Secondary Logo

Journal Logo


Telephone Nurse Counseling Improves HIV Medication Adherence

An Effectiveness Study

Cook, Paul F., PhD; McCabe, Mishcha M., DNP; Emiliozzi, Suzie, RN; Pointer, Lauren, MS

Author Information
Journal of the Association of Nurses in AIDS Care: July-August 2009 - Volume 20 - Issue 4 - p 316-325
  • Free


Despite the benefits that newer antiretroviral treatment (ART) options can provide for persons living with HIV (PLWH), most people have difficulty adhering to ART medications. Baseline ART adherence has been estimated in the range of 50% to 70% during the first 6 months of treatment (Simoni, Pearson, Pantalone, Marks, & Crepaz, 2006), using methods that include electronic medication cap monitoring devices (Rigsby et al., 2000) and participant reports collected in real time (Ickovics et al., 2002). In addition, a significant percentage of PLWH never even begin prescribed treatment: In one study of ART, 22% of participants either declined treatment or agreed to begin but then did not (Maisels, Steinberg, & Tobias, 2001). Although fixed-dose combination medications reduce regimen complexity, nonadherence remains a problem (Parienti et al., 2007).

Nonadherence has a strong direct effect on disease progression and survival in PLWH. In one study, PLWH with CD4 counts lower than 200 cells/cc3 who discontinued ART had a risk of disease progression five times greater than that of PLWH with similar CD4 counts who took their medication, independent of the medication's effect on viral suppression (Lucas, 2005). The consequences of nonadherence can also be measured in terms of health care resources, with nonadherent PLWH in one study having an average of 12.9 hospital days per 1,000, whereas adherent participants' rate was only 2.6 days per 1,000 (Valenti, 2001). Adherence also reduced the number of deaths and eliminated opportunistic infections during the course of that study.

For many medical conditions, partial adherence is acceptable, but this is not the case with ART. Partial adherence to ART regimens may create drug resistant strains of HIV, with intermediate adherence levels being the most problematic: In one study, maximum ART drug resistance occurred at an 87% adherence level (Bangsberg et al., 2004), which is higher than the 80% adherence rate considered acceptable for many chronic medications. Unfortunately, because ART medications fall into only a few broad classes, HIV strains that become resistant to one medication may have resistance to all medications in that class, further limiting treatment options. Nonadherence to ART thus creates a vicious cycle in which increasingly complex medication regimens are required, regimen complexity further disrupts adherence, and increasing drug resistance results (Lucas, 2005).

Interventions to improve ART adherence include a range of techniques borrowed from the psychotherapy literature. Negative beliefs such as skepticism about the efficacy of ART, concerns about medication effects or adverse drug events (ADEs), and inaccurate ideas about using ART intermittently all predict nonadherence (Abel & Painter, 2003; Barfod et al., 2005; Deloria-Knoll et al., 2004). Cognitive-behavioral and educational models have shown some efficacy to address these problems (Holzemer et al., 2006; Koenig et al., 2008; Reynolds et al., 2008). Another technique with proven efficacy to promote ART adherence is motivational interviewing, which involves helping PLWH to clarify valued goals or objectives and then assisting them to see discrepancies between actual and desired behavior (DiIorio et al., 2008; Golin et al., 2006). Because PLWH can have negative emotional reactions such as denial, anger, and depression that result in nonadherence (Enriquez, Lackey, O'Connor, & McKinsey, 2004), they may benefit from a motivational interviewing stance that explicitly acknowledges patients' ambivalence about treatment. Helping PLWH develop stronger working relationships with their health care providers (Douglass, Sowell, & Phillips, 2003) and case management assistance to overcome logistical barriers to treatment (De Bruin, Hospers, de Borne, Kok, & Prins, 2005) are other efficacious interventions to improve ART adherence.

The current study's objective was to evaluate the effectiveness of ScriptAssist, a psychologically based telephone support program offered by CenCorp Health Solutions (St. Louis, MO) as a method to improve ART adherence among PLWH seen in standard clinical care settings. The study's primary hypothesis was that telephonic delivery of psychological counseling would increase the percentage of participants attaining acceptable levels (≥ 95%) of ART adherence over time. In this project, registered nurses (RNs) working at a clinical nursing call center organization provided telephone counseling to PLWH taking a fixed dose combination medication for HIV (lopinavir/ritonavir) in 2006-2007, and followed up with participants over a period of 6 months.



