bs_bs_banner
International Journal of Nursing Practice 2013; 19: 577–583
RESEARCH PAPER
Medication adherence feedback intervention predicts improved human immunodeficiency virus clinical markers Warren A Reich PhD Research and Evaluation Manager, The Family Center, New York, New York, USA
Accepted for publication January 2013 Reich WA. International Journal of Nursing Practice 2013; 19: 577–583 Medication adherence feedback intervention predicts improved HIV clinical markers Thirty-three participants in a human immunodeficiency virus (HIV) medication adherence feedback (MAF) intervention were compared with 58 HIV-positive non-participants in laboratory-tested CD4 and viral load. The intervention provided adherence feedback and counselling based on a visual display from an electronic pill bottle (MEMSTM). Multiple regression controlling for baseline CD4 and showed that postintervention CD4 was higher for MAF participants than for non-MAF participants. Non-MAF participants’ CD4 significantly declined over time. MAF participants were also less likely than non-MAF participants to have a detectable postintervention viral load. Key words: CD4, electronic monitoring, medication adherence, viral load
INTRODUCTION Medical treatment for human immunodeficiency virus (HIV) disease has advanced rapidly and significantly over the past decade. Medication regimens have slowed disease progression and improved survival rate, and HIV-positive persons can now choose from a greater-than-ever number of medications should their existing prescriptions faile
Correspondence: Warren A. Reich, The Family Center, 315 W. 36th Street, New York, NY 10018, USA. Email: wreich@thefamilycenter. org,
[email protected] Warren Reich is the Research and Evaluation Manager at The Family Center in New York City.I wish to thank Paul Boxer, David Holtgrave, Jan Hudis, Andrea Vial, and Dave Nimmons for their comments on an earlier version of this paper, and to Julia Sanchez, Ikponmwosa Edorisiagbon, and Stephanie Mendez for their assistance in data collection. This research was supported in part by a grant from the Pfizer Foundation. doi:10.1111/ijn.12100
or lead to undesirable side effects. The epidemic has accordingly witnessed a paradigmatic shift from a survival orientation to one that promotes healthy living and general well-being. It has been well established that HIV-positive individuals need to maintain stringent adherence to their antiretroviral medications—indeed, most agree no less than 95%—to suppress viral load (vl), boost CD4 counts, and prevent serious life-threatening illness1–3 (but see Shuter4). Yet, optimal adherence remains a significant problem among persons living with HIV/acquired immunodeficiency syndrome (AIDS)5 for a variety of reasons ranging from stress, depressed mood, low self-efficacy, sparse social support and simple forgetfulness.1,6,7 In lieu of a cure for HIV/AIDS, adherence promotion interventions that target these factors have emerged as some of the most effective behavioural weapons for those living with this disease. Adherence monitoring is currently regarded as a key component of a larger repertoire of interventions © 2013 Wiley Publishing Asia Pty Ltd
578
WA Reich
that support the health of HIV-positive persons.8 Such interventions are generally highly cost-effective ways to improve adherence.9–11
THEORETICAL MODEL The present research assumes that one implicit aim of a medication adherence intervention is to engage the client as a collaborator with a vested interest in her or his health. Effective adherence counselling frames communication on medication dosing, side effects, supports and barriers within the context of the client’s unique life circumstances—thereby maximizing the self-relevance of this information. Laboratory research has shown that messages are more systematically processed and better recalled to the extent that they are directly linked to one’s interests, goals and needs.12–14 Consistent with this theoretical model, Noar et al.15 have documented the success of a variety of health interventions that communicate individually tailored information. Personalized feedback has proved to be successful in medication adherence interventions for epilepsy, asthma and mental illness.16–18 A related theoretical assumption draws on selfdetermination theory,19 which hypothesizes a link between autonomous motivation and health-promoting behaviours. This link has been borne out in research on medication adherence generally20 and HIV medication adherence specifically.21 According to this model, clients who actively participate in discussions should be more motivated to maintain adherence than those who simply receive information from an authority. The present intervention sought to provide an autonomy-supportive context by eliciting clients’ accounts of missed doses indicated by electronically generated medication adherence feedback, and their ideas about ways in which identified barriers could be reduced, in a collaborative conversation. This general approach has recently been adopted with positive results in a multifaceted HIV medication adherence intervention.22,23
