy were associated with higher quality of life and health functioning. More highly educated and younger individuals were more confident about the benefits of antiretroviral therapy and their ability to adhere. When age, gender, race, and education were controlled, greater role functioning, better perceptions of one’s health, and lower stress remained the strongest predictors of one’s beliefs about antiretroviral therapy effectiveness and adherence. Social support also proved to be an important explanatory variable in the subset of participants

Fig. 3. Percentage of participants reporting reasons for barriers to adherence to (nonantiretroviral) medications (N = 325).


PP1211-aibe-486486

June 2, 2004

9:51


148 who rated their support from family and friends. In the subset of individuals experienced with medications other than antiretrovirals, depressive symptoms were the strongest correlate of low adherence. Despite wide recognition that adherence to antiretroviral regimens is critical to treatment success, rates of adherence are frequently subtherapeutic. As many as half of the individuals on combination therapies fail to take antiretroviral medication in accordance with dosage, time, and dietary instructions (Murphy et al., 2001b; Nieuwkerk et al., 2001). Because poor adherence places individuals at high risk for treatment failure, the potential for an individual to adhere to an antiretroviral regimen has become an important consideration in decision making surrounding when to start therapy. There has been some question about whether persons at risk for poor adherence can be identified prior to treatment initiation. Demographic characteristics (e.g., age, gender, race/ethnicity) and provider assessment have not been found to be reliable predictors of adherence success or failure. Adherence is also increasingly understood as a dynamic behavior that will change over time in response to a variety of personal and situational factors (Lucas et al., 2002; Spire et al., 2002). While factors influencing adherence behavior may change over time, findings from this study indicate that characteristics associated with lower belief in one’s ability to adhere can be identified in persons naive to therapy. In this analysis we have explored the modifiable factors that affect adherence self-efficacy, namely depression, stress, and social support at baseline. While the associations between variables in this study were modest, the findings are potentially important; early assessment of personal and situational factors may allow identification of persons at greater initial risk for lower adherence and permit focused interventions to improve the likelihood of treatment success. Instituting interventions to improve adherence rates (e.g., memory enhancers, reminders, and more convenient dosing schedules) after the patient has started an antiretroviral therapy regimen, without first addressing the root causes of impaired psychological functioning, might prevent achievement of their full potential. Measures used in this study are quite simple to administer and could be adapted for clinical risk assessment purposes. Since adherence self-efficacy and non-ART medication adherence barriers are associated with actual antiretroviral adherence (Reynolds et al., 2003), it would be worthwhile to test the hypoth-

Reynolds et al. esis that by increasing adherence self-efficacy one increases actual antiretroviral adherence. Research is needed to establish what approach would most effectively address risk factors to enhance adherence prior to treatment initiation. Psychological counseling, stress reduction, and coping with potential barriers to adherence may be viable approaches. Such initial measures to enhance adherence will likely need to be supplemented by strategies designed to support continued adherence behavior. Additional analyses are needed to further establish the relationship between factors identified at baseline in this study and adherence to ACTG 384 ART regimens and health outcomes over time. To maintain the integrity of the primary ACTG 384 analyses, objective, longitudinal indicators of adherence and health outcomes (immunologic/virologic) were not fully available for this analysis. It is, however, anticipated that the factors identified here will also predict subsequent adherence behavior and health outcomes. Depression (Catz et al., 2000; Gordillo et al., 1999; Safren et al., 2001; Schuman et al., 2001), beliefs (e.g., Catz et al., 2000; Gao et al., 2000; Safren et al., 2001; Schroeder et al., 2001), and problems in remembering to take medications (Chesney et al., 2000a; Ostrop et al., 2000; Schuman et al., 2001) are among factors that have repeatedly been associated with adherence over the course of antiretroviral therapy. Less information is available concerning the relationship of role functioning and better perceptions of one’s general health with one’s beliefs about antiretroviral therapy effectiveness. This warrants further consideration. In conclusion, in this large, multicenter antiretroviral strategy trial, personal and situational factors, such as depressive symptoms, perceived stress, and lower education, were associated with less certainty about the potential for antiretroviral therapy effectiveness and one’s perceived ability to adhere to therapy. Findings from these analyses suggest the potential importance of baseline screening for adherence predictors and screening questionnaires and a role for focused interventions to address modifiable factors placing persons at risk for poor adherence prior to antiretroviral treatment initiation.

