PEDIATRICS Vol. 119 No. 4 April 2007, pp. e893-e899 (doi:10.1542/peds.2006-1488)
ARTICLE |
Efficacy of a Pill-Swallowing Training Intervention to Improve Antiretroviral Medication Adherence in Pediatric Patients With HIV/AIDS
a Division of Behavioral Medicine
c Department of Biostatistics, St Jude Children's Research Hospital, Memphis, Tennessee
b Department of Pediatrics, University of Tennessee Health Sciences Center, Memphis, Tennessee
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
OBJECTIVE. We aimed to retrospectively assess the efficacy of pill-swallowing training provided as a clinical intervention to referred pediatric patients with HIV in relation to improved adherence and subsequent related health outcomes. The primary goal of this study was to demonstrate participation in pill-swallowing training is associated with improved medication adherence as documented by routine pharmacy pill counts. Secondary objectives were to assess corresponding improvements in clinically observed biologic indicators of adherence, specifically, immunologic functioning (CD4+ T-cell%) and viral load, over time.
PATIENTS AND METHODS. A retrospective chart review of 23 pediatric patients with HIV aged 4 to 21 years who were clinically referred for pill-swallowing training by an experienced pediatric psychologist for either noted difficulties with currently prescribed antiretroviral regimens and/or desire to change the child's regimen/formulary. Patient demographics, reason(s) for pill-swallowing training referral, number of pill-swallowing training sessions required to attain success, adherence, CD4+ T-cell%, and viral load were abstracted at baseline and at 
3 and 6 months posttraining.
RESULTS. Modal number of sessions required to acquire the pill-swallowing skill was 1 session. Younger children (aged 4–5 years) required a median of 2 training sessions, while older children required 
3 sessions. A significant improvement in adherence from baseline to 6 months post–pill-swallowing training completion was observed, as were significant related improvements in CD4+ T-cell% and viral load.
CONCLUSIONS. Participation in pill-swallowing training related to improved medication adherence at 6 months posttraining. Subsequent improvements in related CD4+ T-cell% and viral load were noted over time, most significantly at 6 months postintervention. These preliminary findings provide justification for additional study via a prospective, randomized, controlled clinical trial. Pill-swallowing training potentially is a successful time-limited, cost-effective intervention to improve adherence to antiretroviral therapies, and thus medical status, in children with HIV.
Key Words: pediatric HIV • pill swallowing • adherence • health outcomes
Abbreviations: HAART—highly active antiretroviral therapy • PST—pill-swallowing training • VL—viral load
Currently available highly active antiretroviral therapies (HAART) to treat HIV/AIDS require combination therapy with multiple daily dosing, few of which come in pediatric-friendly formulations. During routine clinic visits for medical care, parents of younger perinatally HIV-infected patients frequently report difficulty administering prescribed anti-HIV medications to their children. In addition, youth who acquired HIV behaviorally during adolescence report to practitioners difficulty with taking medications as prescribed, citing "the pills are too big," and "I'm afraid I'm gonna choke," as reasons for their own nonadherence. Because of limited available pediatric-friendly antiretroviral formulations, parents are offered alternative methods and techniques by which to administer their children's required medications (eg, opening gel caps and sprinkling contents into foods or liquids). Consequently, children frequently develop and demonstrate behavioral resistance to taking anti-HIV medications in powder or liquid formulation. Such behavioral resistance often is attributed to the medication's bad taste or unpalatable texture and/or is compounded by children's developed aversion to the foods or liquids that are used to mask the taste or conceal the medicine.