Potential participants were identified by physician referral (129 PLWH) or self-referral to the ScriptAssist program (15 PLWH, or 10.4% of the total) using a brochure distributed to HIV care settings by pharmaceutical representatives. Participants receiving ART from any health care provider nationwide were eligible; they were not limited to a specific practice setting, geographic area, or demographic group. Of 144 referrals, 100 could be reached by telephone and were offered the ScriptAssist intervention (see Figure 1). This level of contact difficulty mirrors results for other high-risk populations (e.g., Cook, Emiliozzi, Waters, & El-Hajj, 2008).

Figure 1. Study recruitment and participant flow.
Figure 1. Study recruitment and participant flow.

Physicians were asked to refer PLWH at the start of a new course of treatment, and 92% of participants were new to treatment with ritonavir/lopinavir at the time they enrolled in the ScriptAssist program. All referred PLWH were sent written information and received a telephone outreach attempt from a ScriptAssist nurse to explain the project; those who were not interested could opt out based on verbal or written notification. Only 2 participants withdrew, and both withdrawals were at the time of the initial outreach contact.

The 98 participants had an average age of 43.5 years (SD = 9.3), age range 23 to 73. Participants were 38.8% women (38/98), which is a slightly higher representation than the approximately 29% of all PLWH in the United States who are women (Centers for Disease Control and Prevention, 2006). Race/ethnicity data were not collected. Those patients who were referred but not enrolled had a mean age of 42.4 years (SD = 9.3), and 30.4% (14/46) were female.

Clinically, participants had relatively complex treatment regimens, taking an average of 2.4 other medications (SD = 2.9) in addition to lopinavir/ritonavir. The number of additional medications taken ranged from 0 to 17, with 73% of participants taking at least one additional medication. A total of 9 participants reported taking one or more over-the-counter medications in addition to prescription drugs. Participants reported an average of 1.4 comorbid diagnoses (SD = .7, range: 1-4). A total of 32 of 98 participants (32.7%) reported any ADEs over the course of treatment, with a range from one to six ADEs (Mdn = 1.5 ADE) among participants who had these experiences. During their initial contact with ScriptAssist, 77 participants (78.6%) were classified as at risk for nonadherence using a previously developed predictive algorithm (Cook, 2006), with 21 (21.4%) classified as low risk. The percentage of at-risk participants was similar to reported 12-month nonadherence rates for ART.


Telephone support and counseling were delivered individually by call center-based nurses trained on a psychological model to promote health behavior change. The ScriptAssist nurse assessed the participant's concerns and barriers to adherence at the time of each follow-up call. The nurse then provided psychological counseling interventions based on the participant's assessed readiness for change, using Prochaska's transtheoretical model as a guiding framework (Cook, 2006). Those participants initially screened as being at low risk for nonadherence were scheduled to receive one follow-up call at 6 months, whereas higher risk participants were scheduled to receive four follow-up calls. Participants received further telephone calls from the same nurse for up to 6 months, with a median of three successful contacts per participant (range: 1-14), and an average call length of 7.5 minutes (SD = 11.7 minutes). Nurses made multiple attempts (Mdn = 3.6 attempts per completed call) to enroll and retain participants; contact information could be updated by participants or by their health care providers if they were aware of any changes. Participants were also encouraged to choose call times that were convenient for them. There were no incentives for participation.

Ten nurses were trained to provide the telephone intervention; one had prior HIV clinic experience and acted as a clinical lead. The others had a mix of acute and ambulatory care experience in a variety of clinical settings. All had some level of experience caring for HIV-infected patients and had administered ART during their careers. Nurses were trained to use motivational interviewing strategies to build treatment motivation with participants who were less ready for change or cognitive-behavioral problem-solving strategies with participants who were already motivated to improve their adherence. The ScriptAssist program's psychological counseling model emphasized assessment of individual beliefs, goals, and reactions to treatment. Although nurses did provide some education in the context of their counseling discussions, communicating a standardized set of educational topics was not a focus of the intervention; instead, participants received information in response to their individual concerns and requests. After each telephone call, nurses had the option to mail participants follow-up printed materials to reinforce themes discussed during telephone conversations. The participant's HIV care provider received a written progress note from the ScriptAssist nurse after each call, which had the dual purpose of reporting any adherence issues identified and promoting participants' working relationships with their health care providers.