MEDICATION ADHERENCE FEEDBACK (MAF) INTERVENTION Medication Event Monitoring Systems pill bottle caps (MEMSTM, Aardex Group, Sion, Switzerland) are widely used as measurement devices in adherence research as alternatives to, and sometimes in conjunction with, selfreport adherence measures.5,22,24,25 Some researchers have © 2013 Wiley Publishing Asia Pty Ltd
used electronic monitors as feedback instruments to research participants as well. Three recent studies have shown that delivering such feedback results in improved HIV medication adherence. Smith et al.26 demonstrated that those receiving electronic adherence feedback as part of a behavioural skills ‘self-management’ protocol were far superior at 12 weeks to a control group, with adherence rates of 91% and 37%, respectively. Rosen et al.27 reported a 16-week increase of 15% of HIV medication doses taken in a ‘contingency management’ condition (vs. a decrease of 15% in a supportive-counselling control condition) that included weekly review of electronically reported missed doses and reinforcement for good adherence. A similar intervention was shown to be effective for persons living with HIV in China.28 These results mirror the positive impact of electronically generated feedback on adherence to antihyperglycaemic medication.29 The present analyses document the prospective impact of a brief intervention on laboratory-reported CD4 and vl. The intervention, MAF, was part of a health and wellness programme for HIV-positive persons. It is similar to interventions cited above in that MEMSTM was used to provide tailored HIV medication adherence feedback. Unlike these interventions, however, MAF involved only a single feedback and counselling session. A visual display of 2 weeks’ medication adherence formed the basis of a conversation with staff on the circumstances preceding missed doses, and on problem-solving strategies for avoiding or coping with these situations. It was hypothesized that this brief intervention would predict improved CD4 counts and vl.
METHOD Participants Fifty-four HIV-positive persons (48 female) participated in MAF at The Family Center in New York City. Of these, complete CD4 and viral load data were obtained from 33 (28 female, Mage = 43.38, age range = 32 to 58). Approximately half (47.7%) were African-American, another 47.7% self-identified as ‘more than one race’, and the remainder were ‘unknown’ or unreported. Twenty (45.45%) self-identified as Latino/a. The sample was nearly evenly split between those on a 1 vs. 2 doses per day regimen (48.9% and 51.1%). This variable was not correlated with any measure of adherence (Ps > 0.27), and was therefore not included as a covariate in statistical analyses.
Medication adherence feedback
Procedure Participants were recruited from The Family Center, a community-based social services agency in New York City. Before participating in MAF, all were met by an HIV medication specialist at The Family Center for an individual information session as part of a comprehensive Medical Case Management programme. Two weeks of HIV medication was then loaded into the participant’s medication bottle, which participants were instructed to use to take their HIV medications. MEMSTM caps (affixed to the bottle) electronically record the time and date when they are opened and presumably reflect when the medication was taken. Participants were contacted by phone 1–2 days later to ensure that they were using MEMSTM. After 2 weeks, participants met individually with a staff member for a feedback and counselling session. A graphic illustration was uploaded from the bottle to a laptop computer that displayed the participant’s medication adherence over the past 2 weeks. From this output, the participant and staff member could easily see when a dose was missed. A semi-structured problem-solving discussion ensued that focused on events surrounding missed doses (e.g. argument with partner on that day), why they led to non-adherence (e.g. stressed me out and made me sad) and how to cope with or avoid similar situations in the future. Participants were highly engaged in these interviews, which lasted approximately 60 min.
