Effective antiretroviral therapy (ART) has led to durable viral load suppression and reduced morbidity and mortality.1,2 However, these benefits are dependent on high levels of patient adherence, which is difficult, especially over time.3,4 Although the accepted definition of adherence for nearly all other chronic diseases is ≥80% of doses taken, early studies suggested that near-perfect ART adherence (>95%) was necessary to achieve these positive outcomes.5,6 More recent research indicates that the required level of adherence associated with viral suppression may be lower than originally thought and may vary with the type of ART being taken.7 Nevertheless, poor adherence leads to the development of resistance8,9 and subsequent reduction in viable treatment options for patients9 and is therefore an important clinical issue.
HIV treatment failure without resistance is usually attributable to problems with adherence.10,11 Moreover, poorer adherence may result in treatment failure and counteractively increase drug costs by necessitating the use of more expensive ARVs.11 The economic burden of regimen failure and HIV disease progression is huge. Recently, a study estimated the economic burden of regimen failure and highlighted “the importance of ensuring optimal initial therapy choices and regimen succession.”12 Therefore, investment in adherence improvement interventions may be worthwhile to reduce expensive burden of treatment failure.
Over the last decade, an array of behavioral strategies to increase adherence to ART have been examined. Although several experimental interventions have been associated with improved adherence and/or HIV RNA suppression, findings have been inconsistent across studies.13–18 Most importantly, which intervention components are responsible for observed adherence gains and the costs associated with implementing them is still unclear. Clinicians are left to their own devises to determine which intervention aspects are worth the expense of implementing in their settings, and researchers lack critical cost data to guide the planning of research studies.
Although some cost analyses of behavioral ART adherence interventions have been performed, they rely on simulated models and lack real-world cost assessment.19–22 Although the results of these studies suggest positive cost-effectiveness results of behavioral interventions, the data have limited utility in guiding the design of contemporary adherence interventions for research or practice. Although conventional methods were used, simulation models are based on assumptions and extrapolations. Other shortcomings include outdated source costs, observed therapy costs based on tuberculosis treatment adherence models, and models based on obsolete ART regimens and guidelines.19–22 Likewise, no study has analyzed potential costs of utilizing an electronic drug monitor (the current gold standard) to verify adherence rates.
Given the deleterious impact and cost of nonadherence, clinic settings are going to implement adherence interventions making real-world cost information critical. Both behavioral interventions and modified directly observed therapy (mDOT) have been shown to be feasible adherence interventions in a variety of populations.23–36 Many studies have used electronic drug monitoring (EDM) systems to track adherence; however, few studies have also reported the actual costs of interventions combined with EDM.37–42 Consequently, costs of improving adherence using behavioral interventions, especially when utilizing mDOT and EDM, are underreported in the literature. The goal of this cost analysis was to provide a real-world perspective on not only the costs associated with behavioral adherence intervention, mDOT, and using EDM in a clinical trial, but also the potential costs associated with adapting these interventions and EDM for usual clinical care. By separating out the costs of the adherence intervention components, the research related costs, and adherence monitoring costs, individuals can make informed decisions when selecting adherence support pieces to transfer to future studies or practice.
Project MOTIV8 (R01 MH068197) was a multisite 3-arm randomized controlled ART adherence trial.43 HIV-positive participants (n = 204) from 6 community HIV clinics in Kansas City were randomized to standard care (SC), enhanced counseling (EC), or EC and mDOT interventions. The patients' demographic characteristics are presented in Table 1.
Project MOTIV8 was designed to examine whether EC combined with mDOT is more effective than EC alone or SC for increasing adherence to ART over 48 weeks.43 The length of the EC and mDOT interventions was 24 weeks. The control arm received SC, which was consistent with the treatment that any HIV-infected patient on ART could expect to receive in the United States. The primary adherence measure used for all the participants was EDM (http://www.aardex.ch).