All ScriptAssist program nurses received a standardized training about lopinavir/ritonavir's mechanism of action, features and benefits, and potential ADEs before the start of the project. Education materials were developed for the ART adherence program on topics including effective communication with health care providers, problem solving to address treatment barriers, and managing disclosure risk in HIV. Nurses received 8 hours of training on the psychological counseling model from the first author and subsequent training with role-play exercises from the third author. Intervention fidelity was supported through team meetings with other nurses using the same counseling model, one-on-one supervision, real-time call monitoring, and a half day retraining activity at the midpoint of this 2-year project. The most frequent feedback provided to program nurses through this process was to focus less on patient education and more on listening to patients' current concerns and motivations for treatment. Individual nurses' performance was addressed through ongoing supervision. No major deviations from the program's counseling model were noted by the clinical supervisor or by the first author at the midpoint retraining session.


Interview-based data collection from all participants included demographics, risk factors for nonadherence, and level of adherence assessed at the time of each telephone call. Participants were informed that deidentified aggregate program data would be used for research purposes, and their verbal consent to participate was recorded in the program's clinical database. They were also informed that they could withdraw at any time by notifying the study nurse. Data analysis was approved by the Colorado Multiple Institutional Review Board.

Demographic and clinical variables

Demographic information (age and sex) was collected at the time of program enrollment. Clinical data were recorded at the time of each telephone call, including any comorbid medical conditions reported by the participant, any additional medications reported, and any adverse events noted. The nurse did not systematically inquire about each of these areas, but instead recorded any information that the participant volunteered. Therefore, prevalence rates based on these data are likely to be underestimates.

Primary outcome measure: Medication adherence

Participants' ART adherence was recorded by the ScriptAssist nurse at the time of each telephone call based on the participant's response to the question “What percentage of the time do you currently take your medication as prescribed?” When adherence is assessed in a nonjudgmental interview format by someone other than the medication prescriber, self-report provides a quick and relatively accurate measure of ART adherence (Lu et al., 2008). Self-reported nonadherence predicts treatment failure in ART (Barfod et al., 2005; Ickovics et al., 2002), and this specific interview-based adherence measure has shown 75% agreement with pharmacy data in earlier research (Cook, Emiliozzi, & McCabe, 2007). Interviews were completed by the RN during each call, so interviewers were not blind to the intervention. Two separate measures of adherence were calculated from interview data: treatment persistence, a yes/no variable defined by the participant's report that he or she was continuing to take medication or not, and percent adherence, an ordinal level variable based on the participant's report of the percentage of time he or she was adherent to treatment as prescribed. The following categories were available to record responses: <25%, 25%-49%, 50%-69%, 70%-79%, 80%-89%, 90%-94%, 95%-99%, and 100% adherence. Because adherence of at least 95% is required to gain therapeutic effects and prevent resistance (Bangsberg et al., 2004), the percent adherence measure was dichotomized into adequate adherence (95%-100%) versus inadequate adherence (below 95%) for analysis.

Self-efficacy for treatment adherence

At the time of each telephone contact, participants were also asked, “How confident are you currently in your ability to follow all of your physician's treatment recommendations?” with the answer rated on a five-point Likert-type scale ranging from not at all confident to very confident. Confidence ratings are a standard method for self-efficacy measurement; in the current study, construct validity was supported by significant between-participant differences in self-efficacy, χ2 = 104.9, p = .009, which is understood to be a stable characteristic of persons. However, an intraclass correlation coefficient of just .26 also indicated substantial within-person variability in self-efficacy ratings over time.

Reasons for nonadherence

During each call, the RN asked the participant about any reasons for nonadherence and/or treatment concerns. All reported problems or concerns were analyzed, whether or not they resulted in nonadherence. RNs recorded participants' concerns using six categories: (a) cost of treatment, (b) treatment logistics (e.g., lack of transportation, inconvenience, continuity of care), (c) forgetting to take medication, (d) concern about possible adverse events, (e) ongoing or breakthrough symptoms, or (f) negative beliefs about treatment (e.g., does not think medication is needed, does not think medication will help, does not want to take medication, or is concerned about other people's opinions). RNs have shown interrater reliability of κ = .93 using this tool to rate call transcripts (Cook, 2008).