ACKNOWLEDGMENTS Support was provided by the U.S. NIH, NIAID, Adult AIDS Clinical Trials Group (AI38858), the Ohio State University (AI25924), Harvard University


PP1211-aibe-486486

June 2, 2004

9:51

Factors Influencing Medication Adherence Beliefs (AI27659), and the University of North Carolina (AI25868) AIDS Clinical Trial Units. We thank the participating ACTU personnel and study volunteers for their contributions to this project. We also acknowledge the contributions of the ACTG 384 team and the ACTG 384 adherence substudy (ACTG 731 and 5031s) team investigators, in particular, Jeannette Ickovics, Ph.D., Yale University; Ian Frank, M.D., University of Pennsylvania; and Cheryl Marcus, R.N., University of North Carolina at Chapel Hill. REFERENCES Bangsberg, D. R., Hecht, F. M., Charlebois, E. D., Zolopa, A. R., Holodniy, M., Sheiner, L., Bamberger, J. D., Chesney, M. A., and Moss, A. (2000). Adherence to protease inhibitors, HIV-1 viral load, and development of drug resistance in an indigent population. AIDS, 14, 357–366. Bangsberg, D. R., Hecht, F. M., Clague, H., Charlebois, E. D., Ciccarone, D., Chesney, M., and Moss, A. (2002). Provider assessment of adherence to HIV antiretroviral therapy. Journal of Acquired Immune Deficiency Syndromes, 26, 435–442. Becker, M. H. (1974). The health belief model and personal health behavior. Health Education Monographs, 2, 324–508. Becker, M. H. (1988). AIDS and behavior change. Public Health Reviews, 16, 1–11. Catz, S. L., Kelly, J. A., Bogart, L. M., Benotsch, E. G., and McAliffe, T. L. (2000). Patterns, correlates, and barriers to medication adherence among persons prescribed new treatments for HIV disease. Health Psychology, 19, 124–133. Chesney, M. A., Morin, M., and Sherr, L. (2000a). Adherence to HIV combination therapy. Social Science and Medicine, 50, 1599–1605. Chesney, M. A., Ickovics, J. R., Chambers, D. B., Gifford, A. L., Neidig, J., Zwickl, B., and Wu, A. W. (2000b). Self-reported adherence to antiretroviral medications among participants in HIV clinical trials: The AACTG Adherence Instrument. AIDS Care, 12, 255–266. Gao, X., Nau, D. P., Rosenbluth, S. A., Scott, V., and Woodward, C. (2000). The relationship of disease severity, health beliefs and medication adherence among HIV patients. AIDS Care, 12, 387–398. Gordillo, V., del Amo, J., Soriano, V., and Gonzalez-Lahoz, J. (1999). Socioeconomic and psychological variables influencing adherence to antiretroviral therapy. AIDS, 13, 1763–1769. Haddad, M., Inch, C., Glazier, R. H., Wilkins, A. L., Urbshott, G., Bayoumi, A., and Rourke, S. (2002). Patient support and education for promoting adherence to highly active anitiretroviral therapy for HIV/AIDS. The Cochrane Library, 4, Oxford: Update Software. Ickovics, J. R., and Meade, C. S. (2002). Adherence to HAART among patients with HIV: Breakthroughs and barriers, AIDS Care, 14, 309–318. Janz, N. K., and Becker, M. H. (1984). The health belief model: A decade later. Health Education Quarterly, 11, 1–47. Justice, A. C., Holmes, W., Gifford, A. L., Rabeneck, L., Zackin, R., Sinclair, G., Weissman, S., Neidig, J., Marcus, C., Chesney, M., Cohn, S. E., and Wu, A. (2001). Development and validation of a self-completed HIV symptom index. Journal of Clinical Epidemiology, 54, S77–S90. Karon, J. M., Fleming, P. L., Steketee, R. W., and De Cock, K. M. (2001). HIV in the United States at the turn of the century. American Journal of Public Health, 91, 1060–1068.