The direct costs and time costs associated with Project MOTIV8 were identified and analyzed in this study. The costs related to planning of the trial, research effort for study investigators, research data analysis activities, and other research activities that did not directly impact the participants were not included in this economic evaluation. However, participant-related study costs, such as the study evaluations, recruitment, use of EDM and incentive payouts were included. All direct costs and time costs associated with the study were analyzed from the perspective of an agency that may wish to implement these strategies and are consistent with recently reported economic evaluation studies.44–48
The direct costs included hardware, software, printing, advertising, mileage, and study participation incentive items and payments. Items and services (eg, mileage) that directly impacted the intervention were included in this section. Mileage was based upon the federal rate of reimbursement, which averaged $0.50/mile for an average distance of 13 miles driven to participant homes per visit to conduct counseling sessions, evaluations and/or mDOT. The incentive items (eg, mug, t-shirt, tote bag, folder, keychain, and pen) and payments were distributed to participants upon completion of study evaluations. Both research planning costs and implementation costs were considered for calculating activity and item-based unit costs.
Activity-Based Time Costs
We estimated the amount of time required to carry out each study-related activity and obtained salary information spanning the 5 years of the study so that initial time costs and costs associated with counseling could be determined for each aspect of the intervention. The time-related cost centers were training, evaluation, recruitment, EC (counseling time, setup time, driving time), and observed therapy (driving time and 4 types of visits: observed, electronically recorded, phone, and delivered). To determine staffing time cost, the average yearly salary from the middle of 2004 through the middle of 2009 was calculated by taking the total compensation received by each of the research staff and health educators over the course of the study and dividing by the number of years they worked on the study. The average pay per minute was then calculated based on a 52-week year, 5 work days a week, and 8-hour work days. The research staff and health educators were trained by the study investigators, which included Pharmacists, experts (PhDs), and Psychologists. The salaries can be adjusted to be commensurate with the type of individuals chosen to implement the selected activity in other research or practice settings.
Initial Time Costs
The 3 cost centers under initial time cost were training, evaluation, and recruitment. Training time was a 1-time cost at the beginning of the project involving instruction of 3 staff members in the methods to be employed in carrying out evaluations and interventions. Each staff member received 84.5 hours of training divided among the following categories: HIV 101 for 20 hours, general and safety training for 1 hour, how to perform mDOT for 5.5 hours, and how to carry out EC for 58 hours. Evaluation time cost was the baseline time cost shared by all 3 arms of the study. The participants completed 7 evaluation sessions that comprised demographic, adherence, psychosocial, and medical self-report questionnaires administered via Audio Computer Assisted Self Interview and by staff. Also included were chart reviews and EDM data collection that accounted for the majority of the evaluation time cost. Recruitment time cost involved the amount of time it took to approach all potential participants. Although not every person approached was enrolled, the investigators felt it was necessary to capture the time cost for each attempt and each successful enrollment as both contributed to the overall cost.
The total amount of time the study staff spent on each intervention activity during the entire span of the study was calculated and the average amount of time spent per person. Each study participant received a specified amount of counseling and intervention based upon their study group assignment. The participants in SC received no intervention. Delivery of the EC intervention involved 11 sessions during the first 24 weeks, which were conducted in their home or other desired location. Enhanced counseling sessions focused on increasing motivation and skills for adherence covering a range of topics typically targeted by ART adherence interventions (eg, self-monitoring, problem solving, talking to your doctor).
The mDOT participants received both the 11 EC sessions and mDOT. The mDOT involved direct observation of patients taking 1 of their ART doses Monday through Friday for the first 16 weeks, tapered down by 1 d/wk for weeks 17–19, and ending with 1 observation per week for weeks 20 through 23 for a total of 93 total possible mDOT visits per person. The observed dose was sometimes monitored by phone or other electronic method when direct observation was not possible. Driving time costs associated with the EC sessions, direct observations, and delivery of ART for unobserved doses were also incorporated.