Data Analysis

Descriptive statistics and between-group comparisons were calculated using SPSS for Windows, version 16, and SAS version 9.1. Attrition, which is common in studies with follow-up over multiple months, can lead to biased results if the missing data are not missing at random. To address this possibility, sensitivity analyses were conducted to examine the relationship between attrition, adherence, and demographic variables. Primary outcome analyses were then conducted using a binomial test comparing the 6-month program adherence rate with the expected population rate based on published literature. Participants were identified as nonadherent if they stated that they had stopped treatment and did not intend to resume; participants were identified as lost to follow-up if they failed to provide adherence data in a subsequent study month (e.g., if they could no longer be reached by phone). Nonadherent participants were carried forward into the denominator for all future time periods, whereas participants lost to follow-up were dropped from both the numerator and denominator. This analysis strategy is less conservative than an intent-to-treat analysis in adherence studies with missing data, but it is also less likely to produce biased estimates and has been used successfully in earlier research (Cook et al., 2007; Cook et al., 2008). Power was .80 to detect intervention effects as small as ϕ = .20 with N = 98 and α = .05.


Sample Representativeness and Attrition

Participants were similar to the national HIV epidemic in terms of sex; race/ethnicity data were not collected. Participants were also similar to PLWH who were referred to ScriptAssist but not enrolled because they could not be reached by phone, in terms of both sex, χ2 (1, N = 144) = .94, p = .33, and age, t (141) = .61, p = .54.

In addition to 10 participants who permanently stopped ART during the course of the study (including the 3 who never started treatment), 57 participants (58%) were lost to follow-up by 6 months. This was a higher level of attrition than in ScriptAssist programs for other chronic diseases (Cook, 2006; Cook et al., 2007; Cook et al., 2008). Participants lost to follow-up were similar to those who continued in terms of sex, χ2 (1, N = 98) = .002, p = .97, and age, t (95) = 1.01, p = .32. They were also similar in terms of the number of prescription and over-the-counter medications reported, number of health care providers currently seen, and number of ADEs reported (all ps > .13). Participants who completed the 6-month intervention reported a slightly higher number of comorbid conditions (M = 1.58, versus 1.24 in the group lost to follow-up), but this difference was not significant, t (43.4) = 1.95, p = .058. Most important, participants lost to follow-up were no different from those who completed the intervention in terms of baseline medication adherence, χ2 (1, N = 98) = 2.41, p = .09, or confidence in their ability to take ART medication, t (75) = 1.30, p = .20. These results suggest that attrition was not related to the primary outcome of adherence. Participants' length of stay in the study was also not correlated with their baseline adherence, Spearman r = .09, p = .37.

Treatment Persistence

Cumulative ART persistence and percent adherence by month of treatment are shown in Table 1. Only 9 participants (9.2%) had not filled their initial prescription by the time of their first contact with ScriptAssist, which is a lower initial nonadherence rate than the approximately 20% seen in other chronic diseases (Cook, 2006; Cook et al., 2007) but is consistent with other studies of ART (Maisels et al., 2001). Of those who did not fill their initial prescriptions, 6 (66.7%) expressed an intention to start, and all had in fact started treatment by the next follow-up call. The remaining 3 participants chose not to start treatment. Of the 95 participants who eventually started treatment, 12 (12.6%) stated that they stopped treatment at some point during the study. A total of 5 of these participants (41.7%) expressed an intention to resume. A total of 7 participants stopped and did not resume treatment.

Table 1
Table 1:
Adherence by Month of Intervention

Percent Adherence and Participant Concerns

In all, 87 participants (88.8%) were at least 95% adherent at baseline and 11 (11.2%) were not. In an as-treated analysis, 88 of the 98 participants (92.9%) were 95% adherent or more at the time of their final telephone contact.

Whether or not participants remained fully adherent to treatment, a substantial number did express concerns that represented potential barriers to adherence. Between 0 and 10 concerns were reported per participant (Mdn = 1, SD = 1.9). The two most commonly reported areas of concern were ADEs (97 instances) and treatment logistics (48 instances), with a few participants also reporting concern about feeling too sick or too depressed to take medication (11 instances) or concern about the cost of treatment (6 instances). The number of concerns reported was not correlated to age, sex, or treatment complexity (number of other medications taken or number of comorbid conditions; all ps > .21). Furthermore, the number of concerns reported was also unrelated to the number of ADEs a participant reported, Spearman r = .14, p = .17.