Medication adherence At baseline, 1 and 2 weeks, we assessed self-reported HIV medication adherence over the prior 7 days beginning with pills missed yesterday, then over the past 1, 2, 3 and 7 days.6 The week 1 self-report was completed by phone. The maximum of these responses was used as an indicator of total number of doses missed over the past week. The number of missed doses was also recorded directly from the MEMSTM output.
579
laboratory reports between MAF participation and September 2008. CD4 counts were square root transformed to correct for a skew in the distributions. A second score indicated whether there was at least one detectable vl in the pre- and postperiods (> 75 copies per mm3; vl-det-pre and vl-det-post). A comparison group was comprised of 58 non-MAF participants (55 women) at The Family Center who had laboratory reports in the same time period. The mean age for the non-MAF group was 42.39 in September 2007 (range = 19 to 66). Half (50.0%) were AfricanAmerican, another 41.4% self-identified as ‘more than one race’, 1.1% as Asian, 1.1% as White,and the remainder ‘unknown’ or unreported. Forty-two (44.7%) selfidentified as Latino/a. The median laboratory report date, September 2007, was used to partition the data into ‘pre’ and ‘post’ periods. Thus, CD4-pre for the non-MAF group was the mean of all laboratory reports between February 2006 and September 2007, and CD4-post from that date through September 2008. The mean withinperson standard deviation of the CD4 counts across both groups was 106.25 and 103.12 for pre and post, respectively. A count of vl-det-pre and vl-det-post was also recorded as it was for MAF participants.
RESULTS Descriptive analyses The number of self-reported and MEMS-reported missed doses was correlated, r (44) = 0.30, P = 0.049. However, the MEMS cap revealed a higher number of missed doses than did self-report, MMEMS = 1.07 and Mself = 0.11, paired t (44) = 3.79, P < 0.001. Also, examination of the MEMS outputs revealed that nonadherence was more common on Saturdays, Sundays, and Mondays (17.0% on each day) than on other days (range = 8.5% to 14.9%).
Main analyses: HIV disease markers CD4 and vl The Family Center tracks laboratory-reported CD4 and vl information as part of routine service provision for HIVpositive clients. These records were examined for MAF participants and a comparison group of non-participants between February 2006 and September 2008. Participants’ CD4-pre was the mean CD4 from all laboratory reports on file between February 2006 and date they began MAF. CD4-post was similarly computed from
Because the MAF and non-MAF groups were not randomly selected, it was necessary to control for any pre-existing differences that could have influenced the outcomes. Potential confounding variables included number of services received from The Family Center (all services, and also those specific to adherence), smoking, drinking, hospitalization, medication side effects, pain, physical limitations, other physical or mental illness and visits to healthcare providers during the study period. The © 2013 Wiley Publishing Asia Pty Ltd
580
WA Reich
Table 1 Predicting CD4-post from MAF participation Cumulative R2
B
SE B
b
t
Model 1 0.74** MAF participant CD4-pre Model 2
1.58 0.82
0.76 0.05
0.12 0.85
2.08* 15.42**
2.02 0.83 -0.31 0.49
0.81 0.05 0.29 0.38
0.15 0.85 -0.06 0.07
2.49* 15.48** -1.09 1.29
0.74** MAF participant CD4-pre Number pre-laboratory reports Number postlaboratory reports Note. * P < 0.05; ** P < 0.01. MAF, medication adherence feedback.
MAF and non-MAF groups were equivalent on each of these indices, mean r = 0.07, all Ps > 0.20. Although the two groups were virtually identical in the total number of laboratory reports (reflecting their time in service with The Family Center, MMAF = 4.59 and Mnon-MAF = 4.60), MAF participants had more pre-laboratory reports and fewer postlaboratory reports than non-MAF participants, Ps < 0.07. Number of pre- and postlaboratory reports, then, were entered as covariates in regression analyses.
