Abstract
Background: Long-acting inhaled medications are an important component of the treatment of patients with chronic obstructive pulmonary disease (COPD), yet few studies have examined the determinants of medication adherence among this patient population. Objective: We sought to identify factors associated with adherence to long-acting beta-agonists (LABA) and inhaled corticosteroids (ICS) among patients with COPD. Methods: We performed secondary analysis of baseline data collected in a randomized trial of 376 Veterans with spirometrically confirmed COPD. We used electronic pharmacy records to assess adherence, defined as a medication possession ratio of ≥0.80. We investigated the following exposures: patient characteristics, disease severity, medication regimen complexity, health behaviors, confidence in self-management, and perceptions of provider skill. We performed multivariable logistic regression, clustered by provider, to estimate associations. Results: Of the 167 patients prescribed LABA, 54% (n = 90) were adherent to therapy while only 40% (n = 74) of 184 the patients prescribed ICS were adherent. Higher adherence to LABA and ICS was associated with patient perception of their provider as being an “expert” in diagnosing and managing lung disease [For LABA: OR = 21.70 (95% CI 6.79, 69.37); For ICS OR = 7.93 (95% CI 1.71, 36.67)]. Factors associated with adherence to LABA, but not ICS, included: age, education, race, COPD severity, smoking status, and confidence in self-management. Conclusions: Adherence to long-acting inhaled medications among patients with COPD is poor, and determinants of adherence likely differ by medication class. Patient perception of clinician expertise in lung disease was the factor most highly associated with adherence to long-acting therapies.
Introduction
A number of randomized trials have shown long-acting inhaled medications to lessen symptoms, decrease exacerbations, and improve health-related quality of life among patients with chronic obstructive pulmonary disease (COPD) (Citation1, 2). At the same time, non-adherence to long-acting inhaled therapies has been associated with increased hospitalizations and health-care expenses (Citation3). Together, these studies suggest that adherence to prescribed long-acting pharmacologic therapies is important to improving health outcomes among patients with COPD. However, little is known about the determinants of adherence to long-acting inhaled medications among this patient population, as most previous studies of adherence in COPD included short-acting therapies alone (Citation4–7).
Adherence to a prescribed therapeutic regimen is likely a complex process that incorporates a number of potential factors, including patient characteristics (e.g., age, health beliefs and behaviors, co-morbid illness, self-efficacy), societal interactions, patient's perception and trust, (e.g., the clinician-patient relationship) and the complexity and effectiveness of current medications (Citation8, 9). Although adherence to long-acting inhaled therapies have been extensively investigated among patients with asthma (Citation10–15), extrapolating inferences from this literature for patients with COPD is likely problematic, as many of these studies focused on children or young adults (Citation12–14) and inner-city (Citation15) or low income (Citation14) populations. In contrast, patients with COPD are typically at least middle-aged with multiple co-morbid illnesses. In addition, with the exception of patients with severe persistent asthma, the potential of inhaled therapies to restore normal health status is significantly better among patients with asthma than those with COPD (Citation16).
Understanding of the determinants of adherence to long-acting therapies with proven efficacy for patients with COPD is necessary to help inform interventions to improve adherence to inhaled medications. We sought to identify factors associated with adherence to long-acting beta-agonists (LABA) and inhaled corticosteroids (ICS) among patients with COPD.
Methods
Study Design and Subjects: e performed secondary analysis of baseline data collected in a clinician-clustered, randomized trial to improve the quality of end-of-life communication between patients with COPD and their providers (Citation17). Clinicians (n = 92) and patients (n = 376) were recruited between January 2004 and November 2007, from primary care, pulmonary and geriatric clinics at the Department of Veterans Affairs Puget Sound Health Care System (VAPSHCS). Clinician participants included advanced practice nurses, physicians, and a physician assistant. To be included in the study, patients were required to have spirometric confirmation of obstructive lung disease (Citation18) and to identify a participating provider as the individual responsible for their COPD care. The Institutional Review Boards at the University of Washington and VAPSHCS approved all study protocols and informed consent was obtained from all subjects prior to data collection (VA MIRB # 01378).