Method of Analysis
Microsoft Excel was used to generate the spread sheets for tabulating data, and calculating overall, and patient-specific time and dollar costs. It was also utilized for conducting a sensitivity analysis on which a tornado diagram was generated. The sensitivity analysis was carried out examining the costs and savings associated with 4 different scenarios: excluding all driving related costs in scenario 1 if exclusively electronic methods of delivery could be utilized, using a reduced base labor cost of $0.23/min in scenario 2, which is more indicative of what typical clinic staff are usually paid,36 excluding the cost of EDM in scenario 3 because clinics may not wish to incur the costs associated with that particular type of monitoring, and excluding the cost of the study-related evaluations in scenario 4 because this was empirical research and real-world clinics may not have the time, manpower, funds to do this kind of study-related evaluation. We also calculated the total costs and savings of combining all 4 scenarios with and without direct costs. The tornado diagram (Fig. 1) generated from the sensitivity analysis used a central axis of $881 based upon the cost of SC, thus allowing the diagram to show how costs associated with each of the intervention arms changed in each scenario.
The base cost of labor within the study was calculated as $0.36/min ($21.60/hour). Total calculated direct costs were $126,068 ($618 per patient). Total activity-based time costs totaled $90,658, which was distributed among 3 cost centers: initial time costs $53,590 ($262 per patient), EC cost $18,427 ($137 per patient), and mDOT cost $18,638 ($291 per patient). The total cost per patient per arm of the study was SC = $881 per patient, EC = $1018 per patient, EC/mDOT = $1309 per patient.
Total direct costs, which are summarized in Table 2, equaled $126,618. The direct costs generated from hardware (desktops, laptops, and PDAs), software (Microsoft programs, statistical programs, EDM program), and EDM ($173 for each) was $68,520. Costs for incentive Items and study evaluation payments (Baseline/visit 1 = $20, visit 2 = $20, visit 3 = $40, visit 4 = $20, visit 5 = $65) added $34,870. Other direct costs included printing $570, advertising $298, and mileage $21,810.
Activity-Based Time and Cost by Category
The total individual participant cost based on study group assignment is displayed in Table 3, whereas the total and per person time and costs are displayed for each study-related activity are included in Table 4. The initial costs and intervention-based costs were included in the activity-based time cost category.
Initial Time Cost
The total initial cost was $53,590 ($262 per patient) and involved 148,864 minutes (730 minutes per patient). Initial costs included training, participant evaluation, and recruitment costs. Training costs included 3 staff members, who each received 84.5 hours of training at a total cost of $5470, or $1823 per staff member. Evaluation time costs covered each of the 204 study participants, who ideally, participated in 7 study evaluations. This resulted in an average of 81.6 minutes each at a cost of $29.38 per session; 571 minutes total per person at a cost of $205.63; and total time of 116,554 minutes for the entire study at a total cost of $41,959. Recruitment time costs consisted of 1710 attempts to recruit patients with an average of 10 minutes per recruitment session and 204 participants successfully enrolled in the study. A time expenditure of approximately 83.8 minutes was required for 1 enrollment. The cost of 1 recruitment attempt was $3.60, but the cost of 1 successful enrollment was $30.18.
Enhanced Counseling: One hundred and thirty-four participants completed EC within the study for a total time of 51,188 minutes (382 minutes per patient) and a total cost of $18,427 ($137 per patient). EC was further subdivided into the 3 subsections of counseling, setup, and driving. The EC intervention consisted of 11 sessions varying in length from 10 to 25 minutes for a total time of 26,800 minutes (200 minutes per patient) and cost of $9648 ($72 per patient). Setup time was 7 minutes for each of the 11 sessions with a total time of 10,318 minutes (77 minutes per patient) and cost of $3714 ($27.72 per patient). Driving time for EC consisted of 938 separate trips with an average time of 15 minutes for a total time of 14,070 minutes (105 minutes per patient) and cost of $5065 ($37.80 per patient, $5.40 per trip).