Therefore, for young patients, it is ideal to change the formulary (eg, switch from powders and/or liquids to pills or gel caps) before the child refuses to take medications without having an alternative method of medication administration or substitute formulary readily available. A goal of pill-swallowing training (PST) with young children is optimally to avoid a period of time when the child does not get his or her required medications as a result of refusing to take his or her current formulary (because of texture, taste, volume, etc) and not yet be able to swallow pills, resulting in a twice-daily difficult power struggle between parent and child around medication administration. Therefore, it is desirable to train young children to swallow pills while still compliant with liquid and/or powder formulary to encourage the acquisition of the new skill that, once acquired, will be experienced and thereby perceived by the child as an "easier" way to take medications. Despite parents'/guardians' expressed exasperation and requests for help after repeated difficulty with administering their children's medications in liquid or powder form, when offered, parents/guardians often express reticence to try PST or change their child's formulary. Reasons given for reservations related to parent/guardian perceptions of their child's inability to swallow "those big pills" or their own or their child's previous difficulty with swallowing pills without the benefit of PST, which otherwise would allow for a change in the prescribed regimen.
Children's inability to take antiretroviral medications as prescribed (nonadherence) has significant potential negative health consequences, including the potential for increased viral replication, immune suppression, and, when not taken consistently as prescribed within the acceptable dosing window, the potential to develop viral resistance to treatment, thereby limiting a young person's future treatment options for later life. Therefore, PST may provide an efficacious, time-limited, cost-effective, clinical intervention for young children with HIV to improve adherence to antiretroviral medications and thus related subsequent health benefits as reflected by improvements in CD4 T-cell% and viral load (VL) suppression while also preserving future treatment options.
Data presented herein were obtained via an institutional review board–approved, retrospective chart review study to document the efficacy of a PST technique in relation to improved adherence outcomes. PST was provided routinely as a referral-based clinical intervention at a specialty pediatric/adolescent HIV/AIDS clinic in the midsouthern United States. Others have described the PST procedure1,2 and success in acquiring the pill-swallowing skill,2–4 but no reports to date have documented the subsequent related improvements in medication adherence and secondary improvements in health outcomes in relation to provision of PST as a clinical intervention. Although the primary goal of this exploratory study was to assess improvements in adherence after participation in PST, it also was important to assess changes in commonly monitored biological indicators of antiretroviral adherence, specifically, CD4+ T-cell% and VL in relation to improved adherence observed. Therefore, study hypotheses proposed that participation in PST would relate to improved medication adherence as documented by routine pharmacy pill counts, and, secondarily, as a result of improved adherence, it was anticipated that participation in PST would relate to subsequent improved immunologic functioning (CD4+ T-cell%) and decreased HIV VL at 3 months after training with sustained improvements at 6 months after completion of PST.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Participants
The study cohort was a convenience sample that comprised all participants who were clinically referred for PST within a 2-year period and who were 6 months out from the completion of PST. Twenty-three pediatric and young adult patients with HIV/AIDS had been referred to the clinic pediatric psychologist for PST because of noted difficulties with prescribed antiretroviral regimens and/or in preparation for a desired formulary change. Patient demographics were as follows: male gender, 12 (52.2%); black race, 21 (91.3%); white race, 2 (8.7%); mean age at referral, 7.5 years (SD: 4.6; range: 4.2–21.5 years); perinatally acquired HIV, 21 (91.3%); behaviorally acquired HIV, 2 (8.7%); baseline CD4+ T-cell%, <25%, immunosuppressed, 12 (52.2%).
Procedure
Data were collected in August 2003, and the chart review period covered referrals from June 2001 through May 2003. PST followed a modified version of the shaping protocol documented by Czyzewski et al.1,2 Patients participated in individual training sessions in which the appropriate swallowing technique first was modeled by the trainer (the clinic pediatric psychologist), then practiced by the child by first swallowing pieces of gummy worm candy cut to size to emulate commensurate placebo gel cap sizes before making the transition to lactose-filled placebo gel caps, all of which were swallowed in gradually increasing sizes. Successfully swallowing 2 of each placebo size with ease was required before progressing to the next size. When any swallowing difficulty was apparent on either of the first 2 trials, a third attempt was made before returning to the previous size at which success was achieved or progressing to the next size. The number and the length of sessions were determined by the individual patient's rate of progress and the target gel cap size dictated by their prescribed or desired treatment regimens. Each child participated in as many sessions as needed to achieve success (eg, reach target pill size) or until it was determined the child was not developmentally ready to acquire the skill. In general, following the recommendation of the PST protocol, sessions did not exceed 30 minutes. The length of each session depended on the participant's level of enthusiasm or hesitation to engage in or continue with the training. For those who eagerly engaged in PST and readily grasped the pill-swallowing concept, individual sessions typically lasted 15 to 30 minutes in length. It is interesting that for the 2 young adults reviewed, sessions lasted 45 minutes and typically required considerable time to address cognitive interference and/or anticipatory anxiety issues that contributed to hesitation to initiate the actual swallowing training.