149 Kind, P., Dolan, P., Gudewx, C., and Williams, A. (1998). Variations in population health status: Results from a United Kingdom national questionnaire survey. British Medical Journal, 316, 737–741. Lucas, G. M., Gebo, K. A., Chaisson, R. E., and Moore, R. D. (2002). Longitudinal assessment of the effects of drug and alcohol abuse on HIV-1 treatment outcomes in an urban clinic. AIDS, 16, 767–774. Mannheimer, S., Friedland, G., Matts, J., Child, C., and Chesney, M. (2002). The consistency of adherence to antiretroviral therapy predicts biologic outcomes for human immunodeficiency virus-infected persons in clinical trials. Clinical Infectious Diseases, 34, 1115–1121. Montgomery, S. B., Joseph, J. G., Becker, M. H., Ostrow, D. G., Kessler, R. C., and Kirscht, J. P. (1989). The health belief model in understanding compliance with preventive recommendation for AIDS: How useful? AIDS Education and Prevention, 1, 303–323. Murphy, D. A., Wilson, C. M., Durako, S. J., Muenz, L. R., and Belzer, M. (2001a). Antiretroviral medication adherence among the REACH HIV-infected adolescent cohort in the USA. AIDS Care, 13, 27–40. Murphy, E. L., Collier, A. C., Kalish, L. A., Assmann, S. F., Para, M. F., Flanigan, T. P., Kumar, P. N., Mintz, L., Wallach, F. R., and Nemo, G. J. (2001b). Highly active antiretroviral therapy decreases mortality and morbidity in patients with advanced HIV disease. Annals of Internal Medicine, 135, 17–26. Nieuwkerk, P. Y., Sprangers, M. A., Burger, D. M., Hoetelmans, R. M., Hugen, P. W., Danner, S. A., vab Der Ende, M. E., Schneider, M. M., Schrey, G., Meenhorst, P. L., Sprenger, H. G., Kauffmann, R. H., Jambroes, M., Chesney, M. A., de Wolf, F., and Lange, J. M. (2001). Limited patient adherence to highly active antiretroviral therapy for HIV-1 infection in an observational cohort. Archives of Internal Medicine, 161, 1962– 1968. Ostrop, N. J., Hallet, K. A., and Gill, M. J. (2000). Long-term patient adherence to antiretroviral therapy. Annals of Pharmacotherapy, 34, 703–709. Palella, F. J., Delaney, K. M., Moorman, A. C., Loveless, M. O., Fuher, J., Satten, G. A., Aschman, D. J., and Holmberg, S. D. (1998). Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. New England Journal of Medicine, 338, 853–860. Paterson, D. L., Swindells, S., Mohr, J., Brester, M., Vergis, E. N., Squier, C., Wagener, M. M., and Singh, N. (2000). Adherence to protease inhibitor therapy and outcomes in patients with HIV infection, Annals of Internal Medicine, 133, 21–30. Radloff, L. S. (1977) The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1, 385–401. Reynolds, N. R., Testa, M. A., Su, M., Chesney, M. A., and Robbins, G. K. for the Protocol Teams of ACTG 384. (July, 2003). Baseline predictors of adherence and regimen failure in treatment na¨ıve HIV+ subjects (ACTG 384). Presented at the 2nd IAS Conference on HIV Pathogenesis and Treatment, Paris. Safren, S. A., Worth, J. L., Salomon, E., Johnson, W., Mayer, K., and Boswell, S. (2001). Two strategies to increase adherence to HIV antiretroviral medication: Life steps and medication monitoring. Behavioral Research and Therapy, 39, 11151– 1162. Schroeder, J. R., Latkin, C. A., Hoover, D. R., Knowlton, A. R., Zenilman, J., Strathdee, S., and Celantano, D. D. (2001). Social factors related to antiretroviral therapy use in injection drug users. AIDS and Behavior, 5, 363–369. Schuman, P., Ohmit, S. E., Cohen, M., Sacks, H. S., Richardson, J., Young, M., Schoenbaum, E., Rompalo, A., and Gardner, L. (2001). Prescription of and adherence to antiretroviral therapy among women with AIDS. AIDS and Behavior, 5, 371– 378.


PP1211-aibe-486486

June 2, 2004

9:51


150 Smeaton, L. M., DeGruttola, V., Robbins, G. K., and Shafer, R. W. (2001). ACTG (AIDS Clinical Trials Group 384): A strategy trial comparing consecutive treatments for HIV-1. Controlled Clinical Trials, 22, 142–159. Spire, B., Duran, S., Souville, M., Leport, C., Raffi, F., and Moatti, J. P. (2002). Adherence to highly active antiretroviral therapy: From a predictive to a dynamic approach. Social Science and Medicine, 54, 1481–1496. SPSS Inc. (2002). SPSS version 11.5. Chicago: Author. Volberding, P. (2002). Adherence, resistance and timing: Current issues in the use of new therapies. AIDS Reader, 12, 349–357, 368. Wu, A. W., Rubin, H. R., Matthews, W. C., Ware, J. E., Brysk, L. T., Hardy, W. D., Bozzette, S. A., Spector, S. A., and Richman, D. D. (1991). A health status questionnaire using items from the Medical Outcomes Study. Preliminary valida-

Reynolds et al. tion in persons with early HIV infection. Medical Care, 29, 786–798. Wu, A. W., Jacobson, D. L., Frick, K. D., Clark, R., Revicki, D. A., Freedberg, K. A., Scott-Lennox, J., and Feinberg, J. (2002). Validity and responsiveness of the EuroQol as a measure of health-related quality of life in people enrolled in an AIDS clinical trial. Quality of Life Research, 11, 273– 282. Yeni, P. G., Hammer, S. M., Carpenter, C., Cooper, D. A., Fischl, M. A., Gatell, J. M., Gazzard, B. G., Hirsch, M. S., Jacobsen, D. M., Katzenstein, D. A., Montaner, J. S., Richman, D. D., Saag, M. S., Schechter, M., Schooley, R. T., Thompson, M. A., Vella, S., and Volberding, P. A. (2002). Antiretroviral treatment for adult HIV infection in 2002: Updated recommendations of the International AIDS Society—USA Panel. JAMA, 288, 222–235.

FACTORS INFLUENCING MEDICATION ADHERENCE BELIEFS AND SELF-EFFICACY

Recommend Documents