The mDOT: Sixty-four participants completed mDOT within the study for a total time of 51,773 minutes (809 minutes per patient) and total cost of $18,638 ($291 per patient). Although each participant in the mDOT group should have received 93 delivered and observed doses by protocol, they actually received varied combinations of dose monitoring and medication delivery visits; these included directly observed, “phone contacts' (participant ingested medication during a study staff initiated phone call at the predetermined dose time), ‘med delivery’ (meds delivered outside of target dosing time and participant reported by phone/text when ingested), and ‘PDA visits’ (meds delivered outside of target dosing time and participant retrospectively reported on all unobserved doses using PDA).”43 Directly observed visits were partnered with driving time. There were 1927 directly observed visits with an average of 5.2 minutes (approximate cost of $1.87 per visit) of direct patient contact. The total direct patient contact time was 10,020 minutes with a total cost of $3607. Each of these visits was coupled with a driving time, which averaged 15 minutes and added to the direct contact time for a total time of 28,905 minutes and cost of $10,406. Thus, the average directly observed visits accounted for 20.2 minutes of staff time per patient at an average cost of $7.27 per visit. There were 124 electronically monitored visits with a total time of 496 minutes and cost of $179. There were 1324 phone visits with a total time of 5295 minutes and cost of $1907. There were 1764 visits to deliver medications without a dose observation, which totaled 7096 minutes and a cost of $2540. Each of these kinds of visits: phone, electronic, and delivered averaged 4 minutes in length with an average cost of $1.44 per visit.
The results of the sensitivity analysis are highlighted in Table 4. There was a 10%–23% cost savings associated with each of the 4 scenarios. The biggest cost savings came from scenario 4, which eliminated the initial cost of the study evaluations saving $206 per patient with no difference between study arms. Likewise, scenario 3 showed that omission of adherence monitoring using EDM could save $173 per patient. Scenarios 1 and 2 both showed cost savings, which varied between the study arms. Cutting out driving costs resulted in the most variable impact between the 3 study arms. Exclusion of driving costs impacted both direct and activity-based time costs and resulted in the following savings: SC = $107, EC = $145, EC/mDOT = $307. Similarly, reducing the cost of labor from $0.36 to $0.23/min resulted in variable cost savings across the groups by affecting the time-based initial cost and counseling costs. The savings for each group were as follows: SC = $95, EC = $144, EC/mDOT = $249. The tornado diagram in Figure 1 shows a graphical representation of how each sensitivity assumption reduced costs compared with each other and the resulting comparative costs for each group. To demonstrate additional plausible scenarios, the total savings of adding all of the sensitivity scenarios together and excluding all direct costs were included.
This cost analysis shows the cost components associated with implementation and delivery of 2 established adherence interventions, behavioral counseling and mDOT. in a controlled trial. We also examined the costs of utilizing EDM and conducting study evaluations. Thus, this report is the first to provide detailed and relevant cost information that can inform decisions about implementation of adherence interventions in clinical and research settings.
This analysis compares favorably with that of previous studies, which demonstrated desirable thresholds of cost effectiveness for adherence-improving interventions ranging from $1200 to $1600/year.19–21 Likewise, both of the MOTIV8 interventions fall on either side of the Freedberg et al22 study, which cost $340/year ($28/month). The interventions themselves turn out to be some of the least expensive components (Table 4). The individual patient cost for EC was $137 for the 24-week intervention or $23/month if continued for a full year. EC costs included the cost of driving to participant homes or desired location to provide EC sessions. These costs could be eliminated by combining the intervention with scheduled visits to the clinic or offering them via electronic methods of delivery. The EC monthly cost assumes that the interventions are carried out on continuous 24-week cycles throughout a year, whereas a 1-year price based on providing a single 24-week intervention would correspond with $137/year or $11/month. A sizeable portion of the EC and mDOT expenses resulted from driving to participant homes or preferred locations to conduct counseling sessions and observe medication doses. As shown by scenario 1 in the sensitivity analysis, there would be substantial cost savings if these components were offered at the clinic, by phone, email, or other real-time media. Personnel cost is another variable amenable to savings when designing adherence interventions. Well-trained and supervised community healthcare workers could serve as qualified adherence support staff, and their salaries would be considerably lower than our study staff's.11,14–16 Alternatively, clinics may have existing staff who are able to absorb adherence treatment duties that are brief and/or delivered remotely. In scenario 2, the assumption of personnel compensation rate of $0.23/min (∼$13.71/hour) is based upon the 2000 HIV Costs and Utilization Survey conducted by Agency for Healthcare Research and Quality.49 An alternate staffing cost can be found within the 2009 US Bureau of Labor Statistics National Compensation Survey, which lists total compensation averages for healthcare and social assistance service workers within the $16/hour range.50 This scenario also likely provides a more realistic idea of the true clinical costs for personnel as compared with the much higher $0.36/min (∼$22/hour) costs seen in this study.