Intervention Effect

Participants' adherence at the 6-month study endpoint was compared with an expected 6-month ART adherence rate using the values shown in Table 1. Expected population rates were derived from the extensive ART adherence literature. Although ART adherence rates are highly variable in published reports, the rates used here reflect study samples and meta-analytic results that are considered comparable to the general practice settings in which this clinical effectiveness study was conducted. The percentage of participants still receiving ART (persistence) was 76% at 6 months after the start of treatment, and this was significantly higher than the expected rate of 50%, binomial z = 3.12, p = .001, effect size (ϕ) = .25. Similarly, the percentage of participants reporting a minimum of 95% adherence at 6 months was significantly higher than the expected rate, binomial z = 3.12, p = .001, effect size (ϕ) = .25. Figure 2 shows participants' adherence over time in comparison with expected population rates.

Figure 2. Treatment persistence over time.
Figure 2. Treatment persistence over time.

Predictors of Nonadherence

Participants' baseline adherence and self-efficacy for treatment were significantly correlated, Spearman r = .42, p < .001. Each participant's average level of adherence over the course of the study was inversely related to the number of concerns he or she expressed about treatment, Spearman r = −.43, p < .001. However, participants' average level of adherence was not significantly related to the number of comorbid diagnoses reported, number of additional medications taken, or number of ADEs reported (all ps > .17). Similarly, baseline adherence was not related to the demographic variables age or sex (all ps > .55).


Among participants prescribed ART for HIV, a telephone support program using psychological counseling techniques resulted in a greater percentage of participants at or above 95% adherence for up to 6 months than expected for PLWH starting a new course of treatment based on earlier research. Furthermore, the intervention was successful in retaining two thirds of participants who did not initially start treatment at the time it was prescribed and 5 out of 12 participants who at least temporarily discontinued medication during the course of the study.

Although attrition from the program over time was a clinical concern, analyses suggest that differential attrition cannot explain the result of improved adherence: Participants who attrited were no different in baseline adherence from those who completed the full program, and attrition was not related to participants' demographics, clinical characteristics, or baseline level of confidence in their ability to follow the ART regimen. Program nurses reported that disconnected telephone numbers and other contact difficulties were common, which may partially explain the high attrition rate. Given the high percentage of participants lost to follow-up, self-selection remains a potential limitation of this study. The authors also were unable to evaluate whether participants were representative of underserved groups with HIV because race/ethnicity data were not collected. Because of these limitations, the results may generalize only to those PLWH who are able to participate in and benefit from a telephone intervention. Given that all participants were treated with the same medication regimen throughout this study, changes in the treatment are not a plausible alternative explanation for the obtained improvement in adherence.

Clinical and Research Implications

Results of the current study have clinical relevance for individual PLWH as well as potential relevance for public health. The ScriptAssist telehealth intervention used in this study is relatively brief and has the potential for community based use because it can be delivered without the need for specially trained counselors at each physical location where care is provided. As described previously, the ScriptAssist program is based on the premise that psychological factors are the primary cause of nonadherence, and the intervention therefore focuses on participants' beliefs and reactions to treatment. This theoretical premise was supported by significant relationships between psychological predictors (self-efficacy and participant concerns) and treatment adherence, with an absence of significant relationships between ART adherence and medication regimen characteristics (number of adverse events reported, number of comorbid conditions, or number of other medications taken) or participant demographics (age or sex). This pattern of results supports the ScriptAssist program's overall strategy of focusing primarily on PLWHs' psychological experiences and treatment motivations to promote adherence. A suggestion for future research is to obtain baseline measures of other psychological characteristics beyond self-efficacy (e.g., depression, treatment motivation, strength of relationships with health care providers) that might predict nonadherence.

The current results support those from another recent study (Reynolds et al., 2008) that used psychologically based telephone counseling to improve ART adherence. In that research, which was conducted as part of an AIDS Clinical Trials Group medication study, telephone counseling consisted of a similar number and length of calls based on self-regulation theory. This randomized trial showed improved adherence even in a group of research participants who already had extremely high adherence: 99% in the intervention group versus 97% in the control group. The current study suggests that a similar approach can be effective in a nonselected general clinic population of PLWH. This effectiveness study's results are also consistent with data from efficacy studies of cognitive-behavioral and motivational interviewing interventions (DiIorio et al., 2008; Golin et al., 2006; Holzemer et al., 2006; Koenig et al., 2008) and show that these research-based counseling techniques can be successfully applied in general care settings via telehealth delivery. Telephone delivery of adherence interventions may enhance fidelity to the underlying psychological counseling model by allowing RNs to focus exclusively on counseling as opposed to other aspects of care. Furthermore, because call center-based nurses can work with a large volume of similar participants recruited from multiple locations, they may have greater experience working with HIV than clinic-based providers, particularly in low prevalence areas or in care settings that do not focus exclusively on HIV.