Values that were abstracted from medical charts to assess efficacy of the training for each patient included the baseline levels of adherence to their prescribed antiretroviral regimen; CD4+ T-cell% and VL before the onset of training; and repeated assessments of adherence, CD4+ T-cell%, and VL at 3 months and again at 6 months after completion of training. Adherence was measured by routine pharmacy pill count, which was reported as a percentage on the basis of the number of pills returned in the prescription vial by the patient, subtracted from the number of pills dispensed in the vial, adjusted for the number of days remaining/past in the refill period, divided by the number of pills that should have been taken in the interim period since the last refill was dispensed. Adherence to liquid medications was assessed in a similar manner; however, volume that was returned in the bottle and measured in milliliters was compared with the calculated expected volume to have been consumed by the date of the adherence check. Adherence checks were assessed routinely by a trained pharmacy technician using standardized methods described, which are appropriate regardless of medication formulary being measured. Adherence reported for each patient was derived as an average across individual adherence percentages calculated for each medication prescribed in the patient's regimen. At baseline, medications primarily were in liquid or powder form, with the exception of those who were provided pills with instructions to crush them into a powder or capsules with instructions to open and sprinkle the contents in liquid or over food (eg, in cereal, pudding, apple sauce). At the 3- and 6-month time points, medications primarily were in pill form, with the exception of those few patients who required that a study medication remain in liquid form.
Analyses
The association between age group and number of training sessions was investigated using an exact Mantel-Haenszel 
2 test. Changes from baseline in adherence percent and VL (copies per mL) were analyzed using an exact nonparametric Wilcoxon signed-rank test. CD4+ T-cell% values were categorized into 0% to 14%, 15% to 24%, and >25% corresponding to Centers for Disease Control and Prevention–defined clinical cutoffs for classifying immune suppression5 and analyzed using an exact marginal homogeneity 2 test for paired categorical data. Spearman's rank correlation coefficient was used to investigate the relationship between various continuous measurements. The effects of regimen type, pill burden (number of pills per day), and liquid medication volume adjusted for assessment time point were explored in individual repeated measures mixed models.6 These models included a main effect for treatment variable and assessment time point and accounted for the correlation of repeated assessments over time for a given participant. Exact tests were performed using StatXact-5 for Windows.7 For this exploratory, retrospective, study, P < .10 was considered significant.8
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Patients were referred for PST primarily because of nonadherence (56.5%) or desired formulary change (43.5%) as a result of difficulty with the prescribed regimen or development of viral resistance. Twenty-two (95.7%) referred patients were successful with training (Table 1). One child, who was 5 years of age and presented with significant developmental delay, was not able to grasp the pill-swallowing concept. Because of behavioral refusal, training was discontinued during the first attempted session. As per recommendation in the PST protocol used,1 had this been a prospective trial, PST is not recommended for a child with developmental delay, and, therefore, a child with developmental delay would have been excluded from participation. However, this child also had a gastric tube in place, so this child did not take any medications by mouth (as all other participants did), and supportive community-based health services were placed to ensure medication administration and adherence. As a result of these factors, this child's data were not included in analyses.
|
The modal number of PST sessions that were required to obtain success was 1. The number of sessions that were required to complete training increased with age (Spearman's r = 0.66, P = .001). Younger children (age 4–5 years) required fewer training sessions (Table 1) to reach their target pill size and make the transition to their medication regimen than older children (40% of 4- to 5-year-olds required only 1 session; 50.0% of participants aged
8 years required 4 sessions; P = .038; Table 1). Participation in training resulted in a complete formulary change (eg, powder/liquid, opening and sprinkling gel cap contents to swallowing gel caps whole) for 18 (81.8%) patients and a regimen change for 10 (45.5%) patients; most remained on 3 antiviral medications.