CD4 count Multiple regression analysis indicated that MAF participation was uniquely associated with higher CD4-post controlling for CD4-pre (see Table 1). This effect remained significant after number of pre- and postlaboratory reports were entered as covariates. Figure 1 shows the trend in CD4 over time for both groups. Paired t-tests revealed that MAF participants’ CD4 remained stable, Mpre = 523.48 and Mpost = 517.13, t (32) = 0.18, P = 0.86. However, the non-MAF group showed a significant decline in CD4, Mpre = 515.57 to Mpost = 449.22, t (57) = 3.18, P < 0.01.
Viral load The percentage of MAF participants and non-participants with at least one detectable vl-pre and vl-post are shown in Figure 2. MAF participants were less likely than nonparticipants to have a detectable vl-det-post (31.3% vs. 51.8%), c2 (1) = 3.48, P = 0.06. The decrease among © 2013 Wiley Publishing Asia Pty Ltd
Figure 1. Mean CD4-pre and CD4-post for medication adherence feedback (MAF) and non-MAF participants. ( ) MAF; ( ) non-MAF.
MAF participants from 54.5–31.3% was significant, c2 (1) = 3.60, P = 0.05. The relationship between MAF participation and vl-det-post was marginally significant in a logistic regression, beta = -0.97, P = 0.06. However, this effect became non-significant after entering number of pre- and postlaboratory reports as covariates, beta = -0.08, P = 0.36.
DISCUSSION The main analyses support the efficacy of a single-session feedback and counselling intervention using electronically generated adherence feedback in predicting higher CD4 counts. MAF participants were also marginally less likely
Medication adherence feedback
581
Implications for practice
Figure 2. Percentage of medication adherence feedback (MAF) and non-MAF participants with one or more vl-det-pre and vl-det-post. ( ) MAF; ( ) non-MAF.
than non-MAF participants to have a subsequent undetectable vl. The finding that self-reported missed doses underestimated the electronic report—although they were correlated—is consistent with several other studies comparing these two methods.30,31 The finding that missed doses were more common on weekends is consistent with qualitative and quantitative evidence.32,33 Future research might examine the extent to which a day-to-day change in routine is responsible for such a trend. Because this was a non-randomized study that used archival data, however, firm statements cannot be made about the effect of MAF. Although few between-group differences on any of several background variables were found, and regression analyses controlled for variables on which groups did differ, the possibility remains that there were unmeasured factors that could explain the observed trends in CD4. Optimism and conscientiousness, for example, have been shown to predict improved CD4 counts and vl in several studies.34,35 A similar argument can be made for depression.36,37 These variables might have contributed to CD4-post and vl-det-post in ways that could not be accounted for. Ideally, one would want to test the hypothesis that medication adherence mediated the relationship between the intervention and CD4-post. A significant result would demonstrate that using and discussing electronic feedback promoted adherence, which in turn boosted CD4 and suppressed vl. Unfortunately, complete information on adherence was not available for non-MAF participants, or for MAF participants other than during the 4-week intervention. Thus, mediation analysis was impossible in the present study.
These findings are, nonetheless, encouraging in that they statistically linked a behavioural feedback intervention with a medical outcome drawn from laboratory blood tests. This single-session intervention was relatively easy to implement and generated an enthusiastic response among participants—several of whom expressed a desire to continue to use the electronic pill bottle. This response suggests that the effect of the intervention on gains in HIV medical markers might have been ‘carried’ by participants’ incorporation of the feedback into their most salient personal concerns and identities. This conjecture, drawn from symbolic interaction theory, is in line with recent research on the health benefits of finding personal meaning in HIV medication adherence38,39 and promoting self-determined motivation.21,23 Insofar as participants are fully engaged as collaborators, the tangible consequences of health and wellness discussions should manifest themselves long after the conclusion of the intervention.