Data Collection: e determined medication use by extracting pharmacy data from the Veterans Integrated Service Network (VISN)-20 data warehouse, which contains detailed prescription information for all medications dispensed within the VISN. In the year prior to enrollment, we collected detailed prescription information on inhaled therapies (short-acting beta agonists, short-acting anti-cholinergics, LABA and ICS) oral hypoglycemic agents (metformin or a sulfonylurea), HMG-CoA reductase inhibitors, and six classes of anti-hypertensive medications (beta-blockers, ace-inhibitors, angiotensin II receptor blockers, calcium channel blockers, alpha-1 antagonists and thiazide diuretics). We defined complexity of medication regimen by the number of total inhalers prescribed (with values ranging from 1–4) and the number of additional oral agents prescribed (with values ranging from 0–9).
We used self-report to assess demographic characteristics, tobacco use, and the presence of 26 co-morbid illnesses included in the Seattle Index of Co-morbidity (SIC) score (Citation19). We assessed alcohol use by the AUDIT-C questionnaire (Citation20). Patients were asked to rate their confidence in their ability to self-manage their breathing problems (self-efficacy) as well as their confidence in their provider's skills and knowledge about lung disease. The wording of these questions and possible responses can be found in the online supplement (Table e1). All patients underwent spirometry and the post-bronchodilator percent predicted FEV1 was used to determine disease severity (Citation18).
Outcomes: e determined adherence to LABA and ICS in the 6 month period prior to enrollment using a validated refill compliance measure known as ReComp (Citation21). Over longer periods of time, ReComp is equivalent to a medication possession ratio and represents the actual proportion of days that medications were available (Citation21). We anticipated that determinants of adherence may vary by medication class, and therefore examined LABA and ICS separately, with subjects on both medications contributing to each analysis. Because of formulary decisions, combination therapies of ICS and LABA were not widely available through our VA pharmacy. Likewise, we did not examine the long-acting muscarinic antagonist tiotropium due to its limited use within the VA system during the time of the study. Based upon the previous medication adherence literature (Citation22) and our own prior work (Citation23, 24), we defined adherence to an individual medication as a ReComp score ≥ 0.80.
Exposures of Interest: e identified four major groups of patient factors that may be associated with adherence: 1) patient characteristics (including demographic factors, co-morbid illness and COPD severity) and complexity of medication regimen; 2) patient health behaviors (including tobacco and alcohol use); 3) patient self-efficacy; and, 4) patient perception of provider skill and knowledge.
Statistical Analysis: e performed all statistical analyses using Stata SE-10.0 (StataCorp, College Station, TX). We compared non-adherent and adherent subjects using chi-squared tests and Student's t-tests as appropriate. We used multivariable logistic regression, with robust estimates of variance clustered by clinician, to examine factors associated with adherence (Citation25, 26). We constructed four separate base models for each medication class.
The first model (Model 1) included demographic and socioeconomic characteristics, the co-morbid illness score (SIC), COPD severity (post-bronchodilator FEV1% predicted), and complexity of inhaled and oral medication regimens. The remaining three models included the specific groups of predictors described above. We decided a priori to adjust these base models for all of the factors included in Model 1. There were no differences in significant predictors between the base and adjusted models, and therefore only the adjusted models are presented. We considered a two-tailed p-value of < 0.05 to be significant.
Sensitivity Analyses: e assessed for effect modification of the association of patient perception of provider expertise and adherence by clinician degree (MD or advanced practice nurses) or clinician type (primary care or pulmonary provider). We also examined whether this relationship was “disease-specific” by examining the association of patient perceived pulmonary expertise and adherence to HMG-CoA reductase inhibitors using the same methodology described here.
Results
Characteristics of Non-adherent and Adherent Subjects to LABA and ICS: f the 376 patients who participated in the study, 243 (65%) of patients were prescribed at least one long-acting inhaled medication in the 6 months prior to study entry; 167 (44%) were prescribed a LABA and 184 (49%) were prescribed an ICS, with 108 (29%) prescribed both medications. We found 54% (n = 90) of patients prescribed LABA were adherent, compared to 40% (n = 74) of subjects prescribed ICS.
Patients who were adherent to LABA less frequently reported alcohol misuse or current tobacco use and reported higher confidence in self-management of their breathing problems (). These differences were not seen among patients prescribed ICS. Subjects who were adherent to both LABA and/or ICS described their providers as being more skilled and knowledgeable than non-adherent subjects.