A significant difference between this study and other cost analysis studies is the capture of real-world adherence data using EDM. The EDM has a preferred adherence measure among some researchers.51 It may have some advantages to supply information about therapeutic drug coverage and to be used in providing patient adherence feedback.42,52 However, EDM is not without its own concerns including cost, which was $173 per patient. Clinicians may not choose to include EDM when implementing adherence interventions for their entire clinic population. Rather, it may be wise to include limited use of EDM for specific patients that may benefit most. For example, EDM may be helpful in distinguishing whether a patient is nonadherent or a treatment nonresponder. Further, reviewing EDM results with patients during adherence intervention may be useful as it has been associated with increased adherence.42,52 The incremental estimates for EDM cost provided here will greatly facilitate informed decision making. Potentially less expensive methods of assessing adherence include patient self-report and pharmacy adherence measures.53
The costs associated with study evaluations may have limited real-world application and likely should be excluded from decisions about clinical applications. As previously noted, the greatest cost savings in this analysis resulted from elimination of the study-related costs. These consisted of costs solely related to meeting research needs and were unrelated to providing the adherence intervention. However, this is the first study to report detailed research specific cost information, which is critically important to investigators who are planning new adherence studies.
Due to the variability in resources across HIV treatment settings, it was important to consider the costs and savings that would come from a bare bones adherence intervention, which would combine all 4 cost saving scenarios with or without including direct costs of implementation. Surprisingly these last scenarios may not only be applicable to resource poor settings but also well-funded clinics. For clinics that already have a significant amount of available technology, most of the direct costs for computer equipment and software would already be covered. Also, for those clinics interested in implementing the interventions at their most basic level, performing direct observation through phone calls and electronic devices, may see their costs decrease over time as the initial cost for training paid personnel extends over several years of service. For clinics utilizing volunteers to deliver intervention (eg, peer support programs), the key cost is represented by the time required by paid staff to provide training rather than the costs associated with carrying out the interventions. Although most clinics would likely not employ incentive-based systems, there is limited evidence that incentivizing good adherence may have a positive impact.54 Starting with an empirically supported adherence intervention (ie, cognitive behavioral counseling, mDOT) and then taking an a la carte approach for the selection of other potentially useful services (eg, EDM, evaluations, incentives, in-person direct observation, and travel) will allow clinics to tailor their interventions most effectively and cost efficiently to their patient populations.
One possible limitation to our cost analysis is that costs were not adjusted to account for missed or modified visits. Since this cost analysis is based on actual study data, the total costs reflect completed visits, however; the per patient costs were based on the number of scheduled visits. In some instances, this could result in the underestimation of the total cost and overestimation of the per patient cost. The costs associated with providing mDOT were most vulnerable to these variations. To accommodate for participant schedules (eg, work, vacations), confidentiality, and other situations, the mDOT participants ended up with varied numbers and combinations of observed, delivered, phone, and electronically monitored dose observation visits. Thus, the costs for this intervention may underestimate the true costs if all 93 of the ideally planned mDOT visits were completed. However, it is unlikely that any setting would see 100% appointment completion rate, so the rate observed here provides a reasonable estimate for planning purposes. Another limitation of this cost analysis is that the data were retrospectively collected. However, all cost and time data were concurrently and accurately recorded and available for retrieval. Lastly, this study was not able to examine costs from a societal perspective. In future, cost-effectiveness analysis of intervention with adherence outcome data would be beneficial. These cost data reported in this study are extremely useful for decision making and planning real-world translations of these interventions.