REFERENCES 1 Chesney M. Adherence to HAART regimens. AIDS Patient Care and STDs 2003; 17: 169–177. 2 Robbins G, Daniels B, Zheng H, Chueh H, Meigs J, Freedberg K. Predictors of antiretroviral treatment failure in an urban HIV clinic. Journal of Acquired Immune Deficiency Syndromes 2007; 44: 30–37. 3 Wood E, Hogg R, Yip B, Harrigan P, O’Shaughnessy M, Montaner J. Effect of medication adherence on survival of HIV-infected adults who start highly active antiretroviral therapy when the CD4+ cell count is 0.200 to 0.350 x 10(9) cells/L. Annals of Internal Medicine 2003; 139: 810–816. 4 Shuter J. Forgiveness of non-adherence to HIV-1 antiretroviral therapy. The Journal of Antimicrobial Chemotherapy 2008; 61: 769–773. 5 Levine A, Hinkin C, Marion S et al. Adherence to antiretroviral medications in HIV: Differences in data collected via self-report and electronic monitoring. Health Psychology 2006; 25: 329–335. 6 Chesney M, Ickovics J, Chambers D et al. Self-reported adherence to antiretroviral medications among participants in HIV clinical trials: The AACTG Adherence Instruments. AIDS Care 2000; 12: 255–266. 7 Johnson C, Heckman T, Hansen N, Kochman A, Sikkema K. Adherence to antiretroviral medication in older adults living with HIV/AIDS: A comparison of alternative models. AIDS Care 2009; 21: 541–551. 8 Hammer S, Saag M, Schechter M et al. Treatment for adult HIV infection: 2006 recommendations of the International © 2013 Wiley Publishing Asia Pty Ltd
582
9
10
11
12
13
14
15
16
17
18
19
20
21
22
WA Reich
AIDS Society-USA panel. JAMA: the Journal of the American Medical Association 2006; 296: 827–843. Freedberg K, Hirschhorn L, Schackman B et al. Costeffectiveness of an intervention to improve adherence to antiretroviral therapy in HIV-infected patients. Journal of Acquired Immune Deficiency Syndromes 2006; 43 (Suppl. 1): S113–S118. Goldie S, Paltiel A, Weinstein M et al. Projecting the costeffectiveness of adherence interventions in persons with human immunodeficiency virus infection. The American Journal of Medicine 2003; 115: 632–641. Schackman B, Finkelstein R, Neukermans C, Lewis L, Eldred L. The cost of HIV medication adherence support interventions: Results of a cross-site evaluation. AIDS Care 2005; 17: 927–937. Chaiken S. Heuristic versus systematic information processing and the use of source versus message cues in persuasion. Journal of Personality and Social Psychology 1980; 39: 752– 766. Cacioppo JT, Petty RE, Kao CF, Rodriguez R. Central and peripheral routes to persuasion: An individual difference perspective. Journal of Personality and Social Psychology 1986; 51: 1032–1043. Symonds CS, Johnson BT. The self-reference effect in memory: A meta-analysis. Psychological Bulletin 1997; 121: 371–394. Noar SM, Benac CN, Harris MS. Does tailoring matter? Meta-analytic review of tailored print health behavior change interventions. Psychological Bulletin 2007; 133: 673– 693. Epstein L, Cluss P. A behavioral medicine perspective on adherence to long-term medical regimens. Journal of Consulting and Clinical Psychology 1982; 50: 950–971. Kozuki Y, Schepp K. Visual-feedback therapy for antipsychotic medication adherence. International Clinical Psychopharmacology 2006; 21: 57–61. Otsuki M, Eakin M, Rand C et al. Adherence feedback to improve asthma outcomes among inner-city children: A randomized trial. Pediatrics 2009; 124: 1513–1521. Ryan R, Deci E. Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. The American Psychologist 2000; 55: 68– 78. Williams G, Rodin G, Ryan R, Grolnick W, Deci E. Autonomous regulation and long-term medication adherence in adult outpatients. Health Psychology 1998; 17: 269–276. Kennedy S, Goggin K, Nollen N. Adherence to HIV medications: Utility of the theory of self-determination. Cognitive Therapy and Research 2004; 28: 611–628. Davies G, Koenig L, Stratford D et al. Overview and implementation of an intervention to prevent adherence failure among HIV-infected adults initiating antiretroviral therapy:
© 2013 Wiley Publishing Asia Pty Ltd
23
24
25
26
27
28
29
30
31
32
33
34
35
Lessons learned from Project HEART. AIDS Care 2006; 18: 895–903. Koenig L, Pals S, Bush T, Pratt Palmore M, Stratford D, Ellerbrock T. Randomized controlled trial of an intervention to prevent adherence failure among HIV-infected patients initiating antiretroviral therapy. Health Psychology 2008; 27: 159–169. Kimmerling M, Wagner G, Ghosh-Dastidar B. Factors associated with accurate self-reported adherence to HIV antiretrovirals. International Journal of STD and AIDS 2003; 14: 281–284. Wagner G. Predictors of antiretroviral adherence as measured by self-report, electronic monitoring, and medication diaries. AIDS Patient Care and STDs 2002; 16: 599– 608. Smith S, Rublein J, Marcus C, Brock T, Chesney M. A medication self-management program to improve adherence to HIV therapy regimens. Patient Education and Counseling 2003; 50: 187–199. Rosen M, Dieckhaus K, McMahon T et al. Improved adherence with contingency management. AIDS Patient Care and STDs 2007; 21: 30–40. Sabin L, DeSilva M, Hamer D et al. Using electronic drug monitor feedback to improve adherence to antiretroviral therapy among HIV-positive patients in China. AIDS and Behavior 2010; 14: 580–589. Rosen M, Rigsby M, Salahi J, Ryan C, Cramer J. Electronic monitoring and counseling to improve medication adherence. Behaviour Research and Therapy 2004; 42: 409–422. Arnsten J, Demas P, Farzadegan H et al. Antiretroviral therapy adherence and viral suppression in HIV-infected drug users: Comparison of self-report and electronic monitoring. Clinical Infectious Diseases 2001; 33: 1417– 1423. Martin S, Elliott-DeSorbo D, Calabrese S et al. A comparison of adherence assessment methods utilized in the United States: Perspectives of researchers, HIV-infected children, and their caregivers. AIDS Patient Care and STDs 2009; 23: 593–601. Bachhuber M, Bilker W, Wang H, Chapman J, Gross R. Is antiretroviral therapy adherence substantially worse on weekends than weekdays? Journal of Acquired Immune Deficiency Syndromes 2010; 54: 109–110. Ryan G, Wagner G. Pill taking ‘routinization’: A critical factor to understanding episodic medication adherence. AIDS Care 2003; 15: 795–806. Ironson G, Hayward H. Do positive psychosocial factors predict disease progression in HIV-1? A review of the evidence. Psychosomatic Medicine 2008; 70: 546–554. O’Cleirigh C, Ironson G, Weiss A, Costa P. Conscientiousness predicts disease progression (CD4 number and viral load) in people living with HIV. Health Psychology 2007; 26: 473–480.
Medication adherence feedback
36 Boarts J, Sledjeski E, Bogart L, Delahanty D. The differential impact of PTSD and depression on HIV disease markers and adherence to HAART in people living with HIV. AIDS and Behavior 2006; 10: 253–261. 37 Simoni J, Frick P, Huang B. A longitudinal evaluation of a social support model of medication adherence among HIV-positive men and women on antiretroviral therapy. Health Psychology 2006; 25: 74–81.
583
38 Vervoort S, Grypdonck M, de Grauwe A, Hoepelman A, Borleffs J. Adherence to HAART: Processes explaining adherence behavior in acceptors and non-acceptors. AIDS Care 2009; 21: 431–438. 39 Westling E, Garcia K, Mann T. Discovery of meaning and adherence to medications in HIV-infected women. Journal of Health Psychology 2007; 12: 627–635.
© 2013 Wiley Publishing Asia Pty Ltd