Adherence was significantly improved at 6 months (T3) from baseline (T1), increasing by a median of 9.8% (P = .014; Table 2). An increase in adherence was noted at 3 months, when 41.2% (7 of 17) had improved from baseline, and by 6 months, when 71.4% (10 of 14) had improved (Fig 1). It should be noted that adherence was not fully assessable because patients/parents presented for their scheduled clinic visit without returning medication bottles to the pharmacy. A significant improvement in CD4+ T-cell% emerged from baseline to 3 months (T2; P = .063) and from baseline to 6 months (P = .004; Table 3). The lack of significant difference between T2 and T3 suggests that initial changes in CD4+ T-cell% were sustained over time. The percentage of patients who showed no evidence of immune suppression (25% CD4+ T-cells) increased from 50% (11 of 22) at baseline to 81.8% (18 of 22) at 6 months. There was a corresponding decrease in VL by 6 months, with a median decrease in VL of 1258 copies per mL (P = .093; Table 4). Although trends in improvement over time were noted in CD4+ T-cell% and VL in response to formulary change alone (n = 12), these differences were nonsignificant (data not shown). However, there was a noted improvement from T1 to T3 in CD4+ T-cell% for those who changed their medication regimen after PST (n = 10; exact marginal homogeneity 2 test, P = .031). For patients who had only a formulary change, 83.3% showed no evidence of immunosuppression at T3 compared with 66.7% at T1. For patients who had a medication regimen change, 80% showed no evidence of immunosuppression at T3 compared with 30% at T1. Given the degree of immunosuppression for this group at T1, clearly a regimen change was indicated after PST, and clinical response was evidenced at T3. However, for those who had only a formulary change, a greater percentage was not immunosuppressed at T1, indicating that a regimen change was not clinically warranted for this group.
|
|
|
|
No significant relationships were found between adherence and regimen type or pill burden (number of pills per day). Surprising, however, for those who were prescribed liquid medication, increased liquid volume was related significantly to better adherence (P = .043), possibly because many who retained at least 1 medication in a liquid formulation did so as required by clinical trial research participation. Therefore, those who were reviewed for this study and remained on liquid medication may have been more compliant as a result of their commitment to participate in research in addition to the close follow-up related to study visits and monitoring of study medications. Given the study's small sample size, the distributional assumption of normality for the repeated measures mixed model was difficult to evaluate. Therefore, it should be noted that nonparametric analyses indicated a significant positive relationship between liquid volume and adherence at baseline (Spearman's r = 0.64, P = .004); however, no significant correlations were detected at
3 months and 6 months. It is important to note that no patients who participated in our PST demonstrated a decline in health status during the study period. |
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
All but 1 participant (with significant developmental delay) demonstrated success in acquiring the pill-swallowing skill, resulting in desired formulary change and/or regimen change because of having failed a previous regimen or the development of viral resistance. Younger children acquired the skill more readily, most likely because they were naïve to pill swallowing and therefore were not influenced by previous negative experiences. In addition, younger children typically were of an age at which it is expected they had not yet developed size conservation whereby increasing gel cap size did not create a reticence to attempt each new gel cap size as it was presented during the training. Noted success with PST provides support to indicate that children as young as 4 years1,2 can readily acquire the ability to swallow pills regardless of target size required for their regimen (for some of whom was #00 placebo or the size equivalent of a 1000 IU vitamin E gel cap), contrary to initial caregiver concerns or expectations.