Table 1. Patient characteristics based upon adherence status
Demographic Factors, Co-morbid Illness, COPD Severity, and Complexity of Regimen: very 10% increase in FEV1% predicted was associated with 14% lower odds of adherence to LABA [OR 0.86 (95% CI 0.74, 1.00), p-value = 0.04], but not associated with adherence to ICS. Among patients prescribed LABA, but not those prescribed ICS, older age and higher levels of education were also associated with higher adherence, while being of Black race was associated with lower adherence (). We found no association with adherence to either medication by marital status, income, or co-morbid illness as measured by the SIC score.
Table 2. Associations of patient characteristics and health behaviors with adherence status
Health Behaviors: urrent smokers had a 68% lower odds of adherence to LABA [OR 0.32 (95% CI 0.11, 0.97)], but no difference in adherence to ICS (). Alcohol use was not associated with adherence to either LABA or ICS.
Patient self-efficacy: igher confidence in self-management of their lung disease was associated with adherence to LABA (p-value for test of trend = 0.003). Subjects confident with self-management of their lung disease “a good bit” to “most of the time” had more than a 3-fold higher odds of adherence to LABA [OR 3.57 (95% CI 1.09, 11.71)], and subjects confident “all of the time” demonstrated an almost 6-fold higher odds of adherence to LABA [OR 5.92 (95% CI 1.86, 18.85)] than those who were confident “never” to “some of the time”. This association was not found among patients prescribed ICS ().
Table 3. Associations of patient perceptions with adherence status
Perceptions of Provider Skill and Knowledge: mong subjects prescribed both LABA and ICS, the odds of adherence was higher with increased patient confidence in provider skill in diagnosing and treating lung disease (p-value for trend = < 0.001 for LABA and 0.007 for ICS) (). Among patients prescribed LABA and compared to patients who rated their providers to be “not at all” to “somewhat knowledgeable” in treating lung disease, patients who reported their provider as “very knowledgeable” and those who reported their providers as “expert” had markedly increased odds of adherence [for very knowledgeable: OR 15.28 (95% CI 4.25, 54.97); for expert: OR 21.70 (95% CI 6.79, 69.37)]. However, among patients prescribed ICS, only subjects who described their provider as lung disease “experts” had significantly higher odds of adherence [OR 7.93 (95% CI 1.71, 36.67)].
Sensitivity Analyses-Provider type: e found no effect modification of the association of perceived clinician pulmonary expertise and adherence by provider degree (physician or non-physician) (For LABA, interaction term p-value = 0.276; For ICS, interaction term p-value = 0.201) or provider type (pulmonary or primary care provider) (LABA interaction term p-value = 0.970; ICS interaction term p-value = 0.173). This suggests the patient perception of provider expertise was not limited to physicians or subspecialty providers.
Adherence to non-COPD medication: verall adherence to statins was higher than to the inhaled medications studied, with 68% of the 189 patients prescribed the medication adherent based on a ReComp score ≥ 0.80. We found no association between patient-perceived expertise in lung disease and statin adherence (p-value for trend = 0.794), suggesting a condition specific effect.
Discussion
Among patients with COPD, we found that although adherence to LABA is better than to ICS, adherence to either long-acting medication is poor. Moreover, we found that while there are relatively fewer determinants of adherence to ICS than to LABA, the confidence a patient has in their provider is a common strong predictor of adherence to both medications. This was true regardless of clinician type or specialty training. Our findings are important as they highlight the magnitude of non-adherence to medications with proven efficacy to improve symptoms and prevent exacerbations as well as have the potential to guide the development of interventions to improve adherence to these medications.
Most of the previous studies of medication adherence among patients with COPD were performed prior to the common use of long-acting inhalers, and included only short-acting medications (Citation4–6). Several recent studies examining adherence to inhaled medications have either been restricted to only one medication (Citation27), pooled all therapies together (Citation28), or limited their scope to describing only the patterns of use without examining predictors of adherence for long-acting therapies (Citation29). To our knowledge, this is one of only two studies to examine determinants of adherence to both LABA and ICS among patients with COPD alone (Citation3), and the first to examine factors such as self-efficacy and patient perceptions of providers.