This economic analysis provides guidance on personnel and materials costs for clinicians and researchers wishing to employ behavioral interventions to improve ART adherence for HIV-infected patients. The cost of EC and mDOT compared favorably with other cost-effective interventions. To our knowledge, this is the first published cost analysis of these combined behavioral interventions to improve ART adherence. These cost data are extremely useful for planning real-world translations of these interventions and future studies.
1. Crum NF, Riffenburgh RH, Wegner S, et al.. Comparisons of causes of death and mortality rates among HIV-infected persons: analysis of the pre-, early, and late HAART (highly active antiretroviral therapy) eras. J Acquir Immune Defic Syndr. 2006;41:194–200.
2. Jones J, Taylor B, Wilkin TJ, et al.. Advances in antiretroviral therapy. Top HIV Med. 2007;15:48–83.
3. Ortego C, Huedo-Medina TB, Llorca J, et al.. Adherence to highly active antiretroviral therapy (HAART): a meta-analysis. AIDS Behav. 2011;15:1381–1396.
4. Sethi AK. Adherence and HIV drug resistance. HIV Clin Trials. 2004;5:112–115.
5. Paterson DL, Swindells S, Mohr J, et al.. Adherence to protease inhibitor therapy and outcomes in patients with HIV infection. Ann Intern Med. 2000;133:21–30.
6. Bangsberg DR, Hecht FM, Charlebois ED, et al.. Adherence to protease inhibitors, HIV-1 viral load, and development of drug resistance in an indigent population. AIDS. 2000;14:357–366.
7. Lima VD, Bangsberg DR, Harrigan PR, et al.. Risk of viral failure declines with duration of suppression on highly active antiretroviral therapy irrespective of adherence level. J Acquir Immune Defic Syndr. 2010;55:460–465.
8. Montaner JSG, Reiss P, Cooper D, et al.. A randomized, double-blind trial comparing combinations of nevirapine, didanosine, and zidovudine for HIV-infected patients: the INCAS Trial. Italy, the Netherlands, Canada and Australia Study. JAMA. 1995;279:930–937.
9. Race E, Dam E, Obry V, et al.. Analysis of HIV cross-resistance to protease inhibitors using a rapid single-cycle recombinant virus assay for patients failing on combination therapies. AIDS. 1999;13:2061–2068.
10. Tilghman M, May S, rez-Santiago J, et al.. A combined screening platform for HIV treatment failure and resistance. PLoS One. 2012;7:e35401.
11. Tan J, Sonecha S, Mandalia S, et al.. A survey of patients' willingness to switch from a single-tablet regimen to 2 pills once a day as a cost-saving strategy. J Int AIDS Soc. 2012;15:18388.
12. Snedecor S, Khachatryan A, Nedrow K, et al.. The prevalence and economic burden of 1st-generation NNRTI resistance. J Int AIDS Soc. 2012;15:18204.
13. Amico KR, Harman JJ, Johnson BT. Efficacy of antiretroviral therapy adherence interventions: a research synthesis of trials, 1996 to 2004. J Acquir Immune Defic Syndr. 2006;41:285–297.
14. Simoni JM, Amico KR, Smith L, et al.. Antiretroviral adherence interventions: translating research findings to the real world clinic. Curr HIV/AIDS Rep. 2010;7:44–51.
15. Hart JE, Jeon CY, Ivers LC, et al.. Effect of directly observed therapy for highly active antiretroviral therapy on virologic, immunologic, and adherence outcomes: a meta-analysis and systematic review. J Acquir Immune Defic Syndr. 2010;54:167–179.
16. Hill S, Kavookjian J. Motivational interviewing as a behavioral intervention to increase HAART adherence in patients who are HIV-positive: a systematic review of the literature. AIDS Care. 2012;24:583–592.
17. Ford N, Nachega JB, Engel ME, et al.. Directly observed antiretroviral therapy: a systematic review and meta-analysis of randomized clinical trials. Lancet. 2009;374:2064–2071.