Findings of this retrospective study support the proposed hypotheses. Specifically, success with PST as a clinical intervention related to improved medication adherence as documented by routine pill counts. Secondary to improved adherence, participants demonstrated related improved immunologic functioning (CD4+ T-cell%) and decreased VL after the completion of PST, and these improvements emerged at 3 months and were evident at 6 months after completion of training. In addition, although participation in PST allowed for a formulary change for all participants for whom a formulary change was available, a more noteworthy effect of participation was demonstrated for those who also were able to change their treatment regimen as a result of participation in PST. Those who experienced only a formulary change showed a trend toward improved VL and immunologic functioning. However, for those who also experienced a regimen change subsequent to PST, a greater, although still nonsignificant, trend was noted in improved VL, whereas a significant improvement in immune status over time was demonstrated by a significant proportion of those who showed no evidence of immunosuppression at T3 in comparison with T1. Although the relationship here would be speculative because of the retrospective nature of this study, successful participation in PST seems to provide a clinical intervention that may permit a regimen change to a potentially more therapeutic regimen, which otherwise previously was not an option for these participants because the more potent regimen is available only in pill form and cannot be opened, crushed, squeezed, or chewed.
Although this was an exploratory retrospective study, findings suggest that PST may be a viable, time-limited (1–4 sessions), cost-effective intervention (psychologist's time versus monthly cost of medications, repeated laboratory work to monitor CD4+ T-cell% and VL, and patients' health risk to develop viral resistance associated with nonadherence) to improve adherence to HAART and, secondarily, related to improvements in immunologic and viral status in young children with perinatally acquired HIV. Providing training before a desired formulary change seems to aid in the successful transition from liquid/powder formulations and prepares the child for success with pills, resulting in subsequent improved adherence.
The utility of this training has implications for standard of care in pediatric HIV, especially for children as young as 4 years of age, who readily seem to acquire the skill with successful transfer of training to their prescribed medications. PST has an implication for reducing the cost of HIV patient care through reduced nonadherence (eg, cost of dispensed medications otherwise not taken, thrown away, or having to be redosed because of gagging/vomiting; increased cost because of more frequent medical care visits and monitoring, not to mention additional laboratory work to monitor response via CD4+ T-cell%, VL, and genotyping as a result of viral resistance). These findings provide support for additional investigation via a prospective study of PST as a therapeutic intervention to be provided as a standard of care to young children before size conservation or behavioral refusal that is observed in older children becomes an issue.
Limitations of this study include those related to conducting a retrospective chart review in which prospective data are not gathered uniformly, at predetermined time points, or specifically for each patient included, whereby timing and completeness of data are inexact. Despite that participants typically were scheduled for routine medical care approximately every 3 months, baseline adherence, VL, and CD4+ T-cell% data were not obtained routinely on the date that PST began, and the timing for T2 and T3 data was not exactly at 3 and 6 months after completion of training for desired adherence checks and laboratory values. For the 18 participants with baseline adherence data available for analysis, 72% of assessments were within 30 days of the start of PST; for VL and CD4+ T-cell%, which were assessed in all 22 participants, 73% and 68%, respectively, were obtained within 30 days of the start of PST. Values that subsequently were closest to 3 months (adherence median: 85.0 days; CD4+ T-cell% median: 91.0 days; VL median: 95.5 days) and 6 months (adherence median: 181.0 days; CD4+ T-cell% median: 192.5 days; VL median: 185.5 days) were used. As previously noted, a number of adherence observations were missing, which may be a potential source of bias if those with missing data differed in adherence compared with those whose data were fully assessable. On the basis of clinical experience, those who adhered well were equally as likely not to return medication vials for adherence checks as those who were nonadherent; therefore, no directional bias is expected for these missing data, but this cannot be determined objectively here.
The small sample size should be kept in mind when interpreting these results. The absence of significant findings may be attributable to the limited power to detect important differences. For example, although not significant, the trends for pill burden were in the expected direction. Although corresponding improvements in VL and CD4+ T-cell% were detected, other confounding factors and temporal variations could be at play. PST was provided as a routine clinical intervention on the basis of intention to treat to all eligible patients who demonstrated developmental readiness (as young as 3.5 years). Although it would have been ideal for research purposes, no comparison control group was available for review because all children of similar age in this clinic subsequently had been provided PST and thus transitioned to pill formulary. Therefore, a larger randomized, controlled study is needed to demonstrate clearly the efficacy of pill-swallowing on adherence and to strengthen the link to clinical outcomes (VL and CD4+ T-cell%).