Unlike our study, a secondary analysis of the Towards a Revolution in COPD Health (TORCH) trial found adherence rates to be relatively high (approximately 80%) and did not find a difference in participants’ adherence status based upon age, disease severity, or smoking status (Citation3). However, the TORCH study included patients started on these medications as part of a clinical trial, a factor known to overestimate true adherence in real world clinical settings for several reasons, including patient knowledge that their medication use is likely being monitored (Citation30, 31). Although the patients in our study were also participants in a clinical trial, we assessed medication adherence prior to their enrollment in the study. In addition, the outcome and intervention were unrelated to medication use, and our findings may therefore be more reflective of clinical practice.
Due to formulary restrictions within the VA healthcare system, almost no patients are treated with a combination LABA/ICS inhaler, allowing us the opportunity to assess differences in adherence between the two classes of medication. We found participants prescribed LABA were more likely to be adherent than those prescribed ICS. This difference may be related to the difference in the rapidity of symptomatic effects of the two medications. Almost all patients prescribed a LABA within the VA system are prescribed formoterol, a single dose dry-powdered inhaler with a rapid onset of action. In contrast, inhaled corticosteroids have no immediate effect on symptoms, which may negatively affect adherence among users of the medication.
We also found differences in the determinants of adherence to the two medications. Higher confidence in self-management of breathing problems was associated with adherence to LABA. This could reflect the fact that those with improved LABA adherence attained better symptom control, resulting in increased self-efficacy. Older subjects, those with more education, and those who had achieved smoking cessation were also more likely to adhere to LABA.
Together, these characteristics could also be markers of a patient's enhanced understanding of their lung disease, a factor which has been shown to affect adherence to therapies among patients with COPD (Citation28) and asthma (Citation32). We did not find an association between these same characteristics and adherence to ICS. One explanation for this difference may be the controversy that surrounds the use of ICS for treatment of COPD (Citation33). Patients with a better understanding of their lung disease who do not feel they are deriving symptomatic benefit from the inhaler may stop the medication due to inconvenience or concern over potential side effects.
Perhaps most importantly, only one factor was associated with adherence to both LABA and ICS: confidence in provider expertise. This association was true for both physicians and non-physicians and was not limited to subspecialty providers. There are several different explanations for this finding. First, patients with higher adherence to long-acting inhaled medications may experience decreased symptoms, leading them to perceive their prescribing provider as being more expert in caring for their lung disease. Second, confidence in provider expertise may be a surrogate marker for an improved “clinician-patient” relationship.
This is consistent with a prior study which showed adherent patients with COPD were more likely to describe their physician as being “supportive” (Citation7). Third, patient perception of provider expertise may reflect increased provider self-efficacy in managing lung disease. Clinicians who are more skilled and confident in their ability to treat COPD may also appear to be more knowledgeable to their patients, a perception that may affect both a patient's belief of the effectiveness of prescribed treatments as well as their willingness to adhere to these therapies. This finding is supported by a prior study which suggested patients with self-reported “high-adherence” had higher ratings of satisfaction and “faith” in their providers (Citation28).
This study had important strengths. First, our measure of refill compliance was based on pharmacy data, which has been shown to be more accurate than patient self-report and nearly comparable to electronic monitoring (Citation34). Second, we were able to obtain information regarding patient perceptions of both the management of their lung disease and satisfaction directly to their providers who cared for their COPD. While these factors are known to be important predictors of adherence in other chronic diseases, they have not been previously well-studied among patients with COPD. Third, our study evaluated potentially modifiable factors associated with adherence in order to identify novel mechanisms for improving adherence. Finally, our study was performed in an integrated health care system that allowed for complete ascertainment of outcome without biases from differential access because of insurance status.
This study has potential limitations. First, despite its strengths, pharmacy refill data does not necessarily indicate that the medication was taken by the patient. Second, we were unable to assess adherence to LABA/ICS combination therapy or tiotropium as the medications were uncommonly used within the VA system at the time of this study. Third, subjects may have obtained additional medications at non-VA pharmacies, artificially decreasing their adherence.
However, we feel this is unlikely since LABA and ICS medications are available within the VA for a nominal ($8) or no copayment based on eligibility status, resulting in a substantial cost incentive for Veterans to purchase these medications through the VA. Fourth, our measure of medication regimen complexity did not incorporate all medications or dosing patterns thereby possibly limiting our estimate of overall medication complexity. Last, our study was conducted within the VA, and the large majority of our population is composed of White men, which may make our findings less generalizeable to other populations, such as those which are younger, more affluent, or contain a higher proportion of minorities.