Limitations and Suggestions for Future Research

Although this was a community based evaluation of a clinical program with minimal exclusion criteria, one potential limitation to external validity was a high baseline level of adherence among participants, which may indicate underrepresentation of the groups most in need of adherence support. Another limitation is that participant race/ethnicity data were not collected, so it was impossible to determine the potential impact of health disparities that are present in minority populations with HIV. This is an important potential source of selection bias that future studies with PLWH should address. Other potential types of selection bias could be ruled out based on available data: Participants were representative of PLWH nationally in terms of sex, and were similar to PLWH who were referred to the program but did not participate in terms of both sex and age. A methodological limitation was the lack of a comparison group that did not receive telephone support, so this study cannot rule out alternative explanations such as history or maturation effects. These concerns are mitigated to some extent by the broad literature on ART adherence showing that adherence generally decreases over time to a greater extent than that seen in the current study and that adherence is relatively difficult for ART regimens compared with treatments for other chronic diseases. Finally, although extensive data now support the use of clinical interviews as a valid measure of medication adherence, there remains the possibility that participants could have been dishonest about their medication adherence.


Although this study joins several others that are suggestive rather than conclusive in their support for telehealth counseling support in promoting ART adherence, analysis of potential adherence predictors also supported the psychological approach to adherence used in the current study. Telehealth nurse counseling is a promising modality for treatment delivery that can be implemented as an addition to standard clinical care. Further study using a randomized controlled trial design is needed to verify that psychologically based telehealth interventions have a direct causal role in improving adherence to ART.

Clinical Considerations

  • Medication adherence remains a challenge in HIV, but it can be facilitated by cognitive-behavioral and motivational interviewing counseling methods.
  • Telephone counseling is an effective method to translate research-based counseling methods into care settings.
  • Psychological variables appear to be particularly important in predicting ART adherence.


The authors thank Judy Spencer, RN, who helped develop clinical materials for the intervention manual used in this study. This study was supported by research contract #0506-010-PC with CenCorp Health Solutions/ScriptAssist Medication Adherence Programs.