Aside from other unavoidable logistic constraints related to retrospective chart review data, consistency in PST was obtained because all patients were seen for the clinical intervention by the same pediatric psychologist. It also should be noted that patients with known adherence difficulties receive adherence counseling by their primary care practitioner, pharmacist, and, in many cases, also their social worker as routine standard of care before and in addition to referral to the pediatric psychologist for PST. As a result, it is not possible to assess or account for the role that these supportive services may contribute to the findings.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Although this preliminary retrospective study is the first to relate the efficacy of PST to improved adherence and subsequent related improvements in health outcomes (CD4+ T-cell% and VL), a larger prospective longitudinal study is needed to strengthen the evidence for PST as a cost-effective brief intervention to improve adherence in young patients who have HIV and are prescribed HAART regimens over both the short and long term. One benefit of such a randomized, clinical trial would be the inclusion of a comparison control group by which the treatment effect of PST could be assessed definitively. Although this particular study explored the efficacy of PST for children with HIV, the findings demonstrate that children who have a chronic health condition and are as young as 4 years readily acquire the pill-swallowing skill. The related improved adherence after participation in PST reduces the risk that viral resistance will develop, and as a result of acquiring the pill-swallowing skill, increased regimen options become available to young children. Therefore, with improved and sustained adherence, those regimens remain available as future treatment options for later life as the child ages up. As a result, the noted subsequent improvements in adherence after participation in PST have much farther reaching implications for improved health outcomes not only for children with HIV but also for those with other pediatric conditions for which medication adherence, whether over the short or long term, is critical for symptom management or infectious disease control.
|
ACKNOWLEDGMENTS |
|---|
This research was supported by the American Lebanese-Syrian Associated Charities.
We thank Julie Richardson, PharmD, Susan Carr, PharmD, William T. Dalton, III, PhD, Wally Bitar, MS, and Xin Deng, MS, for respective contributions to data abstraction, verification, and analyses. In addition, we thank Danita Czyzewski, PhD, for sharing the pill-swallowing protocol materials for clinical use and ongoing support of our work with this special pediatric population.
|
FOOTNOTES |
|---|
Accepted Oct 18, 2006.
Address correspondence to Patricia A. Garvie, PhD, Behavioral Medicine Division, St Jude Children's Research Hospital, 332 N Lauderdale St, Mail Stop 740, Memphis, TN 38105. E-mail: patti.garvie{at}stjude.org
The authors have indicated they have no financial relationships relevant to this article to disclose.
This study was presented in part at the annual meeting of the American Psychological Association; July 30, 2004; Honolulu, HI.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
- Czyzewski DI, Runyan RD, Brown G. HIV Therapy in Children. A Primer on Pill Swallowing. Manual and Pill Swallowing Training Video. Basel, Switzerland: F. Hoffman-LaRoche Ltd; 1997
- Czyzewski DI, Runyan RD, Lopez MA, Calles NR. Teaching and maintaining pill swallowing in HIV-infected Children. AIDS Reader. 2000;10 :88 –94
- Blount RL, Dahlquist LM, Baer RA, Wuori D. A brief, effective method for teaching children to swallow pills. Behav Ther. 1984;15 :381 –387[CrossRef][Web of Science]
- Dahlquist LM, Blount RL. Teaching a six-year-old girl to swallow pills. J Behav Ther Exp Psychiatry. 1984;15 :171 –173[CrossRef][Web of Science][Medline]
- Centers for Disease Control and Prevention. Revised classification system for human immunodeficiency virus infection in children less than 13 years of age. MMWR Recomm Rep. 1994;43(RR-12) :1 –10
- Diggle PJ, Liang KY, Zeger SL. Analysis of Longitudinal Data. New York, NY: Oxford University Press; 1994
- StatXact-5 for Windows Statistical Software for Exact Nonparametric Inference [computer program]. Cambridge, MA: Cytel Software Corp; 2001
- Piantadosi S. Clinical Trials: A Methodologic Perspective. New York, NY: Wiley; 1997:162
PEDIATRICS (ISSN 1098-4275). ©2007 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