Conclusions
In summary, overall adherence to long-acting inhaled therapies for COPD is poor, and in clinical practice may be even lower than previously estimated. Factors associated with adherence to long-acting medications for treatment of COPD likely differ by medication class suggesting that clinicians should inquire separately about the use of inhaled therapies and that potential interventions may need to be medication-specific. An important potentially modifiable factor highly-associated with adherence to both LABA and ICS was patient confidence in provider expertise in lung disease. The findings of this study could help inform interventions designed to improve adherence to long-acting inhaled medications with an overall goal of reducing adverse health outcomes, such as hospitalizations and health-care costs, among patients with COPD.
Declaration of Interest Statement
This material is based upon work supported by a VA Advanced Fellowship Program in Health Services Research and Development [TPM 61–037] and a Department of Veterans Affairs, Veterans Health Administration, Office of Research and Development grant [IIR 02–292]. Dr. Slatore is supported by a Veterans Affairs HSR&D Career Development Award and resources from the Portland VA Medical Center, Portland, OR. Additional support also comes from the Firland Foundation. The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs. Some of the results of this work were presented in abstract form at the American Thoracic Society International Conference, in Denver, Colorado on May 15, 2011.
Dr. Cecere, Dr. Slatore, Ms. Uman, Dr. Evans and Mr. Udris have no conflicts of interest to disclose. Dr. Bryson and Dr. Au are Co-Investigators on a grant from Giliad Sciences for work unrelated to this manuscript. Dr. Au also sits on the medical editorial board for Nexcura and is a research consultant for Bosch LLC.
References
- Sin D, McAlister F, Man S, Anthonisen N. Contemporary management of chronic obstructive pulmonary disease: scientific review. JAMA Nov 2003; 290(17):2301–2312.
- Wilt TJ, Niewoehner D, MacDonald R, Kane RL. Management of stable chronic obstructive pulmonary disease: a systematic review for a clinical practice guideline. Ann Intern Med Nov 2007; 147(9):639–653.
- Vestbo J, Anderson J, Calverley P, Adherence to inhaled therapy, mortality and hospital admission in COPD. Thorax Nov 2009; 64(11):939–943.
- Dolce J, Crisp C, Manzella B, Richards J, Hardin J, Bailey W. Medication adherence patterns in chronic obstructive pulmonary disease. Chest Apr 1991; 99(4):837–841.
- Rand C, Nides M, Cowles M, Wise R, Connett J. Long-term metered-dose inhaler adherence in a clinical trial. The Lung Health Study Research Group. Am J Respir Crit Care Med Aug 1995; 152(2):580–588.
- Corden Z, Bosley C, Rees P, Cochrane G. Home nebulized therapy for patients with COPD: patient compliance with treatment and its relation to quality of life. Chest Nov 1997; 112(5):1278–1282.
- Turner J, Wright E, Mendella L, Anthonisen N. Predictors of patient adherence to long-term home nebulizer therapy for COPD. The IPPB Study Group. Intermittent Positive Pressure Breathing. Chest Aug 1995; 108(2):394–400.
- Bourbeau J, Bartlett S. Patient adherence in COPD. Thorax Sep 2008; 63(9):831–838.
- Rand C. Patient adherence with COPD therapy. Eur Respir Rev 2005; 14(96):97–101.
- Bender BG, Bender SE. Patient-identified barriers to asthma treatment adherence: responses to interviews, focus groups, and questionnaires. Immunol Allergy Clin North Am Feb 2005; 25(1):107–130.
- Williams LK, Joseph CL, Peterson EL, Patients with asthma who do not fill their inhaled corticosteroids: a study of primary nonadherence. J Allergy Clin Immunol Nov 2007; 120(5):1153–1159.
- Smith LA, Bokhour B, Hohman KH, Modifiable risk factors for suboptimal control and controller medication underuse among children with asthma. Pediatrics Oct 2008; 122(4):760–769.
- Conn KM, Halterman JS, Lynch K, Cabana MD. The impact of parents’ medication beliefs on asthma management. Pediatrics Sep 2007; 120(3):e521–526.
- Finkelstein JA, Lozano P, Farber HJ, Miroshnik I, Lieu TA. Underuse of controller medications among Medicaid-insured children with asthma. Arch Pediatr Adolesc Med Jun 2002; 156(6):562–567.