Abel, E., Painter, L. (2003). Factors that influence adherence to HIV medications: Perceptions of women and health care providers. Journal of the Association of Nurses in AIDS Care, 14, 61-69.
Bangsberg, D. R., Porco, T. C., Kagay, C., Charlebois, E. D., Deeks, S. G., Guzman, D. (2004). Modeling the HIV protease inhibitor adherence-resistance curve by use of empirically derived estimates. Journal of Infectious Disease, 190, 162-165.
Barfod, T. S., Gerstoft, J., Rodkjaer, L., Pedersen, C., Nielsen, H., Moller, A. (2005). Patients' answers to simple questions about treatment satisfaction and adherence and depression are associated with failure of HAART: A cross-sectional survey. AIDS Patient Care and STDs, 19, 317-325.
Centers for Disease Control and Prevention (2006). HIV/AIDS Surveillance Report: HIV infection and AIDS in the United States and dependent areas, 2005. Washington, DC: Author.
Cook, P. F. (2006). Adherence to medications. In W. T. O'Donohue & E. R. Levensky (Eds.), Promoting treatment adherence: A practical handbook for health care providers (pp. 183-202). Thousand Oaks, CA: Sage.
Cook, P. F. (2008). Patients' and health care practitioners' attributions about adherence problems as predictors of medication adherence. Research in Nursing and Health, 31, 261-273.
Cook, P. F., Emiliozzi, S., McCabe, M. (2007). Telephone counseling to improve osteoporosis treatment adherence: An effectiveness study in community practice settings. American Journal of Medical Quality, 22, 445-456.
Cook, P. F., Emiliozzi, S., Waters, C., El-Hajj, D. (2008). Effects of telephone counseling on antipsychotic adherence and ED utilization. American Journal of Managed Care, 14, 841-846.
De, Bruin M., Hospers, H., de, Borne H. W., Kok, G., Prins, J. (2005). Theory and evidence-based intervention to improve adherence to antiretroviral therapy among HIV-infected patients in the Netherlands: A pilot study. AIDS Patient Care and STDs, 19, 384-394.
Deloria-Knoll, M., Chmiel, J. S., Moorman, A. C., Wood, K. C., Holmberg, S. D., Palella, F. J. (2004). Factors related to and consequences of adherence to antiretroviral therapy in an ambulatory HIV-infected patient cohort. AIDS Patient Care and STDs, 18, 721-727.
DiIorio, C., McCarty, F., Resnicow, K., Holstad, M. M., Soet, J., Yeager, K., et al. (2008). Using motivational interviewing to promote adherence to antiretroviral medications: A randomized controlled study. AIDS Care, 20, 273-283.
Douglass, J. L., Sowell, R. L., Phillips, K. D. (2003). Using Peplau's theory to examine the psychosocial factors associated with HIV-infected women's difficulty in taking their medications. Journal of Theory Construction and Testing, 7, 10-17.
Enriquez, M., Lackey, N. R., O'Connor, M. C., McKinsey, D. S. (2004). Successful adherence after multiple HIV treatment failures. Journal of Advanced Nursing, 45, 438-446.
Golin, C. E., Earp, J., Tien, H. -C., Stewart, P., Porter, C., Howie, L. (2006). A 2-arm, randomized, controlled trial of a motivational interviewing-based intervention to improve adherence to antiretroviral therapy (ART) among patients failing or initiating ART. Journal of Acquired Immune Deficiency Syndromes, 42, 42-51.
Holzemer, W. L., Bakken, S., Portillo, C. J., Grimes, R., Welch, J., Wantland, D., et al. (2006). Testing a nurse-tailored HIV medication adherence intervention. Nursing Research, 55, 189-197.
Ickovics, J. R., Cameron, A., Zackin, R., Bassett, R., Chesney, M., Johnson, V. A. (2002). Consequences and determinants of adherence to antiretroviral medication: Results from Adult AIDS Clinical Trials Group protocol 370. Antiviral Therapy, 7, 185-193.
Koenig, L. J., Pals, S. L., Bush, T., Palmore, M. P., Stratford, D., Ellerbrock, T. V. (2008). Randomized controlled trial of an intervention to prevent adherence failure among HIV-infected patients initiating antiretroviral therapy. Health Psychology, 27, 159-169.
Lu, M., Safren, S. A., Skolnik, P. R., Rogers, W. H., Coady, W., Hardy, H., et al. (2008). Optimal recall period and response task for self-reported HIV medication adherence. AIDS and Behavior, 12, 86-94.
Lucas, G. M. (2005). Antiretroviral adherence, drug resistance, viral fitness and HIV disease progression: A tangled web is woven. Journal of Antimicrobial Chemotherapy, 65, 413-416.
Maisels, L., Steinberg, J., Tobias, C. (2001). An investigation of why eligible patients do not receive HAART. AIDS Patient Care and STDs, 15, 185-191.
Parienti, J. -J., Massari, V., Reliquet, V., Chaillot, F., Le, Moal G., Arvieux, C., et al. (2007). Effect of twice-daily nevirapine on adherence in HIV-1-infected patients: A randomized controlled study. Neurosurgery & Psychiatry, 78, 2217-2222.
Reynolds, N. R., Testa, M. A., Su, M., Chesney, M. A., Neidig, J. L., Frank, I., et al. (2008). Telephone support to improve antiretroviral medication adherence: A multisite, randomized controlled trial. Journal of Acquired Immune Deficiency Syndromes, 47, 62-68.
Rigsby, M. O., Rosen, M. I., Beauvais, J. E., Cramer, J. A., Rainey, P. M., O'Malley, S. S., et al. (2000). Cue-dose training with monetary reinforcement: Pilot study of an antiretroviral adherence intervention. Journal of General Internal Medicine, 15, 841-847.
Simoni, J. M., Pearson, C. R., Pantalone, D. W., Marks, G., Crepaz, N. (2006). Efficacy of interventions in improving highly active antiretroviral therapy adherence and HIV-1 RNA viral load. Journal of Acquired Immune Deficiency Syndromes, 43, S23-S35.
Valenti, W. M. (2001). Treatment adherence improves outcomes and manages costs. The AIDS Reader, 11, 77-80.

adherence; counseling; HIV; medication; self-efficacy

© 2009Elsevier, Inc.