18. Simoni JA, Pearson CR, Panalone DW, et al.. Efficacy of interventions in improving highly active antiretroviral therapy adherence and HIV-1 RNA viral load. J Acquir Immune Defic Syndr. 2006:43;S23–S35.
19. Goldie SJ, Paltiel AD, Weinstein MC, et al.. Projecting the cost-effectiveness of adherence interventions in persons with human immunodeficiency virus infection. Am J Med. 2003;115:632–641.
20. Munakata J, Benner JS, Becker S, et al.. Clinical and economic outcomes of nonadherence to highly active antiretroviral therapy in patients with human immunodeficiency virus. Med Care. 2006;44:893–899.
21. Schackman BR, Finkelstein R, Neukermans CP, et al.. The cost of HIV medication adherence support interventions: results of a cross-site evaluation. AIDS Care. 2005;17:927–937.
22. Freedberg KA, Hirschhorn LR, Schackman BR, et al.. Cost-effectiveness of an intervention to improve adherence to antiretroviral therapy in HIV-infected patients. J Acquir Immune Defic Syndr. 2006;43(suppl 1):S113–S118.
23. Bangsberg D, Mundy L, Tulsky J. Expanding directly observed therapy: tuberculosis to human immunodeficiency virus. Am J Med. 2001;110:664–666.
24. de Bruin M, Hospers HJ, van den Borne HW, et al.. Theory- and evidence-based intervention to improve adherence to antiretroviral therapy among HIV-infected patients in the Netherlands: a pilot study. AIDS Patient Care STDS. 2005;19:384–394.
25. de Bruin M, Hospers HJ, van Breukelen GJ, et al.. Electronic monitoring-based counseling to enhance adherence among HIV-infected patients: a randomized controlled trial. Health Psychol. 2010;29:421–428.
26. DiIorio C, McCarty F, Resnicow K, et al.. Using motivational interviewing to promote adherence to antiretroviral medications: a randomized controlled study. AIDS Care. 2008;20:273–283.
27. Goggin K, Liston RJ, Mitty JA. Modified directly observed therapy for antiretroviral therapy: a primer from the field. Public Health Rep. 2007;122:472–481.
28. Flanigan TP, Taylor LE, Mitty JA. Use of community-based, directly observed therapy for HIV infection: lessons learned for treatment of hepatitis C virus infection. Clin Infect Dis. 2005;40(suppl 5):S346–S348.
29. Hart JE, Jeon CY, Ivers LC, et al.. Effect of directly observed therapy for highly active antiretroviral therapy on virologic, immunologic, and adherence outcomes: a meta-analysis and systematic review. J Acquir Immune Defic Syndr. 2010;54:167–179.
30. Macalino GE, Hogan JW, Mitty JA, et al.. A randomized clinical trial of community-based directly observed therapy as an adherence intervention for HAART among substance users. AIDS. 2007;21:1473–1477.
31. Mitty JA, Macalino GE, Bazerman LB, et al.. The use of community-based modified directly observed therapy for the treatment of HIV-infected persons. J Acquir Immune Defic Syndr. 2005;39:545–550.
32. Mitty JA, Huang D, Loewenthal HG, et al.. Modified directly observed therapy: sustained self-reported adherence and HIV health status. AIDS Patient Care STDS. 2007;21:897–899.
33. Smith SR, Rublein JC, Marcus C, et al.. A medication self-management program to improve adherence to HIV therapy regimens. Patient Educ Couns. 2003;50:187–199.
34. Sorensen JL, Mascovich A, Wall TL, et al.. Medication adherence strategies for drug abusers with HIV/AIDS. AIDS Care. 1998;10:297–312.
35. Tuldra A, Fumax CR, Ferrer MJ, et al.. Prospective randomized two-arm controlled study to determine the efficacy of a specific intervention to improve long-term adherence to highly active antiretroviral therapy. J Acquir Immune Defic Syndr. 2000;25:221–228.