- Ponieman D, Wisnivesky JP, Leventhal H, Musumeci-Szabó TJ, Halm EA. Impact of positive and negative beliefs about inhaled corticosteroids on adherence in inner-city asthmatic patients. Ann Allergy Asthma Immunol Jul 2009; 103(1):38–42.
- Calverley PM. Effect of corticosteroids on exacerbations of asthma and chronic obstructive pulmonary disease. Proc Am Thorac Soc 2004; 1(3):161–166.
- Au DH, Udris EM, Engelberg RA, A randomized trial to ipmrove communication about end-of-life care among patients with COPD. Chest 2011 Sep 22. [Epub ahead of print]
- Rabe K, Hurd S, Anzueto A, Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: GOLD executive summary. Am J Respir Crit Care Med Sep 2007; 176(6):532–555.
- Fan VS, Au D, Heagerty P, Deyo RA, McDonell MB, Fihn SD. Validation of case-mix measures derived from self-reports of diagnoses and health. J Clin Epidemiol Apr 2002; 55(4):371–380.
- Bush K, Kivlahan DR, McDonell MB, Fihn SD, Bradley KA. The AUDIT alcohol consumption questions (AUDIT-C): an effective brief screening test for problem drinking. Ambulatory Care Quality Improvement Project (ACQUIP). Alcohol Use Disorders Identification Test. Arch Intern Med Sep 1998; 158(16):1789–1795.
- Bryson C, Au D, Young B, McDonell M, Fihn S. A refill adherence algorithm for multiple short intervals to estimate refill compliance (ReComp). Med Care Jun 2007; 45(6):497–504.
- Andrade SE, Kahler KH, Frech F, Chan KA. Methods for evaluation of medication adherence and persistence using automated databases. Pharmacoepidemiol Drug Saf Aug 2006; 15(8):565–574; discussion 575–567.
- Slatore C, Bryson C, Au D. The association of inhaled corticosteroid use with serum glucose concentration in a large cohort. Am J Med May 2009; 122(5):472–478.
- Bryson CL, Au DH, Sun H, Williams EC, Kivlahan DR, Bradley KA. Alcohol screening scores and medication nonadherence. Ann Intern Med Dec 2008; 149(11):795–804.
- Vittinghoff E, Glidden DV, Shiboski SC, McCulloch CE. Regression methods in biostatistics: linear, logistic, survival, and repeated measures models. New York: Springer; 2005.
- Huber P. Robust Statistics. New York: John Wiley & Sons; 1981.
- van Grunsven PM, van Schayck CP, van Deuveren M, van Herwaarden C, Akkermans RP, van Weel C. Compliance during long-term treatment with fluticasone propionate in subjects with early signs of asthma or chronic obstructive pulmonary disease (COPD):results of the Detection, Intervention, and Monitoring Program of COPD and Asthma (DIMCA) Study. J Asthma 2000; 37(3):225–234.
- George J, Kong D, Thoman R, Stewart K. Factors associated with medication nonadherence in patients with COPD. Chest Nov 2005; 128(5):3198–3204.
- Cramer JA, Bradley-Kennedy C, Scalera A. Treatment persistence and compliance with medications for chronic obstructive pulmonary disease. Can Respir J 2007 Jan-Feb; 14(1):25–29.
- Nides MA, Tashkin DP, Simmons MS, Wise RA, Li VC, Rand CS. Improving inhaler adherence in a clinical trial through the use of the nebulizer chronolog. Chest Aug 1993; 104(2):501–507.
- Simmons MS, Nides MA, Rand CS, Wise RA, Tashkin DP. Trends in compliance with bronchodilator inhaler use between follow-up visits in a clinical trial. Chest Apr 1996; 109(4):963–968.
- Chambers C, Markson L, Diamond J, Lasch L, Berger M. Health beliefs and compliance with inhaled corticosteroids by asthmatic patients in primary care practices. Respir Med Feb 1999; 93(2):88–94.
- Barnes P. Inhaled corticosteroids in COPD: a controversy. Respiration 2010; 80(2):89–95.
- Choo PW, Rand CS, Inui TS, Validation of patient reports, automated pharmacy records, and pill counts with electronic monitoring of adherence to antihypertensive therapy. Med Care Sep 1999; 37(9):846–857.