36. Weber R, Christen L, Christen S, et al.. Effect of individual cognitive behaviour intervention on adherence to antiretroviral therapy: prospective randomized trial. Antivir Ther. 2004;9:85–95.
37. Arnsten JH, Demas PA, Farzadegan H, et al.. Antiretroviral therapy adherence and viral suppression in HIV-infected drug users: comparison of self-report and electronic monitoring. Clin Infect Dis. 2001;33:1417–1423.
38. Deschamps AE, de Graeve VD, van Wijngaerden E, et al.. Prevalence and correlated of nonadherence to antiretroviral therapy in a population of HIV patients using medication event monitoring system. AIDS Patient Care STDS. 2004;18:644–657.
39. Farley J, Hines S, Musk A, et al.. Assessment of adherence to antiviral therapy in HIV-infected children using the medication event monitoring system pharmacy refill, provider assessment, caregiver self-report, and appointment keeping. J Acquir Immune Defic Syndr. 2003;33:211–218.
40. Müller AD, Bode S, Myer L, et al.. Electronic measurement of adherence to pediatric antiretroviral therapy in south Africa. Ped Infect Dis J. 2008;27:257–262.
41. Rigsby MO, Rosen MI, Beauvais JE, et al.. Cue-dose training with monetary reinforcement: pilot study of an antiretroviral adherence intervention. J Gen Intern Med. 2000;15:841–847.
42. Hughes D. When drugs don't work: economic assessment of enhancing compliance with interventions supported by electronic monitoring devices. Pharmacoeconomics. 2007;25:621–635.
43. Goggin K, Gerkovich MM, Williams K, et al.. A RCT examining the efficacy of motivational counseling with observed therapy for ART adherence. AIDS and Behavior. 2013; In Press.
44. Drummond MF, O'Brien BJ, Stoddart GL, et al.. Methods for the Economic Evaluation of Health Care Programmes. 2nd ed. Oxford, United Kingdom: Oxford University Press; 1997.
45. Edejer TT-T, World Health Organization. Making Choices in Health: WHO Guide to Cost-Effectiveness Analysis. Geneva, Switzerland: World Health Organization; 2003.
46. Franzini L, Boom J, Nelson C. Cost-effectiveness analysis of a practice-based immunization education intervention. Ambul Pediatr. 2007;7:167–175.
47. Rasu RS, Hunter CM, Peterson AL, et al.. Economic evaluation of an internet-based weight management program. Am J Manag Care. 2010;16:e98–e104.
48. Rianon N, Selwyn B, Shahidullah S, et al.. Cost of health education to increase STD awareness in female garment workers in Dhaka, Bangladesh. Int Electron J Health Educ. 2009;12:135–149.
49. HIV Cost and Services Utilization Study. Public Use Interview Data Set. Rockville, MD: Agency for Health Care Research and Quality; 2000.
50. US Department of Labor Bureau of Labor Statistics. National compensation survey. Available at: http://data.bls.gov
. Accessed March 28, 2010.
51. Liu H, Wilson IB, Goggin K, et al.. MACH14: a multi-site collaboration on ART adherence among 14 institutions. AIDS Behav. 2012: DOI 10.1007/s10461-012-0272-4.
52. Blaschke TF, Osterberg L, Vrijens B, et al.. Adherence to medications: insights arising from studies on the unreliable link between prescribed and actual drug dosing histories. Annu Rev Pharmacol Toxicol. 2012;52:275–301.
53. McMahon JH, Jordan MR, Kelley K, et al.. Pharmacy adherence measures to assess adherence to antiretroviral therapy: review of the literature and implications for treatment monitoring. Clin Infect Dis. 2011;52:493.
54. Rosen MI, Dieckhaus K, McMahon TJ, et al.. Improved adherence with contingency management. AIDS Patient Care STDs. 2007;21:30–40.
Keywords:© 2013 Lippincott Williams & Wilkins, Inc.
adherence; EDM; cost analysis; behavioral intervention; HIV; ART; econ omic evaluation